9)_Animal_CNN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [
        "31A3WZP6xTsq"
      ]
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MnyzGHJ2AOdo"
      },
      "source": [
        "# Load Into CNN Stock Prediction G-Drive folder"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zhMBr-FK_4qv",
        "outputId": "db4bf34b-b77c-409a-ea3e-83227638368a"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "NO8DoL8gPMG9",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ead0dd29-99a9-447e-e165-61f099af9998"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "drive  sample_data\n"
          ]
        }
      ],
      "source": [
        "!ls"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "CA2FBK3r_Kgx",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "025fbe5e-9c86-45e0-fc9b-d4048e87cbcc"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "/content/drive/MyDrive/CNN Stock Prediction\n"
          ]
        }
      ],
      "source": [
        "%cd \"drive/MyDrive/CNN Stock Prediction\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "A7tY6y9j_-Hb",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "323d0409-d45d-4302-c367-0ba3ad98f85b"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "4_best_model_keras.h5  cifar10\t\t\t IBM_kbest_selected_features.csv  test_model.png\n",
            "4_model.png\t       Eric_best_model_keras.h5  labeled\n",
            "adjusted\t       features\t\t\t labeled_indicators\n",
            "animals\t\t       historical\t\t split\n"
          ]
        }
      ],
      "source": [
        "!ls data/"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Import Dependencies"
      ],
      "metadata": {
        "id": "z275U2lFmFGU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from PIL import Image"
      ],
      "metadata": {
        "id": "5syBuIPgm_2T"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# optional pandas display options\n",
        "pd.set_option('display.max_rows', 200)\n",
        "pd.set_option('display.max_columns', 200)\n",
        "pd.set_option('display.width', 150)"
      ],
      "metadata": {
        "id": "saLWD5Fb8Lpz"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Load Data"
      ],
      "metadata": {
        "id": "y18DGIe-7dVp"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from keras.datasets import cifar10\n",
        "from skimage.transform import resize"
      ],
      "metadata": {
        "id": "M8WPw-sPjRc9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# load dataset\n",
        "(X_train, y_train), (X_test, y_test) = cifar10.load_data()\n",
        "# summarize loaded dataset\n",
        "print('Train: X=%s, y=%s' % (X_train.shape, y_train.shape))\n",
        "print('Test: X=%s, y=%s' % (X_test.shape, y_test.shape))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OILimMgwBz2Z",
        "outputId": "20430eb4-361a-4dea-d00e-debf891e2adc"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train: X=(50000, 32, 32, 3), y=(50000, 1)\n",
            "Test: X=(10000, 32, 32, 3), y=(10000, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Define the labels of the dataset\n",
        "labels = ['airplane', 'automobile', 'bird', 'cat', 'deer',\n",
        "          'dog', 'frog', 'horse', 'ship', 'truck']"
      ],
      "metadata": {
        "id": "sJu6puuzm1iT"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Subset CIFAR Classes"
      ],
      "metadata": {
        "id": "2uD1tvii7gKr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "bird_index = np.where(y_train.reshape(-1) == 2)\n",
        "X_train_class_0 = X_train[bird_index]\n",
        "y_train_class_0 = y_train[bird_index]\n",
        "\n",
        "dog_index = np.where(y_train.reshape(-1) == 5)\n",
        "X_train_class_1 = X_train[dog_index]\n",
        "y_train_class_1 = y_train[dog_index]\n",
        "\n",
        "horse_index = np.where(y_train.reshape(-1) == 7)\n",
        "X_train_class_2 = X_train[horse_index]\n",
        "y_train_class_2 = y_train[horse_index]\n",
        "\n",
        "X_train_animal = np.concatenate((X_train_class_0, X_train_class_1, X_train_class_2))\n",
        "y_train_animal = np.concatenate((y_train_class_0, y_train_class_1, y_train_class_2)).reshape(-1,1)"
      ],
      "metadata": {
        "id": "vIk2EuNdKlDX"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "bird_test_index = np.where(y_test.reshape(-1) == 2)\n",
        "X_test_class_0 = X_test[bird_test_index]\n",
        "y_test_class_0 = y_test[bird_test_index]\n",
        "\n",
        "dog_test_index = np.where(y_test.reshape(-1) == 5)\n",
        "X_test_class_1 = X_test[dog_test_index]\n",
        "y_test_class_1 = y_test[dog_test_index]\n",
        "\n",
        "horse_test_index = np.where(y_test.reshape(-1) == 7)\n",
        "X_test_class_2 = X_test[horse_test_index]\n",
        "y_test_class_2 = y_test[horse_test_index]\n",
        "\n",
        "X_test_animal = np.concatenate((X_test_class_0, X_test_class_1, X_test_class_2))\n",
        "y_test_animal = np.concatenate((y_test_class_0, y_test_class_1, y_test_class_2)).reshape(-1,1)"
      ],
      "metadata": {
        "id": "cH2L2A3fbEZN"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(f'X train Shape: {X_train_animal.shape}, y train shape: {y_train_animal.shape}')\n",
        "print(f'X test Shape: {X_test_animal.shape}, y test shape: {y_test_animal.shape}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eR9qmCqUK3aZ",
        "outputId": "d6d8c55c-ac59-49b4-e3b3-6f0fef565830"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X train Shape: (15000, 32, 32, 3), y train shape: (15000, 1)\n",
            "X test Shape: (3000, 32, 32, 3), y test shape: (3000, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "pd.DataFrame(y_train_animal).value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3bZ___B6V8mI",
        "outputId": "93e2164d-b85e-4a83-a366-f8aa64d867e1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2    5000\n",
              "5    5000\n",
              "7    5000\n",
              "Name: count, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 54
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Data Preprocessing"
      ],
      "metadata": {
        "id": "23-Wndiw1HQZ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# helper function to plot dataset image\n",
        "def plot_img(img, i):\n",
        "  plt.subplot(331 + i)\n",
        "  # plot raw pixel data\n",
        "  plt.imshow(img)\n",
        "\n",
        "  plt.show()"
      ],
      "metadata": {
        "id": "Qux-wjHZ6rr_"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import random\n",
        "from random import sample\n",
        "\n",
        "img_index = random.randint(0, 15000)\n",
        "\n",
        "print('original CIFAR10 image')\n",
        "ex = X_train_animal[img_index]\n",
        "plot_img(ex, 0)\n",
        "\n",
        "# average RGB channels into one channel (grayscale)\n",
        "ex = np.mean(ex, axis=-1)\n",
        "print(f'shape after averaging 3 layers: {ex.shape}')\n",
        "plot_img(ex, 0)\n",
        "\n",
        "# resize image to 15x15\n",
        "new_shape = (15, 15)\n",
        "X_resized = np.zeros((ex.shape[0], *new_shape))\n",
        "X_resized = resize(ex, new_shape)\n",
        "\n",
        "# add single channel\n",
        "X_resized = np.expand_dims(X_resized, axis=-1)\n",
        "print(f'shape after resizing and adding single channel: {X_resized.shape}')\n",
        "\n",
        "plot_img(X_resized, 0)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 535
        },
        "id": "CTfiFm7Muf-Z",
        "outputId": "c0988cc7-1fa5-4031-9afe-a88aa8d3d23b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "original CIFAR10 image\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAJsAAACcCAYAAABoZNk0AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAZKklEQVR4nO1dXWxc1bVe539mPL/+j7GNQ0sJLYXc6/y56FaoihqJJ0oe+kgRAtHaSDRPzUtRn6L71EptWqlSG55QKh4QKtKNbuW0qbgkTWNuSxPAAZIQJ/aM49gztufvnDln3wc7Z621nYDDjY9Pwv4kK/ucveecPTtr9lp7/WpCCAEKChFA3+wJKHx5oIhNITIoYlOIDIrYFCKDIjaFyKCITSEyKGJTiAyK2BQigyI2hcigiE0hMmwYsR0+fBiGhoYgkUjA7t274fTp0xv1KoW7BBtCbH/4wx/gwIED8Morr8C7774Ljz32GOzbtw9mZ2c34nUKdwm0jTDE7969G3bu3Am/+tWvAAAgCAIYGBiAl156CX7yk5985meDIIDp6WnIZDKgadqdnprCBkAIAUtLS9DX1we6fuv9y7zTL3ZdFyYmJuDgwYPhPV3XYe/evXDy5Mk145vNJjSbzfD66tWr8PWvf/1OT0shAkxNTUF/f/8t++84sc3NzYHv+9DT08Pu9/T0wIcffrhm/KFDh+BnP/vZmvu/ffO3kGpLggZ8dzMNnLJmSNPXcaxhGHjb5OMEeabQAukR+DlHs/G+4PMwyK6byLaxPsvC97VabtiuVatsXN3D65bHnz99oRi2P/rgHOt77+9/C9vn/zlJnueycUO9ybC9b/h+1lcgX3uhVME5Li2zcYHrhW1bWsd0fuV7N1sB/Of4JchkMvBZuOPEdrs4ePAgHDhwILxeXFyEgYEByKSzkEqn1oy3TCtsa4a0ZZMtXCfjDIkdG6IVtgOPL66l4/9CwsblcUxOUPkc/pjSmXY+D8MPm25rKWwvLF1jw8pVnNf89SXWd/GT98J2Y7HI+h4cyoXtpRJ+z6tF/l1sHdevzeRruSWfCNudGWw3q1k2Tni4Vo5jsz5fX/medXfl388Te+44sXV2doJhGFAqldj9UqkEvb29a8Y7jgOO49zpaSjEEHf8NGrbNgwPD8P4+Hh4LwgCGB8fh5GRkTv9OoW7CBvCRg8cOADPPPMM7NixA3bt2gW/+MUvoFqtwrPPPrsRr1O4S7AhxPb9738frl27Bj/96U+hWCzC9u3b4dixY2sODZ8FXTdA1401BwSdCv66JCPo+HWEjnKIoflsmNFEwVyrLrK+tjQ+QycCd2e+g43rLnTiM7Qk6/M1/FxDJ/JhW1qaL87f1izW9cTIcNgW9eusz12eC9tJrRG2//v439g4C/B76wHXcLUl8X1NcihqS/I5WmT93YCv4/VVedcz1qc927ADwtjYGIyNjW3U4xXuQijbqEJk2HTVx62gaTpo2lo2qgFu+bJKAzT8Or6Bx3QDPDbMJ6xTb9RYX7pQwM8R1UfSSbBxOlikzX+zgSC6QMDPmVaBjTNRlw3ucoX1iUYdn9HgKo2Ugax5z45vhu1ikbPbagmvE6ak/qEs3Ma10iR2axL9osRFIWOsqINMrQXrgdrZFCKDIjaFyKCITSEyxFZm0zUDdM1YYwLRdKr64L8VZis1iFnL5zKb8FBYkhdAC9BcZZn4LsviVg5Np6oKPkc/IHIOUYt4AZ9vtYoyVXWRy45UWxM0G6xPtFB1k0ujeWnH9kfZuItn0aaaSvB3m9S0Z6EpS7Qk+auF6yGrTzK1lUka7vq8c9TOphAZFLEpRIbYslHD0MA0NBCaxCoJaxPEmrD6IdIkW37AXW80E/v8GmcN89PomdFzXxc+j2stAASZh8RGgwBZkefju4MWd2dyXWTnQvqfyHTmw3atOcP66mVkuS7xWrovn2fjkg8Mhe00cFZsEjURnVVTcP1Gs4UiiF+ts778/Mq170k6kVtA7WwKkUERm0JkiC8b1TUwdA0CnbNKjRjbQeqjzpOmhuzRlFTfjRqyr+KVadaXbUOWuKUPje+a4CdaINe+xHoaVQzsmS+j46Pr8pNeq4lsKWHx371oEiO6z9mvRp6j+chSHZ0b8/MpPEEbxPUeAMCl10RU8Xw+R5+cztsM/vxCfeXdtrc+Q7za2RQigyI2hcigiE0hMsRWZtM0EzTNBF0zpfsop2mSzKYT2cMm4bDBMld9LM5icEmlyjX3nb3d9IFh03V5QIoVoOdIs8XlofLCBXz+tcthe03kko5BNHaCRya1yLx0SdZLEFWLHuC7fcmZVFh4Xa/xNWgJXCufeHa0JPk2QdY73eB9ucqK3Gq0lOpDIWZQxKYQGWLLRm8Y4gPJgkCDiEFisdQubxIt/nSR5xi5cBE18t0D3ayvrYDXAbESNGucjSZS5bBdb3JWvEjUHYKo+PUW1+L7LWT1tWXe16ohy7WlPcG2UaVBg6ANyUHS88n8Xa66scg6Us2KHE6QbeG703PcgpBdWlljzVdsVCFmUMSmEBkUsSlEhtjKbJpmrAa8SOYqIqdpksxGTVRuA2WsxQaXNbq/8rWw/dAjD7M+dwnlu5liGT9T4POwbXRaXK5zeWtmej5s16soe5kad8D0PZzj8hKPX3WII2g7txKBIAlefIfEypp8YLm0ELbnr3GZ03No4h1s5wK+pskyqlY6r3P1SWaVfGSvl1tB7WwKkUERm0JkiDEbXYkb1WUrAYtBkKwLxBOjvIAspGPLFjau6/6vhG1LesYFkkPuo4lTYfs/dv8bG5dv7wvbtUVuQTj3r4th+4Pz2PYFTzm1/WFMR+GYUmyrjfGmGZN/DkzcI1yBrLPe5CqIjy5eCdvePGfTfg6f6ZDnJRtSfOksiiZZHtoK+qqqZb07ltrZFCKDIjaFyBBbNiqMlT9NMi7rNLRP5+ruFjEIJ0gIXXsPT0IYJNHoXZm+xPrcEp5Gp6fREnBxlqc2eJSwuUqVJz78ZBJZ53unMYNkV/8gG5fc/UDY1iTrglfD7yJsbojXSGZLw8QT7uVPeWbLC0WcV7vFWWyKHCwThD2mFvi77AaSSFM6/d+IWKxryoKgEDMoYlOIDIrYFCJDfGU2XQdh6MwhEoDLcJr0WzFsTCOQ6R8K2wkp3VWjgaqKufffZX2FAL00/v0xrMeQ6epi48BCmbApeT0YxP1k24OYEv7Rndv5uzrQClH8lGv43QX0JMlZUrquJHkf8eyYOv8pG9dYwGcmunl2TFhEq4pRRtk3UZccMImMvCxZMjKrniMtTQW8KMQMitgUIkNs2ShmMZIcJMm1AXxfp5mFmkkSjyBt8/VPPgnb3kVePWWggKyt9yE02BvdQ2ycbaP6JJvhyZ0fHd4etoMWGuL7+rmjpiAqg5qU2qBWRza6ILE2ixS/aMzj8+c/nmLj2girTEqOlZaL77YbJK5DsnIEFonlAA5tNY5BW2f5M7WzKUQGRWwKkUERm0JkiL3MtjZuFK9NKbeFIBnCfRIz6ZMiFQAA8+fOhO12t8z6CjYpiEaq7ZntXC6zSbbGjnYui93/VfQqaTbQhJRNcw+WMimO1mhI2TFJEbia4HtClWQgv3wZ5U9/jnt2bCFOjck5KSs6cUrNEDKwJNlOpzGq0t50QxRep+ZD7WwK0UERm0JkiC8b1e2VP4mNGjRJs8l/KzphnXnyOyqde4+Na5XOh+3uBH9+i8RyJhPIruTkiuU5dJVIJnktUjuNahFPoEWiXufqjQXC9twm9++3EujNEdi8VuiywHkteySuUxIr2ol6IyOxujSJO0gQK4GjcVZP0y+kBO+7kb3SVDEICnGDIjaFyBBbNmoYJpiGxbIWAfATaCCdnKirvnflati+9s8zbFw7oNbdsnjJQ9fHJUkR9uXY3BjeaiDbs5LcyG0mSF2BKvbVa/xEOE/YKHUOAADIduLp18nkWJ+RQLbduxVPvpD/gI1LzqAjaEKqJ5Eg4kmSLFzS4haEFInRaJPYtLH6f2N+TrnuG1A7m0JkUMSmEBkUsSlEhtjKbLplgG4ZYAgpjtEg1waXL2jt0IvvHAvb5sIlNq4tR2JPpfQOLkmRIHyUvQrJdjZOc/B3ep14dgAAeCRTpCDxI1Up++PsdUzTIIeMFDrQKpEk9akAAICkWbCJU+dCjqtI/BI+VU5jYVlEFiNeJG1SnG6SFDKx5Fphq24g66zarXY2heigiE0hMsSXjeorqRdk1Ycgx29HMlBfO3c2bC9fwPZAgu/zFskCFEhs2teRpXhEDfLx5AU2LlNAdYRIS2oRorn3iIG9WORxneUlNMR3d3NjfmcnpmawHM4eSYVJEKRIhpPhlgyfLp3EAg1i6E+Q9UhK620RNrqmSvrq11SGeIXYQRGbQmRQxKYQGeIrs606T/pCMpGQ3Bb1GR7gUfr7/4TtPClelrallFMk9rQqpZlarKOMdfU8Oia+T1JfAQBs6cX8IUPfeIT1aQGR+xoo0Exf4VnLaTXxQoHHpeayqGpJSaoPlxTG8IkclS9w9UzlM4rKGeTaJjKbo/O1ouoOKbUKrFNUw8/fzuBDhw7Bzp07IZPJQHd3Nzz11FMwOTnJxjQaDRgdHYWOjg5Ip9Owf/9+KJVKt3iiwpcJt0VsJ06cgNHRUTh16hT86U9/As/z4Lvf/S5Uq+iz9eMf/xj++Mc/wuuvvw4nTpyA6elpePrpp+/4xBXuPtwWGz127Bi7fvXVV6G7uxsmJibg29/+NlQqFfjd734Hr732GnznO98BAIAjR47Aww8/DKdOnYI9e/as+126tpp6Qfb6ILzn4r/+zvq0WawT1ZVHbwvT4KyyQdIllBs8RdRcgKmr3vnw7bB94SJn2ZkUqiP21PkzvvEQemLMkVRbc5Lqo0BiVPO5TtZHYxzSbXnW5wmM4Ky7JDa0jde/qhBWKZc/52yUWCQkZ1ULaLoLzjiDVZ2HHoXXR6Wy4q3a3r4iK0xMTIDnebB3795wzLZt22BwcBBOnjx502c0m01YXFxkfwr3Jr4wsQVBAC+//DI8/vjj8MgjKwJysVgE27Yhn8+zsT09PVAsFm/ylBU5MJfLhX8DAwNfdEoKMccXJrbR0VE4e/YsHD169P81gYMHD0KlUgn/pqamPv9DCnclvpDqY2xsDN566y3461//Cv39/eH93t5ecF0XyuUy291KpRL09vbe5EkAjuOA4zhr7tdrdTAME2ydx1POXMDcHDPn3mF9XYBeH/UWeuC2JJmCaCNgocblrWw/ZgG3FlA2mrnGY0+vE/nlwWszrG+uHeWt6SuYxsqSAnQSxKPXcbipyTTQBKbrfH004kri+aTYmuTtK4i8KyTVBxW/6Kxk+YuqO2Sz1I10WtpGyGxCCBgbG4M33ngDjh8/Dlu3bmX9w8PDYFkWjI+Ph/cmJyfh8uXLMDIycjuvUrgHcVs72+joKLz22mvw5ptvQiaTCeWwXC4HyWQScrkcPPfcc3DgwAFob2+HbDYLL730EoyMjNzWSVTh3sRtEdtvfvMbAAB44okn2P0jR47AD37wAwAA+PnPfw66rsP+/fuh2WzCvn374Ne//vVtT6xerYGu6VCpzLP7H739X2G7SeI/AQBo+OIi8aiwDMkKQZwFaxr32PgaqWv15Lad+BmbB7U0yqjS6OrKs76rVy/huDqernt6uIa/qwtFC1n1QdmqrkkpHwPkZ3USb1qXaorScty+L2VWJ+ofQZ63xrPjM1QfQvr383BbxCbWkYcrkUjA4cOH4fDhw7fzaIUvAZQhXiEyxNYQf6PqxvTZ/2W365c+Dtspi58kWwaeHqnTYm2e78jVBv7G+ndwWbJz6KthO+vkw/Yzz/D6V9V51BvOzXJ1zfwcnlxNwmRMyYGx576hsJ3J8yxJKeIwaZn8NFr38NRZJ2W7gwI32Hu0VGSDxz9Qm4pLOBZnxAABNcRLaRZaq96TrfUdRtXOphAdFLEpRAZFbAqRIbYyW7O6BKYmoExkNACAlIdSRTYjFZIgMpxHVN+1OpfZFqt43ffQN1hfvgsDTWjZbkvKDJkhMat6jstbA10o35UKaF1oNngxtEIPOkzmc/y7ZFKo7mhpXDYNiGeKbuN/od3JVStAUn6BJLPRQhlNIrM1Ap4T3CFymkws3qoTakuoXB8KMYMiNoXIEFs2ujB7GZrLKagtcL/9nI7bfEouyEGKcDQ0HGdIdv6v7cCaVA9v52zUIEbupIesZ+7SJTbOq6N3MkjxmikLf8OdhM25UtUKsTAdtstF/vwmSYvlSqXFlwnXshwSR5vmzpMGSeXlLvCa276Pk2mRtiu4o2mdrKNspHdXWWxNyOU4bg61sylEBkVsCpFBEZtCZIitzFZbnAPhJaHVqLL7RBwCRzL/0HSbGgnwsB3uOPjV7Rjnqdn8GZ+cR+dMp4bPsxa590ljHq8DN8/6KhUMXVy6julW3WqZjYMaeqYEUjqtZgpziSxJQT9lYobKbHswbGuyEyoJymkI2esD29QBsyHJbFRKo7VeAQCaq3vVsi8n/Lo51M6mEBkUsSlEhtiyUQEChCZAk3zng0C/aRsAmMO8RdhoR6HAhqWyGJ9QnLnK+qY+xpgBbQFjMtulohgp4mHSWuYlt+tN/JxJxIC05GZok3LcdYN/l8oSenbM1/m77f77wnaSeIcst3EVjNaGqg9PUlt4xGGyRdiou8Z5kkKq27rKcqtCsVGFmEERm0JkiC0bTaQzkEglIZnjxuXmdTzpNQK+yRvEuEwd/TIpXliDZp5ckljgEoktqBXRCbJc4Rr4wW58pmPyGIGMjadCnRz7hMQOr8zhu4tlXtfKs/H5yUEexfbA8A58dzuOWyrzBD4mYasNiY22Wje3ILSkVEU6SZgt2wlqq+tdX2fqSbWzKUQGRWwKkUERm0JkiK3MZqfT4LSloOfRb7L718+gE6PlcWfElI3yUUBkNjlzYzaH2nlfcmjUiAdDk2j8t3TwjN12ktQ2TUl9xBulWcM5XS1xK8TfiJrFz/Js4Vu/gnLa4KPbWV/f/UN44aC8dOVTHgOrkyzmLUPS/pPslS7Jut4KpFLoJnrB1GzuQLqQWFFL1VpKZlOIGRSxKUSG2LLRQNMh0HRw7htk9zPLuK1Xzr/P+pwW0dyTc3pXB4/57O0bCtteiWcgsonFIkPqmXbmOavMpTAGwTY4+6JVMVxi5TCSXAUjiAUhSHDtf5rEQnR08uTO2RQpJ07YnGVya4ueQhWMnMmpScSFKlFqOJKxXdgklkMqobWcXVmDulzT/BZQO5tCZFDEphAZFLEpRIbYymwQWACBBb7Bp1ggKa1cj5t/pq98GLbTRGZ7MM3TUbWlMM4zZfJaoV3pfNju6UVZL+9IjppElSAkrwefmoJILo5Mmst9PV1oivtESlx97RqanipLvK83QBmuXkWTlyRugU4CZVqSralBTExlE1UaWkaSv7L4DC/H1SLaquyouUpmU4gZYrez3cgBV6+t5seVyvT6TTx91SWFbIP4nBlE/7hcq7FxNP390jLf2ap1PNHqJEdtUqojYJI6PppUZ8H3yEmPPCPw+HdpkOR9TY8rTBvku9GiJgAASyTRYc3D+ddq3JjPnh9I7yZu4nViiHda0i5FUhS1POmkurqj3TiNfl7+Pk2sJ8NfhLhy5YpKT3+XYmpqiiX0lhE7YguCAKanp0EIAYODgzA1NQXZbPbzP3iPY3FxEQYGBmK5HkIIWFpagr6+PtD1W0tmsWOjuq5Df39/yOqy2WzsFnczEdf1yBF7862gDggKkUERm0JkiC2xOY4Dr7zyyk2rv3wZcS+sR+wOCAr3LmK7synce1DEphAZFLEpRAZFbAqRIZbEdvjwYRgaGoJEIgG7d++G06dPb/aUIsGhQ4dg586dkMlkoLu7G5566imYnJxkYxqNBoyOjkJHRwek02nYv38/lEqlWzwxZhAxw9GjR4Vt2+L3v/+9OHfunHj++edFPp8XpVJps6e24di3b584cuSIOHv2rPjHP/4hnnzySTE4OCiWl5fDMS+++KIYGBgQ4+Pj4syZM2LPnj3iW9/61ibOev2IHbHt2rVLjI6Ohte+74u+vj5x6NChTZzV5mB2dlYAgDhx4oQQQohyuSwsyxKvv/56OOaDDz4QACBOnjy5WdNcN2LFRl3XhYmJCdi7d294T9d12Lt3L5w8eXITZ7Y5qKzmF2lvX3GynJiYAM/z2Pps27YNBgcH74r1iRWxzc3Nge/70NPTw+739PSEVZu/LAiCAF5++WV4/PHH4ZFHVtKyFotFsG0b8vk8G3u3rE/svD4UVjA6Ogpnz56Ft99+e7OncscQq52ts7MTDMNYc7oqlUrQ29t7i0/dexgbG4O33noL/vznPzNnxN7eXnBdF8rlMht/t6xPrIjNtm0YHh6G8fHx8F4QBDA+Pg4jIyObOLNoIISAsbExeOONN+D48eOwdSvPyzY8PAyWZbH1mZychMuXL98d67PZJxQZR48eFY7jiFdffVW8//774oUXXhD5fF4Ui8XNntqG44c//KHI5XLiL3/5i5iZmQn/arVaOObFF18Ug4OD4vjx4+LMmTNiZGREjIyMbOKs14/YEZsQQvzyl78Ug4ODwrZtsWvXLnHq1KnNnlIkgJUMyWv+jhw5Eo6p1+viRz/6kSgUCiKVSonvfe97YmZmZvMmfRtQLkYKkSFWMpvCvQ1FbAqRQRGbQmRQxKYQGRSxKUQGRWwKkUERm0JkUMSmEBkUsSlEBkVsCpFBEZtCZFDEphAZ/g/oML0BVamVlAAAAABJRU5ErkJggg==\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "shape after averaging 3 layers: (32, 32)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "shape after resizing and adding single channel: (15, 15, 1)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAJwAAACaCAYAAABc4SFQAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAANLklEQVR4nO2df2xUVRbHv/Pmx5u2U2aoLVNGWigKLiGxupV2dauRpUm3m7AQE6Ks2TSa2GwibkxjDCRClZDASmIaoSvZP0w1m7WQbMB1Y/oHDVliUmqAsrp0JdCtMLXO1KJDZ4bOj87c/YO0WmfOnQ5O77x2zid5f/Sdd9+97/Y777173r3nmIQQAgyjCC3fDWAKCxYcoxQWHKMUFhyjFBYcoxQWHKMUFhyjFBYcoxQWHKMUFhyjFMtCnbirqwuHDx+Gz+dDbW0tjhw5gvr6+ozlkskkxsbGUFpaCpPJtFDNY3KIEALBYBAejwealuEeJhaAnp4eYbPZxLvvvisuX74sXnjhBeFyuYTf789Y1uv1CgC8LcLN6/Vm/P+ahMj9x/uGhgZs2rQJR48eBXDnrlVVVYWXXnoJu3fvlpa9desWXC4Xfv7XP8BcrKfYHbaotHyJNUbaLFpyHq1PZYPja6n9iZIrpG04voK0nfI/TNrG/rFaWueKYwNSO0VwxybSVjQel5YVlvRPnOnpCAb+dQiBQABOp1N6jpw/UmOxGC5cuIA9e/bM7tM0DU1NTejv7085PhqNIhr9XkTBYBAAYC7WYSlJFZzFJq/faqVtdys4u0NyUgAlDvoxUhSju9gaoi/GrNuldVpM8jaR57XS57VYzNKylOBmmM8rUM4HDRMTE0gkEnC73XP2u91u+Hy+lOMPHjwIp9M5u1VVVeW6SYyByPsodc+ePbh169bs5vV6890kZgHJ+SO1vLwcZrMZfr9/zn6/34/KysqU43Vdh66nPjqZpUnOBWez2VBXV4e+vj5s374dwJ1BQ19fH3bt2jXv85RYY7BYU98JHFb5oGG5bYq0ecMu0lam3yZtmx1D0jp/aacfFBXmEdJ2qbSatF2uvVdaJ/74GGm698QwaYvcQ7c1slz+wzdH048vE7GEtNwPWRA/XHt7O1pbW/HII4+gvr4enZ2dCIfDeO655xaiOmYRsSCCe/rpp/HNN99g37598Pl8eOihh9Db25sykGAKjwX70rBr166sHqFMYZD3USpTWLDgGKWw4BilsOAYpSzYoOGnUmyNpf0uWprBD7fMQvvhQjF6lHw7Tn/XLNMi0jqBYtISTNLfPMcjpfQpNfmcitsrafv4b9aStmm6qdDkXQsr4aoU83fD8R2OUQsLjlEKC45RCguOUQoLjlEKC45RimHdIg5rFFZr6tC/1CJ3UcimkX/19XLSZnfQayHGEhL3BYDrt+lu/NPINtIWmZZ0f0I+XVuL0fYpehkFrEHaZkrKXTGCuD1R+9PBdzhGKSw4RiksOEYpLDhGKSw4RiksOEYphnWL2LU4bObUof+yDG6RryIu0lZUSk+HcBTRtp6JX0jr9Etmfdz4z0q6oMQLYUrjEvohSRttt4/S5Ypu0m6jeLHcFZNM8/8AABPPFmGMCguOUQoLjlEKC45RCguOUQoLjlFKzt0ir7/+Ot544405+x544AF88cUXWZ2nxByDbk4d+hdr9KwOAHBa6UU0bRs+IW1He39N2s6uKpLWGfuWDvJ3/99pN87oFnpFS0KXu0VMktiKK86kxuGbIemk64w75UEQpyqIBUHx+QdRXRA/3MaNG3H69OnvK7EY1t3HKGZBlGCxWNLGgmOYBXmHu3r1KjweD9auXYtnn30WN27cII+NRqOYnJycszFLl5wLrqGhAd3d3ejt7cU777yDkZERPP7447PBon8Mx/gtLHIuuJaWFuzYsQMPPvggmpub8fHHHyMQCODEiRNpj+cYv4XFgr/Nu1wurF+/HteuXUtr5xi/hcWC++FCoRCGh4excqVk1gRTMOT8DvfKK69g69atWL16NcbGxtDR0QGz2YydO3dmdR6HJQrdkupsclro4M8AsFYfJ22dw1tIW2L5NGn7/c/OS+v8ZOI+0nbtd/QPzU67y2CfkE8VsoZo31fyS/q1xLzKQ9oSRfJkI1oivVy0RB79cKOjo9i5cydu3ryJiooKNDY24ty5c6ioqMh1VcwiJOeC6+npyfUpmSUEf0tllMKCY5TCgmOUwoJjlGLYaRwuy23Y08wyuccckpbzxstI283P6ZGyuYqe1rTK9q20zs0V9LKlL/33kDb7EN39sWXSKmG7Rds0RwlpEzba9TFdIpfDtD29qyahzT9VPN/hGKWw4BilsOAYpbDgGKWw4BilsOAYpRjWLbLCMoliizllvzVD5JQ/9/+KtMkuVjawvxGjXRsAEJqm5/MlbtI2y216lkWgTr46LVJOuzdW/JOuU7ZqK+rK4BbRCbeIid0ijEFhwTFKYcExSmHBMUphwTFKYcExSjGsW8RuisOeJo3RX0afkJazTtCXJFK9LLNoI3TAmtMnG6V1jrXQrppl1+hKZSmKHr6PjlYAAFdv0jNfTMX0tSSK6f6Jl8jdG1T/yfr1x/AdjlEKC45RCguOUQoLjlEKC45RCguOUUrWgjt79iy2bt0Kj8cDk8mEU6dOzbELIbBv3z6sXLkSRUVFaGpqwtWrV3PVXmaRk7UfLhwOo7a2Fs8//zyeeuqpFPubb76Jt99+G++99x5qamqwd+9eNDc3Y2hoCHa7PGjxDzk+vgnWsC1l/+hHa6TlHJIpPwlieg0AiM3fkbbQBJ26HAD0Ufp3aw3T7ZncQPvv7nNMSOv8r/8uQ9pKglFn8qflIgV51oJraWlBS0tL+oqFQGdnJ1577TVs23Yn1/v7778Pt9uNU6dO4Zlnnsm2OmaJkdN3uJGREfh8PjQ1Nc3uczqdaGhoQH9/fy6rYhYpOf205fPdCXjmdrvn7He73bO2HxONRhGNfp86koNKL23yPkrloNKFRU4FN5Obwe/3z9nv9/vJvA0cVLqwyKngampqUFlZib6+vtl9k5OTGBgYwKOPPpq2jK7rWLZs2ZyNWbpk/Q4XCoXmRCQfGRnBpUuXUFZWhurqarz88ss4cOAA1q1bN+sW8Xg82L59e1b1/PvSWmhFqW6Ussm7z0FlC9LG3Rs/Im39q++X1vl5gI6bOzxeTto2rKBdH7pGxxwGACHpBlFMu5+0OO2KyZDGDEj1Ut2pL4sU5FkL7vz589i8efPs3+3t7QCA1tZWdHd349VXX0U4HEZbWxsCgQAaGxvR29ublQ+OWbpkLbgnn3wSQvLzMplM2L9/P/bv3/+TGsYsTfI+SmUKCxYcoxQWHKMUFhyjFMOu2ioZ1WDWU38PiQx54JKSKwpvodOB/83fQNo+/5p2ewBAMknPQpGlJ782spq2CdoGAPEVcdpWRneCZZLuA13iNgKApCX9dU5nkYKc73CMUlhwjFJYcIxSWHCMUlhwjFJYcIxSDOsWiZQLaPbU4bbDKw+4kpS4TTzlAdJ2cWAdabMF5L/L2L30zA5ziF6ZUvo/+pyBDXJXQ936L0mbv5zOUF3qo3MmmaNyt4gWI/o+Li835xzzPpJhcgALjlEKC45RCguOUQoLjlEKC45RCguOUYph/XC2+4MwF6cuI5rKkJvbEqb9dK3VdLiJAyO/JW0RW4aVYkW0H077jvbD6ZIVaGXr5GnPH1tOO/GOO2ifomkqStq0DNOMyGBA80+1xXc4Ri0sOEYpLDhGKSw4RiksOEYphhulzqzqT9xOP5pKROhFIABgitJDpqkQPZpMTtHnNcUzDMOs9IKWZIQe+ckWn1DXP0NEci2JGH0t00n6vNPT8r5NaOnvT9PxO+VkERlmMIn5HKWQ0dFRjhG3SPF6vVi1apX0GMMJLplMYmxsDKWlpTCZTJicnERVVRW8Xi+H8kqDEfpHCIFgMAiPxwONuAvOYLhHqqZpaX8lHDtOTr77x+l0zus4HjQwSmHBMUoxvOB0XUdHRwd0PUOMhwJlsfWP4QYNzNLG8Hc4ZmnBgmOUwoJjlMKCY5RiaMF1dXVhzZo1sNvtaGhowKeffprvJuWNpZKn1rCCO378ONrb29HR0YGLFy+itrYWzc3NGB8fz3fT8sJMntqurq609pk8tceOHcPAwABKSkrQ3NyMSIbJDsoRBqW+vl68+OKLs38nEgnh8XjEwYMH89gqYwBAnDx5cvbvZDIpKisrxeHDh2f3BQIBoeu6+OCDD/LQQhpD3uFisRguXLgwJ++qpmloamrivKtpWEx5ag0puImJCSQSiazyrhYyd5OnNl8YUnDM0sWQgisvL4fZbM4q72ohczd5avOFIQVns9lQV1c3J+9qMplEX18fmXe1kLmbPLV5I9+jFoqenh6h67ro7u4WQ0NDoq2tTbhcLuHz+fLdtLwQDAbF4OCgGBwcFADEW2+9JQYHB8X169eFEEIcOnRIuFwu8eGHH4rPPvtMbNu2TdTU1Iipqak8t3wuhhWcEEIcOXJEVFdXC5vNJurr68W5c+fy3aS8cebMGQEgZWttbRVC3HGN7N27V7jdbqHrutiyZYu4cuVKfhudBp6exCjFkO9wzNKFBccohQXHKIUFxyiFBccohQXHKIUFxyiFBccohQXHKIUFxyiFBccohQXHKOX/gGKa2ZnktQYAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Process Single Channel Images"
      ],
      "metadata": {
        "id": "31A3WZP6xTsq"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Average RGB Layer\n",
        "Averages three layers of images in CIFAR10 data set into one layer"
      ],
      "metadata": {
        "id": "q7qEaSptMust"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X_train_gray = np.mean(X_train_animal, axis=-1)\n",
        "X_test_gray = np.mean(X_test_animal, axis=-1)\n",
        "print(f'X train shape: {X_train_gray.shape}')\n",
        "print(f'X test shape: {X_test_gray.shape}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8jlu_QR5wtNW",
        "outputId": "80bfad07-73f4-4964-d017-920fecfabecd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X train shape: (15000, 32, 32)\n",
            "X test shape: (3000, 32, 32)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Resize Images\n",
        "Resize 32x32 images to 15x15 for CNN input layer"
      ],
      "metadata": {
        "id": "UmQ3wrLCNp2S"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# resize image to 15x15\n",
        "new_shape = (15, 15)\n",
        "X_train_resized = np.zeros((X_train_gray.shape[0], *new_shape))\n",
        "X_test_resized = np.zeros((X_test_gray.shape[0], *new_shape))\n",
        "\n",
        "# reshaping\n",
        "for i in range(X_train_gray.shape[0]):\n",
        "  X_train_resized[i] = resize(X_train_gray[i], new_shape)\n",
        "\n",
        "for i in range(X_test_gray.shape[0]):\n",
        "  X_test_resized[i] = resize(X_test_gray[i], new_shape)\n",
        "\n",
        "# add single channel\n",
        "X_train_single_final = X_train_resized\n",
        "X_test_single_final = X_test_resized\n",
        "\n",
        "print(f'X train shape: {X_train_single_final.shape}')\n",
        "print(f'X train shape: {X_test_single_final.shape}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MeK4mwuXN02-",
        "outputId": "764631b8-a060-4845-95ad-48fbacb42538"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X train shape: (15000, 15, 15)\n",
            "X train shape: (3000, 15, 15)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Process RGB Images"
      ],
      "metadata": {
        "id": "lSgWN1hQL9KV"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# resize image to 15x15\n",
        "new_shape = (15, 15, 3)\n",
        "X_train_rgb_resized = np.zeros((X_train_animal.shape[0], *new_shape))\n",
        "X_test_rgb_resized = np.zeros((X_test_animal.shape[0], *new_shape))\n",
        "\n",
        "# reshaping\n",
        "for i in range(X_train_animal.shape[0]):\n",
        "  X_train_rgb_resized[i] = resize(X_train_animal[i], new_shape)\n",
        "\n",
        "for i in range(X_test_animal.shape[0]):\n",
        "  X_test_rgb_resized[i] = resize(X_test_animal[i], new_shape)\n",
        "\n",
        "# add single channel\n",
        "X_train_final = X_train_rgb_resized\n",
        "X_test_final = X_test_rgb_resized\n",
        "\n",
        "print(f'X train shape: {X_train_final.shape}')\n",
        "print(f'X train shape: {X_test_final.shape}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1yqqnStfL9dU",
        "outputId": "67dd1f42-6d69-4557-ba38-07e2cd12d209"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X train shape: (15000, 15, 15, 3)\n",
            "X train shape: (3000, 15, 15, 3)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Show Images"
      ],
      "metadata": {
        "id": "clzrRJrLxfU1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# plot 9 random images\n",
        "rand_img_ind = sample(range(0, 15000),9)\n",
        "\n",
        "print(f\"Original CIFAR10 Images: {X_train_animal[0].shape}\")\n",
        "for ind in zip(range(9), rand_img_ind):\n",
        "\tplt.subplot(330 + 1 + ind[0])\n",
        "\t# plot raw pixel data\n",
        "\tplt.imshow(X_train_animal[ind[1]])\n",
        "\n",
        "plt.show()\n",
        "\n",
        "for ind in zip(range(9), rand_img_ind):\n",
        "\tplt.subplot(330 + 1 + ind[0])\n",
        "\t# plot raw pixel data\n",
        "\tplt.imshow(X_train_single_final[ind[1]])\n",
        "\n",
        "print(f\"Processed CIFAR10 Images: {X_train_single_final[0].shape}\")\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 802
        },
        "id": "uB19XKZVN5zn",
        "outputId": "c570ad39-1304-4871-c897-136dba628047"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Original CIFAR10 Images: (32, 32, 3)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 9 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "name 'X_train_single_final' is not defined",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-17-f98bd9dd2392>\u001b[0m in \u001b[0;36m<cell line: 12>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     13\u001b[0m         \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msubplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m330\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;36m1\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mind\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     14\u001b[0m         \u001b[0;31m# plot raw pixel data\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 15\u001b[0;31m         \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mimshow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_train_single_final\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mind\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     16\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     17\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"Processed CIFAR10 Images: {X_train_single_final[0].shape}\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'X_train_single_final' is not defined"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# plot 9 random images\n",
        "rand_img_ind = sample(range(0, 15000),9)\n",
        "\n",
        "print(f\"Original CIFAR10 Images: {X_train_animal[0].shape}\")\n",
        "for ind in zip(range(9), rand_img_ind):\n",
        "\tplt.subplot(330 + 1 + ind[0])\n",
        "\t# plot raw pixel data\n",
        "\tplt.imshow(X_train_animal[ind[1]])\n",
        "\n",
        "plt.show()\n",
        "\n",
        "for ind in zip(range(9), rand_img_ind):\n",
        "\tplt.subplot(330 + 1 + ind[0])\n",
        "\t# plot raw pixel data\n",
        "\tplt.imshow(X_train_final[ind[1]])\n",
        "\n",
        "print(f\"Processed CIFAR10 Images: {X_train_final[0].shape}\")\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 884
        },
        "id": "OaV9WfZRh3hP",
        "outputId": "f45c1126-67de-4c03-e7ad-6768f7ab81d3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Original CIFAR10 Images: (32, 32, 3)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 9 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Processed CIFAR10 Images: (15, 15, 3)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 9 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Train Test Split"
      ],
      "metadata": {
        "id": "q4lYKnX1Y8DO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split"
      ],
      "metadata": {
        "id": "6Jd0PwQEZB-I"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Training and Cross Validation Data\n",
        "X_train_split, X_val, y_train_split, y_val = train_test_split(X_train_final, y_train_animal, train_size=0.8,\n",
        "                                                    test_size=0.2, random_state=2, shuffle=True,\n",
        "                                                    stratify=y_train_animal)"
      ],
      "metadata": {
        "id": "TsJ3oMpXY9jh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(f\"X_train shape: {X_train_split.shape}\")\n",
        "print(f\"y_train shape: {y_train_split.shape}\")\n",
        "print(f\"X_val shape: {X_val.shape}\")\n",
        "print(f\"y_val shape: {y_val.shape}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "G8Z_4qjNZi4O",
        "outputId": "affca506-f01f-4159-bc4d-782e801f4899"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X_train shape: (12000, 15, 15, 3)\n",
            "y_train shape: (12000, 1)\n",
            "X_val shape: (3000, 15, 15, 3)\n",
            "y_val shape: (3000, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Normalize and Encode Data"
      ],
      "metadata": {
        "id": "LBUeMGwsdUGx"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Normalize Images"
      ],
      "metadata": {
        "id": "bOPBtoT7c-Ll"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.preprocessing import MinMaxScaler, OneHotEncoder"
      ],
      "metadata": {
        "id": "P-X1rsvPmzxH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Encode Labels"
      ],
      "metadata": {
        "id": "j5VSoD63dB65"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Define dictionary for reassigning class labels\n",
        "labels = {2: 0, 5: 1, 7: 2}\n",
        "\n",
        "# Reassign labels\n",
        "y_train_split = np.vectorize(labels.get)(y_train_split)\n",
        "y_val = np.vectorize(labels.get)(y_val)\n",
        "y_test_animal = np.vectorize(labels.get)(y_test_animal)"
      ],
      "metadata": {
        "id": "GAedn7Fgc9tY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print('y train values')\n",
        "print(pd.DataFrame(y_train_split).value_counts())\n",
        "print('\\ny validation values')\n",
        "print(pd.DataFrame(y_val).value_counts())\n",
        "print('\\ny test values')\n",
        "print(pd.DataFrame(y_test_animal).value_counts())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iciZ5HYAd83v",
        "outputId": "a693f31e-a710-4b4a-bb9b-846c7561e123"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "y train values\n",
            "0    4000\n",
            "1    4000\n",
            "2    4000\n",
            "Name: count, dtype: int64\n",
            "\n",
            "y validation values\n",
            "0    1000\n",
            "1    1000\n",
            "2    1000\n",
            "Name: count, dtype: int64\n",
            "\n",
            "y test values\n",
            "0    1000\n",
            "1    1000\n",
            "2    1000\n",
            "Name: count, dtype: int64\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# One Hot Encode Labels\n",
        "\"\"\"\n",
        "0 -> [1, 0, 0]\n",
        "1 -> [0, 1, 0]\n",
        "2 -> [0, 0, 1]\n",
        "\"\"\"\n",
        "OHE = OneHotEncoder(sparse=False, categories='auto')\n",
        "\n",
        "y_train_split = OHE.fit_transform(y_train_split.reshape(-1, 1)) # Reshapes to a column vector\n",
        "y_val = OHE.transform(y_val.reshape(-1, 1))\n",
        "y_test_animal = OHE.transform(y_test_animal.reshape(-1, 1))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "49Rkx4vyc0iF",
        "outputId": "f025ecad-b82d-40ba-eca1-e16749fd83b6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/sklearn/preprocessing/_encoders.py:868: FutureWarning: `sparse` was renamed to `sparse_output` in version 1.2 and will be removed in 1.4. `sparse_output` is ignored unless you leave `sparse` to its default value.\n",
            "  warnings.warn(\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Add Single Channel to Images\n",
        "Reshape input images to include a single channel to fit CNN Sequential Layer"
      ],
      "metadata": {
        "id": "z95UxZY1fIaT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# add single channel if necessary\n",
        "# X_train_split = np.expand_dims(X_train_split, axis=-1)\n",
        "# X_val = np.expand_dims(X_val, axis=-1)\n",
        "# X_test_final = np.expand_dims(X_test_final, axis=-1)"
      ],
      "metadata": {
        "id": "mJSmz80beylm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(f\"X_train Images shape: {X_train_split.shape}\")\n",
        "print(f\"y_train labels shape: {y_train_split.shape}\\n\")\n",
        "\n",
        "print(f\"X_val Images shape: {X_val.shape}\")\n",
        "print(f\"y_val labels shape: {y_val.shape}\\n\")\n",
        "\n",
        "print(f\"X_test Images shape: {X_test_final.shape}\")\n",
        "print(f\"y_test labels shape: {y_test_animal.shape}\\n\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "v69YoasygfMr",
        "outputId": "6c0bfec9-781c-4991-9952-86eedeae8379"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X_train Images shape: (12000, 15, 15, 3)\n",
            "y_train labels shape: (12000, 3)\n",
            "\n",
            "X_val Images shape: (3000, 15, 15, 3)\n",
            "y_val labels shape: (3000, 3)\n",
            "\n",
            "X_test Images shape: (3000, 15, 15, 3)\n",
            "y_test labels shape: (3000, 3)\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# CNN Model"
      ],
      "metadata": {
        "id": "8rAlehV1NmGf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import keras\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Dropout, Activation, Flatten\n",
        "from keras.layers import Conv2D, MaxPooling2D\n",
        "\n",
        "!pip install h5py"
      ],
      "metadata": {
        "id": "ClKLwxI5NwKM",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8df48547-eb13-4133-9a70-6fc3cd5c5b63"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: h5py in /usr/local/lib/python3.10/dist-packages (3.11.0)\n",
            "Requirement already satisfied: numpy>=1.17.3 in /usr/local/lib/python3.10/dist-packages (from h5py) (1.25.2)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## CNN Model Implementation"
      ],
      "metadata": {
        "id": "c7g300nEN-D-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "params = {\n",
        "    \"input_w\": 15,\n",
        "    \"input_h\": 15,\n",
        "    \"input_c\": 3,\n",
        "    \"num_classes\": 3,\n",
        "    \"batch_size\": 1024,\n",
        "    \"epochs\": 2000\n",
        "    }"
      ],
      "metadata": {
        "id": "J5C3ZWfuUUV2"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras import regularizers\n",
        "\n",
        "# CNN model\n",
        "model = Sequential()\n",
        "# Conv2D\n",
        "model.add(Conv2D(filters = 20,\n",
        "                 kernel_size = 2,\n",
        "                 strides = 1,\n",
        "                 padding = 'valid',\n",
        "                 activation='relu',\n",
        "                 use_bias = True,\n",
        "                 kernel_regularizer=regularizers.l2(0.0),\n",
        "                 kernel_initializer='glorot_uniform',\n",
        "                 input_shape=(params[\"input_w\"],\n",
        "                              params[\"input_h\"],\n",
        "                              params[\"input_c\"])\n",
        "                  ))\n",
        "\n",
        "# Dropout\n",
        "model.add(Dropout(0.22))\n",
        "\n",
        "# Conv2D\n",
        "model.add(Conv2D(filters = 40,\n",
        "                 kernel_size = 2,\n",
        "                 strides = 2,\n",
        "                 padding = 'valid',\n",
        "                 activation='relu',\n",
        "                 kernel_regularizer=regularizers.l2(0.0),\n",
        "                 kernel_initializer='glorot_uniform',\n",
        "                 ))\n",
        "\n",
        "# Dropout\n",
        "model.add(Dropout(0.05))\n",
        "\n",
        "# Flatten\n",
        "model.add(Flatten())\n",
        "\n",
        "# Dense\n",
        "model.add(Dense(100, activation='relu'))\n",
        "\n",
        "# Dropout\n",
        "model.add(Dropout(0.22))\n",
        "\n",
        "# Dense\n",
        "model.add(Dense(3, activation='softmax'))\n",
        "\n",
        "# Compile Model\n",
        "model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy', 'mae', 'mse'])"
      ],
      "metadata": {
        "id": "zsT6aunf8pLQ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot Model\n",
        "from IPython.display import SVG\n",
        "from tensorflow.keras.utils import model_to_dot, plot_model\n",
        "\n",
        "plot_model(model, show_shapes=True, show_layer_names=False)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TTi979wcWVzv",
        "outputId": "85a76481-2366-4197-dd68-28e23ca599df"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "You must install pydot (`pip install pydot`) and install graphviz (see instructions at https://graphviz.gitlab.io/download/) for plot_model to work.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Model Checkpoint Callbacks"
      ],
      "metadata": {
        "id": "lPmv0U-nXFu9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint, EarlyStopping, ReduceLROnPlateau, CSVLogger, Callback\n",
        "\n",
        "best_model_path = \"models/animal_cnn.h5\"\n",
        "es = EarlyStopping(monitor='val_loss', mode='min', verbose=1,\n",
        "                   patience=100, min_delta=0.0001) # Stops the training early when validation loss stops improving, prevent overfitting\n",
        "\n",
        "rlp = ReduceLROnPlateau(monitor='val_loss', factor=0.02, patience=20, verbose=1, mode='min',\n",
        "                        min_delta=0.001, cooldown=1, min_lr=0.0001) # Adjusts Learning Rate on validation loss\n",
        "\n",
        "mcp = ModelCheckpoint(best_model_path, monitor='val_loss', verbose=1,\n",
        "                      save_best_only=True, save_weights_only=False, mode='min', save_freq=\"epoch\")  # val_f1_metric"
      ],
      "metadata": {
        "id": "Lb60bddQXFgU"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Training Model"
      ],
      "metadata": {
        "id": "8AevTBfgOBlU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def createCNN():\n",
        "  # CNN model\n",
        "  model = Sequential()\n",
        "  # Conv2D\n",
        "  model.add(Conv2D(filters = 20,\n",
        "                  kernel_size = 2,\n",
        "                  strides = 1,\n",
        "                  padding = 'valid',\n",
        "                  activation='relu',\n",
        "                  use_bias = True,\n",
        "                  kernel_regularizer=regularizers.l2(0.0),\n",
        "                  kernel_initializer='glorot_uniform',\n",
        "                  input_shape=(params[\"input_w\"],\n",
        "                                params[\"input_h\"],\n",
        "                                params[\"input_c\"])\n",
        "                    ))\n",
        "\n",
        "  # Dropout\n",
        "  model.add(Dropout(0.22))\n",
        "\n",
        "  # Conv2D\n",
        "  model.add(Conv2D(filters = 40,\n",
        "                  kernel_size = 2,\n",
        "                  strides = 2,\n",
        "                  padding = 'valid',\n",
        "                  activation='relu',\n",
        "                  kernel_regularizer=regularizers.l2(0.0),\n",
        "                  kernel_initializer='glorot_uniform',\n",
        "                  ))\n",
        "\n",
        "  # Dropout\n",
        "  model.add(Dropout(0.05))\n",
        "\n",
        "  # Flatten\n",
        "  model.add(Flatten())\n",
        "\n",
        "  # Dense\n",
        "  model.add(Dense(100, activation='relu'))\n",
        "\n",
        "  # Dropout\n",
        "  model.add(Dropout(0.22))\n",
        "\n",
        "  # Dense\n",
        "  model.add(Dense(3, activation='softmax'))\n",
        "\n",
        "  # Compile Model\n",
        "  model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy', 'mae', 'mse'])\n",
        "\n",
        "  return model"
      ],
      "metadata": {
        "id": "vc5nMsL6kIpw"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(f\"X_train Images shape: {X_train_split.shape}\")\n",
        "print(f\"y_train labels shape: {y_train_split.shape}\\n\")\n",
        "\n",
        "print(f\"X_val Images shape: {X_val.shape}\")\n",
        "print(f\"y_val labels shape: {y_val.shape}\\n\")\n",
        "\n",
        "print(f\"X_test Images shape: {X_test_final.shape}\")\n",
        "print(f\"y_test labels shape: {y_test_animal.shape}\\n\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fsJhJSAioawk",
        "outputId": "655bd1c9-ecde-47dc-fa65-aab20f32a2ff"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X_train Images shape: (12000, 15, 15, 3)\n",
            "y_train labels shape: (12000, 3)\n",
            "\n",
            "X_val Images shape: (3000, 15, 15, 3)\n",
            "y_val labels shape: (3000, 3)\n",
            "\n",
            "X_test Images shape: (3000, 15, 15, 3)\n",
            "y_test labels shape: (3000, 3)\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "trained_model = createCNN()"
      ],
      "metadata": {
        "id": "z83l9aywkTWp"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "hist = trained_model.fit(X_train_split, y_train_split,\n",
        "          batch_size=params[\"batch_size\"],\n",
        "          epochs=params[\"epochs\"],\n",
        "          callbacks=[mcp, rlp],\n",
        "          validation_data = (X_val, y_val),\n",
        "          verbose=2)"
      ],
      "metadata": {
        "id": "qNBCC2GxN816",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "be98330d-6f35-4fe7-9067-11e1eaf5d743"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/2000\n",
            "\n",
            "Epoch 1: val_loss improved from inf to 1.03081, saving model to models/animal_cnn.h5\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py:3103: UserWarning: You are saving your model as an HDF5 file via `model.save()`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')`.\n",
            "  saving_api.save_model(\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1;30;43mStreaming output truncated to the last 5000 lines.\u001b[0m\n",
            "Epoch 751: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0975 - accuracy: 0.9673 - mae: 0.0441 - mse: 0.0172 - val_loss: 0.7987 - val_accuracy: 0.7780 - val_mae: 0.1631 - val_mse: 0.1166 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 752/2000\n",
            "\n",
            "Epoch 752: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0887 - accuracy: 0.9718 - mae: 0.0412 - mse: 0.0153 - val_loss: 0.7987 - val_accuracy: 0.7750 - val_mae: 0.1625 - val_mse: 0.1164 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 753/2000\n",
            "\n",
            "Epoch 753: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0942 - accuracy: 0.9670 - mae: 0.0433 - mse: 0.0166 - val_loss: 0.8025 - val_accuracy: 0.7760 - val_mae: 0.1623 - val_mse: 0.1164 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 754/2000\n",
            "\n",
            "Epoch 754: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.1013 - accuracy: 0.9645 - mae: 0.0445 - mse: 0.0179 - val_loss: 0.8016 - val_accuracy: 0.7763 - val_mae: 0.1626 - val_mse: 0.1164 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 755/2000\n",
            "\n",
            "Epoch 755: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0968 - accuracy: 0.9679 - mae: 0.0435 - mse: 0.0171 - val_loss: 0.7954 - val_accuracy: 0.7777 - val_mae: 0.1626 - val_mse: 0.1161 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 756/2000\n",
            "\n",
            "Epoch 756: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0938 - accuracy: 0.9699 - mae: 0.0432 - mse: 0.0161 - val_loss: 0.7960 - val_accuracy: 0.7747 - val_mae: 0.1630 - val_mse: 0.1162 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 757/2000\n",
            "\n",
            "Epoch 757: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0918 - accuracy: 0.9707 - mae: 0.0421 - mse: 0.0160 - val_loss: 0.8012 - val_accuracy: 0.7747 - val_mae: 0.1628 - val_mse: 0.1164 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 758/2000\n",
            "\n",
            "Epoch 758: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0955 - accuracy: 0.9689 - mae: 0.0430 - mse: 0.0166 - val_loss: 0.7995 - val_accuracy: 0.7737 - val_mae: 0.1624 - val_mse: 0.1161 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 759/2000\n",
            "\n",
            "Epoch 759: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0912 - accuracy: 0.9691 - mae: 0.0420 - mse: 0.0162 - val_loss: 0.7990 - val_accuracy: 0.7747 - val_mae: 0.1625 - val_mse: 0.1161 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 760/2000\n",
            "\n",
            "Epoch 760: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0941 - accuracy: 0.9679 - mae: 0.0424 - mse: 0.0164 - val_loss: 0.7964 - val_accuracy: 0.7757 - val_mae: 0.1624 - val_mse: 0.1160 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 761/2000\n",
            "\n",
            "Epoch 761: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0948 - accuracy: 0.9681 - mae: 0.0434 - mse: 0.0168 - val_loss: 0.8020 - val_accuracy: 0.7747 - val_mae: 0.1624 - val_mse: 0.1164 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 762/2000\n",
            "\n",
            "Epoch 762: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0948 - accuracy: 0.9699 - mae: 0.0431 - mse: 0.0165 - val_loss: 0.7990 - val_accuracy: 0.7743 - val_mae: 0.1622 - val_mse: 0.1160 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 763/2000\n",
            "\n",
            "Epoch 763: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0945 - accuracy: 0.9686 - mae: 0.0436 - mse: 0.0167 - val_loss: 0.7972 - val_accuracy: 0.7747 - val_mae: 0.1623 - val_mse: 0.1160 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 764/2000\n",
            "\n",
            "Epoch 764: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0953 - accuracy: 0.9676 - mae: 0.0432 - mse: 0.0166 - val_loss: 0.8001 - val_accuracy: 0.7750 - val_mae: 0.1622 - val_mse: 0.1160 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 765/2000\n",
            "\n",
            "Epoch 765: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0927 - accuracy: 0.9694 - mae: 0.0426 - mse: 0.0161 - val_loss: 0.8066 - val_accuracy: 0.7730 - val_mae: 0.1623 - val_mse: 0.1166 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 766/2000\n",
            "\n",
            "Epoch 766: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0938 - accuracy: 0.9685 - mae: 0.0428 - mse: 0.0166 - val_loss: 0.8090 - val_accuracy: 0.7733 - val_mae: 0.1617 - val_mse: 0.1163 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 767/2000\n",
            "\n",
            "Epoch 767: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0918 - accuracy: 0.9705 - mae: 0.0422 - mse: 0.0162 - val_loss: 0.8106 - val_accuracy: 0.7737 - val_mae: 0.1617 - val_mse: 0.1164 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 768/2000\n",
            "\n",
            "Epoch 768: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0902 - accuracy: 0.9692 - mae: 0.0416 - mse: 0.0159 - val_loss: 0.8076 - val_accuracy: 0.7753 - val_mae: 0.1620 - val_mse: 0.1164 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 769/2000\n",
            "\n",
            "Epoch 769: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0915 - accuracy: 0.9691 - mae: 0.0425 - mse: 0.0162 - val_loss: 0.8077 - val_accuracy: 0.7743 - val_mae: 0.1619 - val_mse: 0.1163 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 770/2000\n",
            "\n",
            "Epoch 770: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0903 - accuracy: 0.9699 - mae: 0.0411 - mse: 0.0159 - val_loss: 0.8088 - val_accuracy: 0.7737 - val_mae: 0.1617 - val_mse: 0.1164 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 771/2000\n",
            "\n",
            "Epoch 771: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0896 - accuracy: 0.9695 - mae: 0.0410 - mse: 0.0157 - val_loss: 0.8058 - val_accuracy: 0.7753 - val_mae: 0.1618 - val_mse: 0.1161 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 772/2000\n",
            "\n",
            "Epoch 772: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0912 - accuracy: 0.9699 - mae: 0.0411 - mse: 0.0159 - val_loss: 0.8099 - val_accuracy: 0.7753 - val_mae: 0.1617 - val_mse: 0.1163 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 773/2000\n",
            "\n",
            "Epoch 773: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0926 - accuracy: 0.9703 - mae: 0.0417 - mse: 0.0162 - val_loss: 0.8111 - val_accuracy: 0.7747 - val_mae: 0.1619 - val_mse: 0.1165 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 774/2000\n",
            "\n",
            "Epoch 774: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0938 - accuracy: 0.9688 - mae: 0.0420 - mse: 0.0162 - val_loss: 0.8094 - val_accuracy: 0.7747 - val_mae: 0.1619 - val_mse: 0.1164 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 775/2000\n",
            "\n",
            "Epoch 775: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0920 - accuracy: 0.9694 - mae: 0.0423 - mse: 0.0161 - val_loss: 0.8109 - val_accuracy: 0.7763 - val_mae: 0.1618 - val_mse: 0.1165 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 776/2000\n",
            "\n",
            "Epoch 776: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0868 - accuracy: 0.9713 - mae: 0.0400 - mse: 0.0149 - val_loss: 0.8079 - val_accuracy: 0.7743 - val_mae: 0.1619 - val_mse: 0.1164 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 777/2000\n",
            "\n",
            "Epoch 777: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0926 - accuracy: 0.9689 - mae: 0.0416 - mse: 0.0160 - val_loss: 0.8109 - val_accuracy: 0.7767 - val_mae: 0.1615 - val_mse: 0.1164 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 778/2000\n",
            "\n",
            "Epoch 778: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0936 - accuracy: 0.9686 - mae: 0.0414 - mse: 0.0163 - val_loss: 0.8098 - val_accuracy: 0.7780 - val_mae: 0.1618 - val_mse: 0.1164 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 779/2000\n",
            "\n",
            "Epoch 779: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0881 - accuracy: 0.9709 - mae: 0.0404 - mse: 0.0152 - val_loss: 0.8114 - val_accuracy: 0.7727 - val_mae: 0.1624 - val_mse: 0.1168 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 780/2000\n",
            "\n",
            "Epoch 780: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0861 - accuracy: 0.9735 - mae: 0.0396 - mse: 0.0146 - val_loss: 0.8160 - val_accuracy: 0.7740 - val_mae: 0.1622 - val_mse: 0.1171 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 781/2000\n",
            "\n",
            "Epoch 781: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0915 - accuracy: 0.9713 - mae: 0.0414 - mse: 0.0158 - val_loss: 0.8156 - val_accuracy: 0.7700 - val_mae: 0.1622 - val_mse: 0.1172 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 782/2000\n",
            "\n",
            "Epoch 782: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0922 - accuracy: 0.9687 - mae: 0.0418 - mse: 0.0164 - val_loss: 0.8176 - val_accuracy: 0.7753 - val_mae: 0.1621 - val_mse: 0.1170 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 783/2000\n",
            "\n",
            "Epoch 783: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0877 - accuracy: 0.9709 - mae: 0.0403 - mse: 0.0153 - val_loss: 0.8176 - val_accuracy: 0.7733 - val_mae: 0.1619 - val_mse: 0.1169 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 784/2000\n",
            "\n",
            "Epoch 784: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0905 - accuracy: 0.9697 - mae: 0.0417 - mse: 0.0159 - val_loss: 0.8146 - val_accuracy: 0.7740 - val_mae: 0.1619 - val_mse: 0.1167 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 785/2000\n",
            "\n",
            "Epoch 785: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0870 - accuracy: 0.9713 - mae: 0.0397 - mse: 0.0149 - val_loss: 0.8114 - val_accuracy: 0.7750 - val_mae: 0.1617 - val_mse: 0.1165 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 786/2000\n",
            "\n",
            "Epoch 786: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0923 - accuracy: 0.9691 - mae: 0.0414 - mse: 0.0162 - val_loss: 0.8122 - val_accuracy: 0.7763 - val_mae: 0.1622 - val_mse: 0.1169 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 787/2000\n",
            "\n",
            "Epoch 787: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0908 - accuracy: 0.9706 - mae: 0.0411 - mse: 0.0156 - val_loss: 0.8057 - val_accuracy: 0.7760 - val_mae: 0.1622 - val_mse: 0.1165 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 788/2000\n",
            "\n",
            "Epoch 788: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0893 - accuracy: 0.9699 - mae: 0.0409 - mse: 0.0157 - val_loss: 0.8092 - val_accuracy: 0.7747 - val_mae: 0.1621 - val_mse: 0.1166 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 789/2000\n",
            "\n",
            "Epoch 789: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0907 - accuracy: 0.9708 - mae: 0.0408 - mse: 0.0156 - val_loss: 0.8065 - val_accuracy: 0.7753 - val_mae: 0.1620 - val_mse: 0.1164 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 790/2000\n",
            "\n",
            "Epoch 790: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0909 - accuracy: 0.9697 - mae: 0.0415 - mse: 0.0159 - val_loss: 0.8082 - val_accuracy: 0.7750 - val_mae: 0.1619 - val_mse: 0.1164 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 791/2000\n",
            "\n",
            "Epoch 791: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0865 - accuracy: 0.9713 - mae: 0.0405 - mse: 0.0151 - val_loss: 0.8147 - val_accuracy: 0.7747 - val_mae: 0.1618 - val_mse: 0.1167 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 792/2000\n",
            "\n",
            "Epoch 792: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0858 - accuracy: 0.9724 - mae: 0.0396 - mse: 0.0151 - val_loss: 0.8137 - val_accuracy: 0.7747 - val_mae: 0.1616 - val_mse: 0.1166 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 793/2000\n",
            "\n",
            "Epoch 793: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0886 - accuracy: 0.9718 - mae: 0.0404 - mse: 0.0154 - val_loss: 0.8190 - val_accuracy: 0.7770 - val_mae: 0.1620 - val_mse: 0.1171 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 794/2000\n",
            "\n",
            "Epoch 794: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0895 - accuracy: 0.9709 - mae: 0.0407 - mse: 0.0157 - val_loss: 0.8103 - val_accuracy: 0.7747 - val_mae: 0.1618 - val_mse: 0.1166 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 795/2000\n",
            "\n",
            "Epoch 795: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0851 - accuracy: 0.9722 - mae: 0.0396 - mse: 0.0148 - val_loss: 0.8183 - val_accuracy: 0.7760 - val_mae: 0.1615 - val_mse: 0.1168 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 796/2000\n",
            "\n",
            "Epoch 796: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0873 - accuracy: 0.9710 - mae: 0.0395 - mse: 0.0153 - val_loss: 0.8197 - val_accuracy: 0.7750 - val_mae: 0.1613 - val_mse: 0.1167 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 797/2000\n",
            "\n",
            "Epoch 797: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0900 - accuracy: 0.9703 - mae: 0.0401 - mse: 0.0155 - val_loss: 0.8159 - val_accuracy: 0.7750 - val_mae: 0.1613 - val_mse: 0.1165 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 798/2000\n",
            "\n",
            "Epoch 798: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0895 - accuracy: 0.9703 - mae: 0.0402 - mse: 0.0157 - val_loss: 0.8175 - val_accuracy: 0.7787 - val_mae: 0.1616 - val_mse: 0.1167 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 799/2000\n",
            "\n",
            "Epoch 799: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0893 - accuracy: 0.9716 - mae: 0.0397 - mse: 0.0154 - val_loss: 0.8126 - val_accuracy: 0.7750 - val_mae: 0.1617 - val_mse: 0.1166 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 800/2000\n",
            "\n",
            "Epoch 800: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0907 - accuracy: 0.9690 - mae: 0.0413 - mse: 0.0162 - val_loss: 0.8116 - val_accuracy: 0.7763 - val_mae: 0.1618 - val_mse: 0.1166 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 801/2000\n",
            "\n",
            "Epoch 801: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0830 - accuracy: 0.9722 - mae: 0.0388 - mse: 0.0145 - val_loss: 0.8124 - val_accuracy: 0.7753 - val_mae: 0.1615 - val_mse: 0.1164 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 802/2000\n",
            "\n",
            "Epoch 802: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0876 - accuracy: 0.9711 - mae: 0.0395 - mse: 0.0153 - val_loss: 0.8166 - val_accuracy: 0.7753 - val_mae: 0.1613 - val_mse: 0.1164 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 803/2000\n",
            "\n",
            "Epoch 803: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0917 - accuracy: 0.9704 - mae: 0.0411 - mse: 0.0160 - val_loss: 0.8150 - val_accuracy: 0.7750 - val_mae: 0.1616 - val_mse: 0.1165 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 804/2000\n",
            "\n",
            "Epoch 804: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0845 - accuracy: 0.9724 - mae: 0.0387 - mse: 0.0145 - val_loss: 0.8197 - val_accuracy: 0.7743 - val_mae: 0.1619 - val_mse: 0.1170 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 805/2000\n",
            "\n",
            "Epoch 805: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0816 - accuracy: 0.9730 - mae: 0.0383 - mse: 0.0142 - val_loss: 0.8204 - val_accuracy: 0.7747 - val_mae: 0.1616 - val_mse: 0.1171 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 806/2000\n",
            "\n",
            "Epoch 806: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0917 - accuracy: 0.9703 - mae: 0.0413 - mse: 0.0163 - val_loss: 0.8235 - val_accuracy: 0.7753 - val_mae: 0.1617 - val_mse: 0.1173 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 807/2000\n",
            "\n",
            "Epoch 807: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0857 - accuracy: 0.9738 - mae: 0.0388 - mse: 0.0147 - val_loss: 0.8186 - val_accuracy: 0.7743 - val_mae: 0.1616 - val_mse: 0.1170 - lr: 1.0000e-04 - 282ms/epoch - 24ms/step\n",
            "Epoch 808/2000\n",
            "\n",
            "Epoch 808: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0857 - accuracy: 0.9710 - mae: 0.0391 - mse: 0.0150 - val_loss: 0.8189 - val_accuracy: 0.7757 - val_mae: 0.1617 - val_mse: 0.1169 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 809/2000\n",
            "\n",
            "Epoch 809: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0803 - accuracy: 0.9738 - mae: 0.0378 - mse: 0.0136 - val_loss: 0.8201 - val_accuracy: 0.7740 - val_mae: 0.1617 - val_mse: 0.1171 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 810/2000\n",
            "\n",
            "Epoch 810: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0858 - accuracy: 0.9718 - mae: 0.0395 - mse: 0.0150 - val_loss: 0.8251 - val_accuracy: 0.7760 - val_mae: 0.1617 - val_mse: 0.1173 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 811/2000\n",
            "\n",
            "Epoch 811: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0887 - accuracy: 0.9706 - mae: 0.0400 - mse: 0.0154 - val_loss: 0.8231 - val_accuracy: 0.7753 - val_mae: 0.1618 - val_mse: 0.1175 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 812/2000\n",
            "\n",
            "Epoch 812: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0868 - accuracy: 0.9703 - mae: 0.0398 - mse: 0.0152 - val_loss: 0.8193 - val_accuracy: 0.7740 - val_mae: 0.1613 - val_mse: 0.1168 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 813/2000\n",
            "\n",
            "Epoch 813: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0879 - accuracy: 0.9707 - mae: 0.0395 - mse: 0.0152 - val_loss: 0.8181 - val_accuracy: 0.7723 - val_mae: 0.1616 - val_mse: 0.1170 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 814/2000\n",
            "\n",
            "Epoch 814: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0823 - accuracy: 0.9736 - mae: 0.0380 - mse: 0.0141 - val_loss: 0.8189 - val_accuracy: 0.7763 - val_mae: 0.1615 - val_mse: 0.1170 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 815/2000\n",
            "\n",
            "Epoch 815: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0859 - accuracy: 0.9708 - mae: 0.0392 - mse: 0.0151 - val_loss: 0.8185 - val_accuracy: 0.7743 - val_mae: 0.1615 - val_mse: 0.1168 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 816/2000\n",
            "\n",
            "Epoch 816: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0849 - accuracy: 0.9722 - mae: 0.0390 - mse: 0.0149 - val_loss: 0.8193 - val_accuracy: 0.7750 - val_mae: 0.1614 - val_mse: 0.1169 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 817/2000\n",
            "\n",
            "Epoch 817: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0883 - accuracy: 0.9700 - mae: 0.0398 - mse: 0.0157 - val_loss: 0.8216 - val_accuracy: 0.7740 - val_mae: 0.1616 - val_mse: 0.1170 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 818/2000\n",
            "\n",
            "Epoch 818: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0906 - accuracy: 0.9700 - mae: 0.0408 - mse: 0.0159 - val_loss: 0.8184 - val_accuracy: 0.7763 - val_mae: 0.1617 - val_mse: 0.1169 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 819/2000\n",
            "\n",
            "Epoch 819: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0843 - accuracy: 0.9716 - mae: 0.0390 - mse: 0.0147 - val_loss: 0.8212 - val_accuracy: 0.7743 - val_mae: 0.1615 - val_mse: 0.1170 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 820/2000\n",
            "\n",
            "Epoch 820: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0877 - accuracy: 0.9720 - mae: 0.0397 - mse: 0.0152 - val_loss: 0.8216 - val_accuracy: 0.7733 - val_mae: 0.1620 - val_mse: 0.1173 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 821/2000\n",
            "\n",
            "Epoch 821: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0875 - accuracy: 0.9726 - mae: 0.0395 - mse: 0.0151 - val_loss: 0.8207 - val_accuracy: 0.7730 - val_mae: 0.1624 - val_mse: 0.1176 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 822/2000\n",
            "\n",
            "Epoch 822: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0846 - accuracy: 0.9728 - mae: 0.0392 - mse: 0.0146 - val_loss: 0.8258 - val_accuracy: 0.7727 - val_mae: 0.1620 - val_mse: 0.1176 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 823/2000\n",
            "\n",
            "Epoch 823: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0874 - accuracy: 0.9712 - mae: 0.0394 - mse: 0.0153 - val_loss: 0.8213 - val_accuracy: 0.7733 - val_mae: 0.1617 - val_mse: 0.1171 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 824/2000\n",
            "\n",
            "Epoch 824: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0844 - accuracy: 0.9725 - mae: 0.0390 - mse: 0.0147 - val_loss: 0.8229 - val_accuracy: 0.7727 - val_mae: 0.1620 - val_mse: 0.1174 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 825/2000\n",
            "\n",
            "Epoch 825: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0883 - accuracy: 0.9715 - mae: 0.0398 - mse: 0.0155 - val_loss: 0.8221 - val_accuracy: 0.7740 - val_mae: 0.1616 - val_mse: 0.1171 - lr: 1.0000e-04 - 320ms/epoch - 27ms/step\n",
            "Epoch 826/2000\n",
            "\n",
            "Epoch 826: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0885 - accuracy: 0.9705 - mae: 0.0400 - mse: 0.0156 - val_loss: 0.8220 - val_accuracy: 0.7750 - val_mae: 0.1614 - val_mse: 0.1170 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 827/2000\n",
            "\n",
            "Epoch 827: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0883 - accuracy: 0.9695 - mae: 0.0398 - mse: 0.0156 - val_loss: 0.8237 - val_accuracy: 0.7730 - val_mae: 0.1614 - val_mse: 0.1171 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 828/2000\n",
            "\n",
            "Epoch 828: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0890 - accuracy: 0.9690 - mae: 0.0403 - mse: 0.0159 - val_loss: 0.8243 - val_accuracy: 0.7740 - val_mae: 0.1614 - val_mse: 0.1171 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 829/2000\n",
            "\n",
            "Epoch 829: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0892 - accuracy: 0.9708 - mae: 0.0394 - mse: 0.0156 - val_loss: 0.8249 - val_accuracy: 0.7747 - val_mae: 0.1617 - val_mse: 0.1174 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 830/2000\n",
            "\n",
            "Epoch 830: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0828 - accuracy: 0.9736 - mae: 0.0372 - mse: 0.0141 - val_loss: 0.8234 - val_accuracy: 0.7757 - val_mae: 0.1619 - val_mse: 0.1175 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 831/2000\n",
            "\n",
            "Epoch 831: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0788 - accuracy: 0.9748 - mae: 0.0373 - mse: 0.0137 - val_loss: 0.8227 - val_accuracy: 0.7727 - val_mae: 0.1617 - val_mse: 0.1172 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 832/2000\n",
            "\n",
            "Epoch 832: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0810 - accuracy: 0.9732 - mae: 0.0370 - mse: 0.0139 - val_loss: 0.8299 - val_accuracy: 0.7733 - val_mae: 0.1614 - val_mse: 0.1176 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 833/2000\n",
            "\n",
            "Epoch 833: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0833 - accuracy: 0.9728 - mae: 0.0383 - mse: 0.0145 - val_loss: 0.8322 - val_accuracy: 0.7740 - val_mae: 0.1612 - val_mse: 0.1176 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 834/2000\n",
            "\n",
            "Epoch 834: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0856 - accuracy: 0.9706 - mae: 0.0390 - mse: 0.0151 - val_loss: 0.8307 - val_accuracy: 0.7740 - val_mae: 0.1612 - val_mse: 0.1176 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 835/2000\n",
            "\n",
            "Epoch 835: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0798 - accuracy: 0.9752 - mae: 0.0372 - mse: 0.0137 - val_loss: 0.8285 - val_accuracy: 0.7700 - val_mae: 0.1616 - val_mse: 0.1178 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 836/2000\n",
            "\n",
            "Epoch 836: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0823 - accuracy: 0.9740 - mae: 0.0373 - mse: 0.0142 - val_loss: 0.8272 - val_accuracy: 0.7743 - val_mae: 0.1620 - val_mse: 0.1177 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 837/2000\n",
            "\n",
            "Epoch 837: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0820 - accuracy: 0.9734 - mae: 0.0380 - mse: 0.0142 - val_loss: 0.8317 - val_accuracy: 0.7737 - val_mae: 0.1615 - val_mse: 0.1178 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 838/2000\n",
            "\n",
            "Epoch 838: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0803 - accuracy: 0.9743 - mae: 0.0371 - mse: 0.0139 - val_loss: 0.8339 - val_accuracy: 0.7750 - val_mae: 0.1608 - val_mse: 0.1175 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 839/2000\n",
            "\n",
            "Epoch 839: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0807 - accuracy: 0.9737 - mae: 0.0373 - mse: 0.0142 - val_loss: 0.8350 - val_accuracy: 0.7723 - val_mae: 0.1613 - val_mse: 0.1179 - lr: 1.0000e-04 - 306ms/epoch - 25ms/step\n",
            "Epoch 840/2000\n",
            "\n",
            "Epoch 840: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0838 - accuracy: 0.9707 - mae: 0.0380 - mse: 0.0148 - val_loss: 0.8334 - val_accuracy: 0.7707 - val_mae: 0.1614 - val_mse: 0.1177 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 841/2000\n",
            "\n",
            "Epoch 841: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0813 - accuracy: 0.9740 - mae: 0.0364 - mse: 0.0139 - val_loss: 0.8290 - val_accuracy: 0.7717 - val_mae: 0.1618 - val_mse: 0.1177 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 842/2000\n",
            "\n",
            "Epoch 842: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0812 - accuracy: 0.9731 - mae: 0.0375 - mse: 0.0141 - val_loss: 0.8290 - val_accuracy: 0.7700 - val_mae: 0.1615 - val_mse: 0.1175 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 843/2000\n",
            "\n",
            "Epoch 843: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0823 - accuracy: 0.9718 - mae: 0.0376 - mse: 0.0145 - val_loss: 0.8341 - val_accuracy: 0.7740 - val_mae: 0.1610 - val_mse: 0.1176 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 844/2000\n",
            "\n",
            "Epoch 844: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0813 - accuracy: 0.9739 - mae: 0.0366 - mse: 0.0141 - val_loss: 0.8326 - val_accuracy: 0.7710 - val_mae: 0.1610 - val_mse: 0.1174 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 845/2000\n",
            "\n",
            "Epoch 845: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0786 - accuracy: 0.9757 - mae: 0.0361 - mse: 0.0134 - val_loss: 0.8379 - val_accuracy: 0.7733 - val_mae: 0.1614 - val_mse: 0.1180 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 846/2000\n",
            "\n",
            "Epoch 846: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0776 - accuracy: 0.9758 - mae: 0.0362 - mse: 0.0134 - val_loss: 0.8333 - val_accuracy: 0.7717 - val_mae: 0.1610 - val_mse: 0.1175 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 847/2000\n",
            "\n",
            "Epoch 847: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0796 - accuracy: 0.9740 - mae: 0.0369 - mse: 0.0138 - val_loss: 0.8356 - val_accuracy: 0.7753 - val_mae: 0.1607 - val_mse: 0.1175 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 848/2000\n",
            "\n",
            "Epoch 848: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0798 - accuracy: 0.9738 - mae: 0.0365 - mse: 0.0138 - val_loss: 0.8306 - val_accuracy: 0.7737 - val_mae: 0.1607 - val_mse: 0.1171 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 849/2000\n",
            "\n",
            "Epoch 849: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0819 - accuracy: 0.9733 - mae: 0.0369 - mse: 0.0142 - val_loss: 0.8321 - val_accuracy: 0.7737 - val_mae: 0.1607 - val_mse: 0.1173 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 850/2000\n",
            "\n",
            "Epoch 850: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0831 - accuracy: 0.9732 - mae: 0.0374 - mse: 0.0145 - val_loss: 0.8327 - val_accuracy: 0.7720 - val_mae: 0.1611 - val_mse: 0.1178 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 851/2000\n",
            "\n",
            "Epoch 851: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0826 - accuracy: 0.9731 - mae: 0.0374 - mse: 0.0143 - val_loss: 0.8329 - val_accuracy: 0.7727 - val_mae: 0.1614 - val_mse: 0.1179 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 852/2000\n",
            "\n",
            "Epoch 852: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0859 - accuracy: 0.9738 - mae: 0.0384 - mse: 0.0149 - val_loss: 0.8328 - val_accuracy: 0.7737 - val_mae: 0.1618 - val_mse: 0.1180 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 853/2000\n",
            "\n",
            "Epoch 853: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0803 - accuracy: 0.9753 - mae: 0.0364 - mse: 0.0137 - val_loss: 0.8301 - val_accuracy: 0.7747 - val_mae: 0.1615 - val_mse: 0.1176 - lr: 1.0000e-04 - 313ms/epoch - 26ms/step\n",
            "Epoch 854/2000\n",
            "\n",
            "Epoch 854: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0841 - accuracy: 0.9737 - mae: 0.0383 - mse: 0.0148 - val_loss: 0.8329 - val_accuracy: 0.7710 - val_mae: 0.1616 - val_mse: 0.1177 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 855/2000\n",
            "\n",
            "Epoch 855: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0780 - accuracy: 0.9756 - mae: 0.0367 - mse: 0.0133 - val_loss: 0.8325 - val_accuracy: 0.7713 - val_mae: 0.1614 - val_mse: 0.1177 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 856/2000\n",
            "\n",
            "Epoch 856: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0812 - accuracy: 0.9733 - mae: 0.0370 - mse: 0.0142 - val_loss: 0.8372 - val_accuracy: 0.7737 - val_mae: 0.1611 - val_mse: 0.1178 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 857/2000\n",
            "\n",
            "Epoch 857: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0806 - accuracy: 0.9735 - mae: 0.0370 - mse: 0.0140 - val_loss: 0.8355 - val_accuracy: 0.7743 - val_mae: 0.1615 - val_mse: 0.1180 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 858/2000\n",
            "\n",
            "Epoch 858: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0821 - accuracy: 0.9720 - mae: 0.0375 - mse: 0.0146 - val_loss: 0.8332 - val_accuracy: 0.7737 - val_mae: 0.1612 - val_mse: 0.1176 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 859/2000\n",
            "\n",
            "Epoch 859: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0827 - accuracy: 0.9731 - mae: 0.0378 - mse: 0.0145 - val_loss: 0.8378 - val_accuracy: 0.7770 - val_mae: 0.1610 - val_mse: 0.1178 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 860/2000\n",
            "\n",
            "Epoch 860: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0833 - accuracy: 0.9722 - mae: 0.0378 - mse: 0.0146 - val_loss: 0.8375 - val_accuracy: 0.7760 - val_mae: 0.1609 - val_mse: 0.1176 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 861/2000\n",
            "\n",
            "Epoch 861: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0788 - accuracy: 0.9747 - mae: 0.0363 - mse: 0.0137 - val_loss: 0.8368 - val_accuracy: 0.7737 - val_mae: 0.1608 - val_mse: 0.1176 - lr: 1.0000e-04 - 282ms/epoch - 23ms/step\n",
            "Epoch 862/2000\n",
            "\n",
            "Epoch 862: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0834 - accuracy: 0.9717 - mae: 0.0379 - mse: 0.0147 - val_loss: 0.8383 - val_accuracy: 0.7740 - val_mae: 0.1605 - val_mse: 0.1175 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 863/2000\n",
            "\n",
            "Epoch 863: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0758 - accuracy: 0.9748 - mae: 0.0353 - mse: 0.0132 - val_loss: 0.8402 - val_accuracy: 0.7740 - val_mae: 0.1604 - val_mse: 0.1176 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 864/2000\n",
            "\n",
            "Epoch 864: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0787 - accuracy: 0.9731 - mae: 0.0365 - mse: 0.0137 - val_loss: 0.8361 - val_accuracy: 0.7733 - val_mae: 0.1608 - val_mse: 0.1176 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 865/2000\n",
            "\n",
            "Epoch 865: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0802 - accuracy: 0.9732 - mae: 0.0366 - mse: 0.0141 - val_loss: 0.8409 - val_accuracy: 0.7733 - val_mae: 0.1608 - val_mse: 0.1177 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 866/2000\n",
            "\n",
            "Epoch 866: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0815 - accuracy: 0.9731 - mae: 0.0373 - mse: 0.0142 - val_loss: 0.8338 - val_accuracy: 0.7723 - val_mae: 0.1611 - val_mse: 0.1175 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 867/2000\n",
            "\n",
            "Epoch 867: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0805 - accuracy: 0.9709 - mae: 0.0374 - mse: 0.0145 - val_loss: 0.8400 - val_accuracy: 0.7737 - val_mae: 0.1610 - val_mse: 0.1179 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 868/2000\n",
            "\n",
            "Epoch 868: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0801 - accuracy: 0.9736 - mae: 0.0369 - mse: 0.0140 - val_loss: 0.8386 - val_accuracy: 0.7747 - val_mae: 0.1611 - val_mse: 0.1180 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 869/2000\n",
            "\n",
            "Epoch 869: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0799 - accuracy: 0.9742 - mae: 0.0371 - mse: 0.0140 - val_loss: 0.8389 - val_accuracy: 0.7727 - val_mae: 0.1610 - val_mse: 0.1179 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 870/2000\n",
            "\n",
            "Epoch 870: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0772 - accuracy: 0.9743 - mae: 0.0361 - mse: 0.0136 - val_loss: 0.8400 - val_accuracy: 0.7720 - val_mae: 0.1607 - val_mse: 0.1177 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 871/2000\n",
            "\n",
            "Epoch 871: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0789 - accuracy: 0.9746 - mae: 0.0358 - mse: 0.0136 - val_loss: 0.8398 - val_accuracy: 0.7707 - val_mae: 0.1610 - val_mse: 0.1179 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 872/2000\n",
            "\n",
            "Epoch 872: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0797 - accuracy: 0.9746 - mae: 0.0362 - mse: 0.0139 - val_loss: 0.8420 - val_accuracy: 0.7710 - val_mae: 0.1609 - val_mse: 0.1179 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 873/2000\n",
            "\n",
            "Epoch 873: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0799 - accuracy: 0.9738 - mae: 0.0368 - mse: 0.0140 - val_loss: 0.8406 - val_accuracy: 0.7723 - val_mae: 0.1610 - val_mse: 0.1178 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 874/2000\n",
            "\n",
            "Epoch 874: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0782 - accuracy: 0.9747 - mae: 0.0356 - mse: 0.0134 - val_loss: 0.8426 - val_accuracy: 0.7723 - val_mae: 0.1609 - val_mse: 0.1179 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 875/2000\n",
            "\n",
            "Epoch 875: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0802 - accuracy: 0.9743 - mae: 0.0366 - mse: 0.0138 - val_loss: 0.8405 - val_accuracy: 0.7720 - val_mae: 0.1614 - val_mse: 0.1181 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 876/2000\n",
            "\n",
            "Epoch 876: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0794 - accuracy: 0.9741 - mae: 0.0364 - mse: 0.0137 - val_loss: 0.8443 - val_accuracy: 0.7723 - val_mae: 0.1615 - val_mse: 0.1184 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 877/2000\n",
            "\n",
            "Epoch 877: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0764 - accuracy: 0.9772 - mae: 0.0353 - mse: 0.0130 - val_loss: 0.8410 - val_accuracy: 0.7703 - val_mae: 0.1609 - val_mse: 0.1179 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 878/2000\n",
            "\n",
            "Epoch 878: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0830 - accuracy: 0.9724 - mae: 0.0370 - mse: 0.0146 - val_loss: 0.8416 - val_accuracy: 0.7723 - val_mae: 0.1613 - val_mse: 0.1181 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 879/2000\n",
            "\n",
            "Epoch 879: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0763 - accuracy: 0.9777 - mae: 0.0354 - mse: 0.0130 - val_loss: 0.8409 - val_accuracy: 0.7733 - val_mae: 0.1608 - val_mse: 0.1178 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 880/2000\n",
            "\n",
            "Epoch 880: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0790 - accuracy: 0.9743 - mae: 0.0355 - mse: 0.0137 - val_loss: 0.8430 - val_accuracy: 0.7710 - val_mae: 0.1610 - val_mse: 0.1180 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 881/2000\n",
            "\n",
            "Epoch 881: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0801 - accuracy: 0.9727 - mae: 0.0365 - mse: 0.0139 - val_loss: 0.8438 - val_accuracy: 0.7727 - val_mae: 0.1606 - val_mse: 0.1179 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 882/2000\n",
            "\n",
            "Epoch 882: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0758 - accuracy: 0.9738 - mae: 0.0348 - mse: 0.0132 - val_loss: 0.8450 - val_accuracy: 0.7727 - val_mae: 0.1606 - val_mse: 0.1179 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 883/2000\n",
            "\n",
            "Epoch 883: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0827 - accuracy: 0.9738 - mae: 0.0365 - mse: 0.0143 - val_loss: 0.8458 - val_accuracy: 0.7720 - val_mae: 0.1607 - val_mse: 0.1181 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 884/2000\n",
            "\n",
            "Epoch 884: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0800 - accuracy: 0.9740 - mae: 0.0361 - mse: 0.0141 - val_loss: 0.8468 - val_accuracy: 0.7707 - val_mae: 0.1610 - val_mse: 0.1184 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 885/2000\n",
            "\n",
            "Epoch 885: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0788 - accuracy: 0.9729 - mae: 0.0364 - mse: 0.0137 - val_loss: 0.8399 - val_accuracy: 0.7730 - val_mae: 0.1612 - val_mse: 0.1181 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 886/2000\n",
            "\n",
            "Epoch 886: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0811 - accuracy: 0.9729 - mae: 0.0370 - mse: 0.0142 - val_loss: 0.8387 - val_accuracy: 0.7740 - val_mae: 0.1608 - val_mse: 0.1177 - lr: 1.0000e-04 - 306ms/epoch - 25ms/step\n",
            "Epoch 887/2000\n",
            "\n",
            "Epoch 887: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0764 - accuracy: 0.9765 - mae: 0.0354 - mse: 0.0130 - val_loss: 0.8403 - val_accuracy: 0.7743 - val_mae: 0.1603 - val_mse: 0.1174 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 888/2000\n",
            "\n",
            "Epoch 888: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0772 - accuracy: 0.9753 - mae: 0.0353 - mse: 0.0132 - val_loss: 0.8465 - val_accuracy: 0.7753 - val_mae: 0.1601 - val_mse: 0.1176 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 889/2000\n",
            "\n",
            "Epoch 889: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0771 - accuracy: 0.9753 - mae: 0.0355 - mse: 0.0135 - val_loss: 0.8474 - val_accuracy: 0.7723 - val_mae: 0.1603 - val_mse: 0.1179 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 890/2000\n",
            "\n",
            "Epoch 890: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0766 - accuracy: 0.9762 - mae: 0.0349 - mse: 0.0131 - val_loss: 0.8492 - val_accuracy: 0.7743 - val_mae: 0.1602 - val_mse: 0.1178 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 891/2000\n",
            "\n",
            "Epoch 891: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0734 - accuracy: 0.9746 - mae: 0.0338 - mse: 0.0127 - val_loss: 0.8473 - val_accuracy: 0.7730 - val_mae: 0.1601 - val_mse: 0.1177 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 892/2000\n",
            "\n",
            "Epoch 892: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0757 - accuracy: 0.9737 - mae: 0.0341 - mse: 0.0132 - val_loss: 0.8458 - val_accuracy: 0.7727 - val_mae: 0.1602 - val_mse: 0.1177 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 893/2000\n",
            "\n",
            "Epoch 893: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0735 - accuracy: 0.9765 - mae: 0.0342 - mse: 0.0127 - val_loss: 0.8494 - val_accuracy: 0.7743 - val_mae: 0.1600 - val_mse: 0.1178 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 894/2000\n",
            "\n",
            "Epoch 894: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0793 - accuracy: 0.9736 - mae: 0.0362 - mse: 0.0140 - val_loss: 0.8454 - val_accuracy: 0.7730 - val_mae: 0.1603 - val_mse: 0.1176 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 895/2000\n",
            "\n",
            "Epoch 895: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0735 - accuracy: 0.9760 - mae: 0.0343 - mse: 0.0126 - val_loss: 0.8422 - val_accuracy: 0.7713 - val_mae: 0.1608 - val_mse: 0.1178 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 896/2000\n",
            "\n",
            "Epoch 896: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0774 - accuracy: 0.9750 - mae: 0.0353 - mse: 0.0135 - val_loss: 0.8455 - val_accuracy: 0.7707 - val_mae: 0.1605 - val_mse: 0.1180 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 897/2000\n",
            "\n",
            "Epoch 897: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0730 - accuracy: 0.9770 - mae: 0.0342 - mse: 0.0125 - val_loss: 0.8445 - val_accuracy: 0.7717 - val_mae: 0.1605 - val_mse: 0.1180 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 898/2000\n",
            "\n",
            "Epoch 898: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0788 - accuracy: 0.9735 - mae: 0.0351 - mse: 0.0138 - val_loss: 0.8483 - val_accuracy: 0.7743 - val_mae: 0.1609 - val_mse: 0.1183 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 899/2000\n",
            "\n",
            "Epoch 899: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0788 - accuracy: 0.9743 - mae: 0.0357 - mse: 0.0138 - val_loss: 0.8439 - val_accuracy: 0.7713 - val_mae: 0.1609 - val_mse: 0.1181 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 900/2000\n",
            "\n",
            "Epoch 900: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0775 - accuracy: 0.9759 - mae: 0.0350 - mse: 0.0135 - val_loss: 0.8483 - val_accuracy: 0.7730 - val_mae: 0.1612 - val_mse: 0.1184 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 901/2000\n",
            "\n",
            "Epoch 901: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0768 - accuracy: 0.9750 - mae: 0.0352 - mse: 0.0133 - val_loss: 0.8505 - val_accuracy: 0.7747 - val_mae: 0.1611 - val_mse: 0.1186 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 902/2000\n",
            "\n",
            "Epoch 902: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0815 - accuracy: 0.9722 - mae: 0.0366 - mse: 0.0143 - val_loss: 0.8477 - val_accuracy: 0.7723 - val_mae: 0.1611 - val_mse: 0.1186 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 903/2000\n",
            "\n",
            "Epoch 903: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0699 - accuracy: 0.9781 - mae: 0.0334 - mse: 0.0119 - val_loss: 0.8504 - val_accuracy: 0.7747 - val_mae: 0.1607 - val_mse: 0.1183 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 904/2000\n",
            "\n",
            "Epoch 904: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0798 - accuracy: 0.9752 - mae: 0.0356 - mse: 0.0138 - val_loss: 0.8521 - val_accuracy: 0.7747 - val_mae: 0.1604 - val_mse: 0.1181 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 905/2000\n",
            "\n",
            "Epoch 905: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0795 - accuracy: 0.9734 - mae: 0.0358 - mse: 0.0138 - val_loss: 0.8511 - val_accuracy: 0.7733 - val_mae: 0.1603 - val_mse: 0.1181 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 906/2000\n",
            "\n",
            "Epoch 906: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0795 - accuracy: 0.9733 - mae: 0.0359 - mse: 0.0138 - val_loss: 0.8517 - val_accuracy: 0.7727 - val_mae: 0.1606 - val_mse: 0.1182 - lr: 1.0000e-04 - 313ms/epoch - 26ms/step\n",
            "Epoch 907/2000\n",
            "\n",
            "Epoch 907: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0773 - accuracy: 0.9754 - mae: 0.0353 - mse: 0.0134 - val_loss: 0.8460 - val_accuracy: 0.7720 - val_mae: 0.1607 - val_mse: 0.1179 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 908/2000\n",
            "\n",
            "Epoch 908: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0728 - accuracy: 0.9769 - mae: 0.0340 - mse: 0.0124 - val_loss: 0.8512 - val_accuracy: 0.7747 - val_mae: 0.1606 - val_mse: 0.1181 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 909/2000\n",
            "\n",
            "Epoch 909: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0812 - accuracy: 0.9734 - mae: 0.0364 - mse: 0.0141 - val_loss: 0.8505 - val_accuracy: 0.7753 - val_mae: 0.1607 - val_mse: 0.1182 - lr: 1.0000e-04 - 315ms/epoch - 26ms/step\n",
            "Epoch 910/2000\n",
            "\n",
            "Epoch 910: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0765 - accuracy: 0.9746 - mae: 0.0351 - mse: 0.0134 - val_loss: 0.8486 - val_accuracy: 0.7717 - val_mae: 0.1607 - val_mse: 0.1182 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 911/2000\n",
            "\n",
            "Epoch 911: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0756 - accuracy: 0.9750 - mae: 0.0344 - mse: 0.0131 - val_loss: 0.8506 - val_accuracy: 0.7763 - val_mae: 0.1607 - val_mse: 0.1183 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 912/2000\n",
            "\n",
            "Epoch 912: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0731 - accuracy: 0.9755 - mae: 0.0341 - mse: 0.0126 - val_loss: 0.8534 - val_accuracy: 0.7730 - val_mae: 0.1608 - val_mse: 0.1185 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 913/2000\n",
            "\n",
            "Epoch 913: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0758 - accuracy: 0.9741 - mae: 0.0346 - mse: 0.0134 - val_loss: 0.8530 - val_accuracy: 0.7730 - val_mae: 0.1606 - val_mse: 0.1184 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 914/2000\n",
            "\n",
            "Epoch 914: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0733 - accuracy: 0.9760 - mae: 0.0338 - mse: 0.0127 - val_loss: 0.8569 - val_accuracy: 0.7733 - val_mae: 0.1605 - val_mse: 0.1186 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 915/2000\n",
            "\n",
            "Epoch 915: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0724 - accuracy: 0.9777 - mae: 0.0335 - mse: 0.0123 - val_loss: 0.8588 - val_accuracy: 0.7740 - val_mae: 0.1602 - val_mse: 0.1186 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 916/2000\n",
            "\n",
            "Epoch 916: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0711 - accuracy: 0.9779 - mae: 0.0331 - mse: 0.0121 - val_loss: 0.8604 - val_accuracy: 0.7743 - val_mae: 0.1600 - val_mse: 0.1183 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 917/2000\n",
            "\n",
            "Epoch 917: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0751 - accuracy: 0.9752 - mae: 0.0344 - mse: 0.0131 - val_loss: 0.8536 - val_accuracy: 0.7723 - val_mae: 0.1601 - val_mse: 0.1181 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 918/2000\n",
            "\n",
            "Epoch 918: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0742 - accuracy: 0.9753 - mae: 0.0337 - mse: 0.0129 - val_loss: 0.8550 - val_accuracy: 0.7727 - val_mae: 0.1604 - val_mse: 0.1181 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 919/2000\n",
            "\n",
            "Epoch 919: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0748 - accuracy: 0.9760 - mae: 0.0346 - mse: 0.0130 - val_loss: 0.8536 - val_accuracy: 0.7713 - val_mae: 0.1608 - val_mse: 0.1184 - lr: 1.0000e-04 - 320ms/epoch - 27ms/step\n",
            "Epoch 920/2000\n",
            "\n",
            "Epoch 920: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0711 - accuracy: 0.9772 - mae: 0.0330 - mse: 0.0122 - val_loss: 0.8555 - val_accuracy: 0.7720 - val_mae: 0.1607 - val_mse: 0.1185 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 921/2000\n",
            "\n",
            "Epoch 921: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0740 - accuracy: 0.9757 - mae: 0.0336 - mse: 0.0127 - val_loss: 0.8564 - val_accuracy: 0.7733 - val_mae: 0.1607 - val_mse: 0.1185 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 922/2000\n",
            "\n",
            "Epoch 922: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0751 - accuracy: 0.9747 - mae: 0.0345 - mse: 0.0132 - val_loss: 0.8594 - val_accuracy: 0.7707 - val_mae: 0.1606 - val_mse: 0.1187 - lr: 1.0000e-04 - 315ms/epoch - 26ms/step\n",
            "Epoch 923/2000\n",
            "\n",
            "Epoch 923: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0775 - accuracy: 0.9744 - mae: 0.0345 - mse: 0.0136 - val_loss: 0.8570 - val_accuracy: 0.7707 - val_mae: 0.1607 - val_mse: 0.1187 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 924/2000\n",
            "\n",
            "Epoch 924: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0737 - accuracy: 0.9766 - mae: 0.0338 - mse: 0.0127 - val_loss: 0.8552 - val_accuracy: 0.7733 - val_mae: 0.1603 - val_mse: 0.1184 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 925/2000\n",
            "\n",
            "Epoch 925: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0752 - accuracy: 0.9747 - mae: 0.0349 - mse: 0.0133 - val_loss: 0.8547 - val_accuracy: 0.7730 - val_mae: 0.1608 - val_mse: 0.1186 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 926/2000\n",
            "\n",
            "Epoch 926: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0749 - accuracy: 0.9746 - mae: 0.0338 - mse: 0.0132 - val_loss: 0.8574 - val_accuracy: 0.7730 - val_mae: 0.1611 - val_mse: 0.1189 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 927/2000\n",
            "\n",
            "Epoch 927: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0747 - accuracy: 0.9752 - mae: 0.0337 - mse: 0.0128 - val_loss: 0.8576 - val_accuracy: 0.7713 - val_mae: 0.1608 - val_mse: 0.1187 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 928/2000\n",
            "\n",
            "Epoch 928: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0710 - accuracy: 0.9778 - mae: 0.0328 - mse: 0.0121 - val_loss: 0.8595 - val_accuracy: 0.7713 - val_mae: 0.1604 - val_mse: 0.1185 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 929/2000\n",
            "\n",
            "Epoch 929: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0724 - accuracy: 0.9768 - mae: 0.0326 - mse: 0.0125 - val_loss: 0.8589 - val_accuracy: 0.7740 - val_mae: 0.1601 - val_mse: 0.1183 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 930/2000\n",
            "\n",
            "Epoch 930: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0682 - accuracy: 0.9793 - mae: 0.0320 - mse: 0.0116 - val_loss: 0.8636 - val_accuracy: 0.7710 - val_mae: 0.1600 - val_mse: 0.1185 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 931/2000\n",
            "\n",
            "Epoch 931: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0722 - accuracy: 0.9763 - mae: 0.0331 - mse: 0.0126 - val_loss: 0.8598 - val_accuracy: 0.7747 - val_mae: 0.1600 - val_mse: 0.1184 - lr: 1.0000e-04 - 314ms/epoch - 26ms/step\n",
            "Epoch 932/2000\n",
            "\n",
            "Epoch 932: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0786 - accuracy: 0.9735 - mae: 0.0351 - mse: 0.0138 - val_loss: 0.8597 - val_accuracy: 0.7723 - val_mae: 0.1606 - val_mse: 0.1188 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 933/2000\n",
            "\n",
            "Epoch 933: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0772 - accuracy: 0.9747 - mae: 0.0345 - mse: 0.0135 - val_loss: 0.8570 - val_accuracy: 0.7730 - val_mae: 0.1610 - val_mse: 0.1187 - lr: 1.0000e-04 - 314ms/epoch - 26ms/step\n",
            "Epoch 934/2000\n",
            "\n",
            "Epoch 934: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0711 - accuracy: 0.9766 - mae: 0.0326 - mse: 0.0122 - val_loss: 0.8576 - val_accuracy: 0.7720 - val_mae: 0.1609 - val_mse: 0.1187 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 935/2000\n",
            "\n",
            "Epoch 935: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0732 - accuracy: 0.9771 - mae: 0.0332 - mse: 0.0126 - val_loss: 0.8643 - val_accuracy: 0.7733 - val_mae: 0.1609 - val_mse: 0.1189 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 936/2000\n",
            "\n",
            "Epoch 936: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0755 - accuracy: 0.9772 - mae: 0.0342 - mse: 0.0130 - val_loss: 0.8563 - val_accuracy: 0.7743 - val_mae: 0.1608 - val_mse: 0.1187 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 937/2000\n",
            "\n",
            "Epoch 937: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0710 - accuracy: 0.9773 - mae: 0.0329 - mse: 0.0123 - val_loss: 0.8598 - val_accuracy: 0.7747 - val_mae: 0.1607 - val_mse: 0.1188 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 938/2000\n",
            "\n",
            "Epoch 938: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0707 - accuracy: 0.9787 - mae: 0.0327 - mse: 0.0121 - val_loss: 0.8633 - val_accuracy: 0.7757 - val_mae: 0.1596 - val_mse: 0.1183 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 939/2000\n",
            "\n",
            "Epoch 939: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0727 - accuracy: 0.9774 - mae: 0.0330 - mse: 0.0125 - val_loss: 0.8701 - val_accuracy: 0.7760 - val_mae: 0.1597 - val_mse: 0.1187 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 940/2000\n",
            "\n",
            "Epoch 940: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0739 - accuracy: 0.9762 - mae: 0.0330 - mse: 0.0129 - val_loss: 0.8627 - val_accuracy: 0.7753 - val_mae: 0.1601 - val_mse: 0.1185 - lr: 1.0000e-04 - 314ms/epoch - 26ms/step\n",
            "Epoch 941/2000\n",
            "\n",
            "Epoch 941: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0756 - accuracy: 0.9753 - mae: 0.0343 - mse: 0.0133 - val_loss: 0.8565 - val_accuracy: 0.7730 - val_mae: 0.1601 - val_mse: 0.1182 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 942/2000\n",
            "\n",
            "Epoch 942: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0709 - accuracy: 0.9772 - mae: 0.0324 - mse: 0.0119 - val_loss: 0.8594 - val_accuracy: 0.7727 - val_mae: 0.1598 - val_mse: 0.1180 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 943/2000\n",
            "\n",
            "Epoch 943: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0677 - accuracy: 0.9783 - mae: 0.0318 - mse: 0.0116 - val_loss: 0.8671 - val_accuracy: 0.7760 - val_mae: 0.1597 - val_mse: 0.1185 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 944/2000\n",
            "\n",
            "Epoch 944: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0726 - accuracy: 0.9770 - mae: 0.0323 - mse: 0.0124 - val_loss: 0.8638 - val_accuracy: 0.7743 - val_mae: 0.1598 - val_mse: 0.1184 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 945/2000\n",
            "\n",
            "Epoch 945: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0719 - accuracy: 0.9750 - mae: 0.0333 - mse: 0.0126 - val_loss: 0.8648 - val_accuracy: 0.7750 - val_mae: 0.1597 - val_mse: 0.1184 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 946/2000\n",
            "\n",
            "Epoch 946: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0714 - accuracy: 0.9763 - mae: 0.0328 - mse: 0.0125 - val_loss: 0.8648 - val_accuracy: 0.7730 - val_mae: 0.1597 - val_mse: 0.1185 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 947/2000\n",
            "\n",
            "Epoch 947: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0702 - accuracy: 0.9776 - mae: 0.0325 - mse: 0.0121 - val_loss: 0.8706 - val_accuracy: 0.7750 - val_mae: 0.1598 - val_mse: 0.1187 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 948/2000\n",
            "\n",
            "Epoch 948: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0715 - accuracy: 0.9768 - mae: 0.0329 - mse: 0.0126 - val_loss: 0.8647 - val_accuracy: 0.7733 - val_mae: 0.1599 - val_mse: 0.1184 - lr: 1.0000e-04 - 322ms/epoch - 27ms/step\n",
            "Epoch 949/2000\n",
            "\n",
            "Epoch 949: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0696 - accuracy: 0.9772 - mae: 0.0325 - mse: 0.0120 - val_loss: 0.8687 - val_accuracy: 0.7743 - val_mae: 0.1604 - val_mse: 0.1189 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 950/2000\n",
            "\n",
            "Epoch 950: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0721 - accuracy: 0.9772 - mae: 0.0328 - mse: 0.0123 - val_loss: 0.8649 - val_accuracy: 0.7730 - val_mae: 0.1599 - val_mse: 0.1184 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 951/2000\n",
            "\n",
            "Epoch 951: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0715 - accuracy: 0.9760 - mae: 0.0323 - mse: 0.0124 - val_loss: 0.8694 - val_accuracy: 0.7743 - val_mae: 0.1596 - val_mse: 0.1186 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 952/2000\n",
            "\n",
            "Epoch 952: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0698 - accuracy: 0.9778 - mae: 0.0322 - mse: 0.0119 - val_loss: 0.8630 - val_accuracy: 0.7740 - val_mae: 0.1596 - val_mse: 0.1183 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 953/2000\n",
            "\n",
            "Epoch 953: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0748 - accuracy: 0.9751 - mae: 0.0338 - mse: 0.0131 - val_loss: 0.8693 - val_accuracy: 0.7763 - val_mae: 0.1597 - val_mse: 0.1188 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 954/2000\n",
            "\n",
            "Epoch 954: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0725 - accuracy: 0.9780 - mae: 0.0334 - mse: 0.0125 - val_loss: 0.8627 - val_accuracy: 0.7737 - val_mae: 0.1597 - val_mse: 0.1184 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 955/2000\n",
            "\n",
            "Epoch 955: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0716 - accuracy: 0.9757 - mae: 0.0329 - mse: 0.0126 - val_loss: 0.8663 - val_accuracy: 0.7767 - val_mae: 0.1598 - val_mse: 0.1184 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 956/2000\n",
            "\n",
            "Epoch 956: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0657 - accuracy: 0.9789 - mae: 0.0310 - mse: 0.0112 - val_loss: 0.8677 - val_accuracy: 0.7743 - val_mae: 0.1597 - val_mse: 0.1185 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 957/2000\n",
            "\n",
            "Epoch 957: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0713 - accuracy: 0.9771 - mae: 0.0323 - mse: 0.0123 - val_loss: 0.8700 - val_accuracy: 0.7743 - val_mae: 0.1594 - val_mse: 0.1185 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 958/2000\n",
            "\n",
            "Epoch 958: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0719 - accuracy: 0.9762 - mae: 0.0327 - mse: 0.0126 - val_loss: 0.8744 - val_accuracy: 0.7740 - val_mae: 0.1596 - val_mse: 0.1189 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 959/2000\n",
            "\n",
            "Epoch 959: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0720 - accuracy: 0.9762 - mae: 0.0326 - mse: 0.0126 - val_loss: 0.8704 - val_accuracy: 0.7737 - val_mae: 0.1596 - val_mse: 0.1187 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 960/2000\n",
            "\n",
            "Epoch 960: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0732 - accuracy: 0.9751 - mae: 0.0335 - mse: 0.0129 - val_loss: 0.8702 - val_accuracy: 0.7743 - val_mae: 0.1600 - val_mse: 0.1187 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 961/2000\n",
            "\n",
            "Epoch 961: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0712 - accuracy: 0.9756 - mae: 0.0329 - mse: 0.0126 - val_loss: 0.8688 - val_accuracy: 0.7723 - val_mae: 0.1603 - val_mse: 0.1188 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 962/2000\n",
            "\n",
            "Epoch 962: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0708 - accuracy: 0.9773 - mae: 0.0320 - mse: 0.0123 - val_loss: 0.8675 - val_accuracy: 0.7730 - val_mae: 0.1598 - val_mse: 0.1184 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 963/2000\n",
            "\n",
            "Epoch 963: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0695 - accuracy: 0.9772 - mae: 0.0315 - mse: 0.0121 - val_loss: 0.8723 - val_accuracy: 0.7743 - val_mae: 0.1599 - val_mse: 0.1189 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 964/2000\n",
            "\n",
            "Epoch 964: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0694 - accuracy: 0.9781 - mae: 0.0316 - mse: 0.0119 - val_loss: 0.8724 - val_accuracy: 0.7740 - val_mae: 0.1598 - val_mse: 0.1188 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 965/2000\n",
            "\n",
            "Epoch 965: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0719 - accuracy: 0.9767 - mae: 0.0325 - mse: 0.0127 - val_loss: 0.8724 - val_accuracy: 0.7767 - val_mae: 0.1599 - val_mse: 0.1189 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 966/2000\n",
            "\n",
            "Epoch 966: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0674 - accuracy: 0.9773 - mae: 0.0310 - mse: 0.0117 - val_loss: 0.8717 - val_accuracy: 0.7750 - val_mae: 0.1599 - val_mse: 0.1188 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 967/2000\n",
            "\n",
            "Epoch 967: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0753 - accuracy: 0.9741 - mae: 0.0334 - mse: 0.0132 - val_loss: 0.8705 - val_accuracy: 0.7757 - val_mae: 0.1598 - val_mse: 0.1186 - lr: 1.0000e-04 - 317ms/epoch - 26ms/step\n",
            "Epoch 968/2000\n",
            "\n",
            "Epoch 968: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0713 - accuracy: 0.9763 - mae: 0.0327 - mse: 0.0125 - val_loss: 0.8680 - val_accuracy: 0.7750 - val_mae: 0.1601 - val_mse: 0.1188 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 969/2000\n",
            "\n",
            "Epoch 969: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0690 - accuracy: 0.9778 - mae: 0.0318 - mse: 0.0119 - val_loss: 0.8668 - val_accuracy: 0.7753 - val_mae: 0.1604 - val_mse: 0.1190 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 970/2000\n",
            "\n",
            "Epoch 970: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0690 - accuracy: 0.9786 - mae: 0.0314 - mse: 0.0119 - val_loss: 0.8691 - val_accuracy: 0.7740 - val_mae: 0.1600 - val_mse: 0.1190 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 971/2000\n",
            "\n",
            "Epoch 971: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0700 - accuracy: 0.9778 - mae: 0.0320 - mse: 0.0120 - val_loss: 0.8704 - val_accuracy: 0.7737 - val_mae: 0.1600 - val_mse: 0.1190 - lr: 1.0000e-04 - 324ms/epoch - 27ms/step\n",
            "Epoch 972/2000\n",
            "\n",
            "Epoch 972: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0698 - accuracy: 0.9758 - mae: 0.0318 - mse: 0.0124 - val_loss: 0.8760 - val_accuracy: 0.7760 - val_mae: 0.1599 - val_mse: 0.1190 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 973/2000\n",
            "\n",
            "Epoch 973: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0708 - accuracy: 0.9772 - mae: 0.0321 - mse: 0.0124 - val_loss: 0.8744 - val_accuracy: 0.7730 - val_mae: 0.1603 - val_mse: 0.1194 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 974/2000\n",
            "\n",
            "Epoch 974: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0672 - accuracy: 0.9788 - mae: 0.0314 - mse: 0.0115 - val_loss: 0.8729 - val_accuracy: 0.7717 - val_mae: 0.1603 - val_mse: 0.1193 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 975/2000\n",
            "\n",
            "Epoch 975: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0740 - accuracy: 0.9760 - mae: 0.0329 - mse: 0.0130 - val_loss: 0.8715 - val_accuracy: 0.7730 - val_mae: 0.1606 - val_mse: 0.1195 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 976/2000\n",
            "\n",
            "Epoch 976: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0722 - accuracy: 0.9752 - mae: 0.0328 - mse: 0.0128 - val_loss: 0.8675 - val_accuracy: 0.7687 - val_mae: 0.1608 - val_mse: 0.1194 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 977/2000\n",
            "\n",
            "Epoch 977: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0715 - accuracy: 0.9766 - mae: 0.0321 - mse: 0.0123 - val_loss: 0.8683 - val_accuracy: 0.7753 - val_mae: 0.1605 - val_mse: 0.1191 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 978/2000\n",
            "\n",
            "Epoch 978: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0713 - accuracy: 0.9768 - mae: 0.0320 - mse: 0.0123 - val_loss: 0.8709 - val_accuracy: 0.7720 - val_mae: 0.1602 - val_mse: 0.1191 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 979/2000\n",
            "\n",
            "Epoch 979: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0692 - accuracy: 0.9767 - mae: 0.0323 - mse: 0.0120 - val_loss: 0.8718 - val_accuracy: 0.7737 - val_mae: 0.1602 - val_mse: 0.1190 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 980/2000\n",
            "\n",
            "Epoch 980: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0680 - accuracy: 0.9787 - mae: 0.0314 - mse: 0.0118 - val_loss: 0.8726 - val_accuracy: 0.7713 - val_mae: 0.1600 - val_mse: 0.1190 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 981/2000\n",
            "\n",
            "Epoch 981: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0698 - accuracy: 0.9769 - mae: 0.0322 - mse: 0.0122 - val_loss: 0.8790 - val_accuracy: 0.7767 - val_mae: 0.1599 - val_mse: 0.1193 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 982/2000\n",
            "\n",
            "Epoch 982: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0710 - accuracy: 0.9772 - mae: 0.0320 - mse: 0.0123 - val_loss: 0.8751 - val_accuracy: 0.7767 - val_mae: 0.1597 - val_mse: 0.1189 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 983/2000\n",
            "\n",
            "Epoch 983: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0691 - accuracy: 0.9773 - mae: 0.0311 - mse: 0.0120 - val_loss: 0.8765 - val_accuracy: 0.7737 - val_mae: 0.1601 - val_mse: 0.1191 - lr: 1.0000e-04 - 315ms/epoch - 26ms/step\n",
            "Epoch 984/2000\n",
            "\n",
            "Epoch 984: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0683 - accuracy: 0.9767 - mae: 0.0312 - mse: 0.0119 - val_loss: 0.8731 - val_accuracy: 0.7723 - val_mae: 0.1598 - val_mse: 0.1188 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 985/2000\n",
            "\n",
            "Epoch 985: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0720 - accuracy: 0.9755 - mae: 0.0329 - mse: 0.0127 - val_loss: 0.8743 - val_accuracy: 0.7723 - val_mae: 0.1602 - val_mse: 0.1191 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 986/2000\n",
            "\n",
            "Epoch 986: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0639 - accuracy: 0.9803 - mae: 0.0304 - mse: 0.0108 - val_loss: 0.8719 - val_accuracy: 0.7723 - val_mae: 0.1600 - val_mse: 0.1190 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 987/2000\n",
            "\n",
            "Epoch 987: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0666 - accuracy: 0.9786 - mae: 0.0305 - mse: 0.0114 - val_loss: 0.8768 - val_accuracy: 0.7753 - val_mae: 0.1598 - val_mse: 0.1190 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 988/2000\n",
            "\n",
            "Epoch 988: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0677 - accuracy: 0.9776 - mae: 0.0310 - mse: 0.0117 - val_loss: 0.8764 - val_accuracy: 0.7743 - val_mae: 0.1596 - val_mse: 0.1189 - lr: 1.0000e-04 - 283ms/epoch - 24ms/step\n",
            "Epoch 989/2000\n",
            "\n",
            "Epoch 989: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0710 - accuracy: 0.9768 - mae: 0.0321 - mse: 0.0125 - val_loss: 0.8798 - val_accuracy: 0.7757 - val_mae: 0.1597 - val_mse: 0.1192 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 990/2000\n",
            "\n",
            "Epoch 990: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0702 - accuracy: 0.9767 - mae: 0.0321 - mse: 0.0123 - val_loss: 0.8772 - val_accuracy: 0.7750 - val_mae: 0.1597 - val_mse: 0.1189 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 991/2000\n",
            "\n",
            "Epoch 991: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0672 - accuracy: 0.9787 - mae: 0.0310 - mse: 0.0116 - val_loss: 0.8767 - val_accuracy: 0.7757 - val_mae: 0.1597 - val_mse: 0.1187 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 992/2000\n",
            "\n",
            "Epoch 992: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0707 - accuracy: 0.9769 - mae: 0.0319 - mse: 0.0122 - val_loss: 0.8779 - val_accuracy: 0.7743 - val_mae: 0.1600 - val_mse: 0.1190 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 993/2000\n",
            "\n",
            "Epoch 993: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0709 - accuracy: 0.9774 - mae: 0.0312 - mse: 0.0121 - val_loss: 0.8726 - val_accuracy: 0.7743 - val_mae: 0.1600 - val_mse: 0.1188 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 994/2000\n",
            "\n",
            "Epoch 994: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0713 - accuracy: 0.9775 - mae: 0.0320 - mse: 0.0124 - val_loss: 0.8759 - val_accuracy: 0.7737 - val_mae: 0.1600 - val_mse: 0.1189 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 995/2000\n",
            "\n",
            "Epoch 995: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0683 - accuracy: 0.9780 - mae: 0.0315 - mse: 0.0117 - val_loss: 0.8687 - val_accuracy: 0.7730 - val_mae: 0.1597 - val_mse: 0.1184 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 996/2000\n",
            "\n",
            "Epoch 996: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0660 - accuracy: 0.9791 - mae: 0.0313 - mse: 0.0112 - val_loss: 0.8810 - val_accuracy: 0.7737 - val_mae: 0.1594 - val_mse: 0.1186 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 997/2000\n",
            "\n",
            "Epoch 997: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0644 - accuracy: 0.9795 - mae: 0.0294 - mse: 0.0109 - val_loss: 0.8852 - val_accuracy: 0.7740 - val_mae: 0.1590 - val_mse: 0.1186 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 998/2000\n",
            "\n",
            "Epoch 998: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0650 - accuracy: 0.9793 - mae: 0.0297 - mse: 0.0111 - val_loss: 0.8846 - val_accuracy: 0.7737 - val_mae: 0.1594 - val_mse: 0.1190 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 999/2000\n",
            "\n",
            "Epoch 999: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0665 - accuracy: 0.9797 - mae: 0.0305 - mse: 0.0113 - val_loss: 0.8869 - val_accuracy: 0.7757 - val_mae: 0.1591 - val_mse: 0.1189 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1000/2000\n",
            "\n",
            "Epoch 1000: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0687 - accuracy: 0.9773 - mae: 0.0311 - mse: 0.0118 - val_loss: 0.8845 - val_accuracy: 0.7747 - val_mae: 0.1592 - val_mse: 0.1188 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1001/2000\n",
            "\n",
            "Epoch 1001: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0651 - accuracy: 0.9798 - mae: 0.0298 - mse: 0.0111 - val_loss: 0.8857 - val_accuracy: 0.7727 - val_mae: 0.1593 - val_mse: 0.1190 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1002/2000\n",
            "\n",
            "Epoch 1002: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0681 - accuracy: 0.9783 - mae: 0.0305 - mse: 0.0117 - val_loss: 0.8849 - val_accuracy: 0.7750 - val_mae: 0.1596 - val_mse: 0.1191 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1003/2000\n",
            "\n",
            "Epoch 1003: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0688 - accuracy: 0.9777 - mae: 0.0315 - mse: 0.0119 - val_loss: 0.8834 - val_accuracy: 0.7757 - val_mae: 0.1594 - val_mse: 0.1190 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1004/2000\n",
            "\n",
            "Epoch 1004: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0706 - accuracy: 0.9753 - mae: 0.0319 - mse: 0.0125 - val_loss: 0.8816 - val_accuracy: 0.7763 - val_mae: 0.1597 - val_mse: 0.1190 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1005/2000\n",
            "\n",
            "Epoch 1005: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0674 - accuracy: 0.9768 - mae: 0.0308 - mse: 0.0118 - val_loss: 0.8841 - val_accuracy: 0.7737 - val_mae: 0.1597 - val_mse: 0.1190 - lr: 1.0000e-04 - 282ms/epoch - 23ms/step\n",
            "Epoch 1006/2000\n",
            "\n",
            "Epoch 1006: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0745 - accuracy: 0.9753 - mae: 0.0324 - mse: 0.0132 - val_loss: 0.8839 - val_accuracy: 0.7723 - val_mae: 0.1598 - val_mse: 0.1190 - lr: 1.0000e-04 - 316ms/epoch - 26ms/step\n",
            "Epoch 1007/2000\n",
            "\n",
            "Epoch 1007: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0658 - accuracy: 0.9794 - mae: 0.0306 - mse: 0.0112 - val_loss: 0.8806 - val_accuracy: 0.7733 - val_mae: 0.1600 - val_mse: 0.1190 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1008/2000\n",
            "\n",
            "Epoch 1008: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0633 - accuracy: 0.9796 - mae: 0.0297 - mse: 0.0109 - val_loss: 0.8813 - val_accuracy: 0.7723 - val_mae: 0.1598 - val_mse: 0.1190 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1009/2000\n",
            "\n",
            "Epoch 1009: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0639 - accuracy: 0.9796 - mae: 0.0298 - mse: 0.0108 - val_loss: 0.8855 - val_accuracy: 0.7740 - val_mae: 0.1593 - val_mse: 0.1190 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1010/2000\n",
            "\n",
            "Epoch 1010: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0658 - accuracy: 0.9784 - mae: 0.0303 - mse: 0.0115 - val_loss: 0.8864 - val_accuracy: 0.7737 - val_mae: 0.1592 - val_mse: 0.1189 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1011/2000\n",
            "\n",
            "Epoch 1011: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0623 - accuracy: 0.9798 - mae: 0.0291 - mse: 0.0107 - val_loss: 0.8905 - val_accuracy: 0.7750 - val_mae: 0.1591 - val_mse: 0.1191 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1012/2000\n",
            "\n",
            "Epoch 1012: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0703 - accuracy: 0.9765 - mae: 0.0306 - mse: 0.0122 - val_loss: 0.8925 - val_accuracy: 0.7740 - val_mae: 0.1592 - val_mse: 0.1193 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1013/2000\n",
            "\n",
            "Epoch 1013: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0640 - accuracy: 0.9803 - mae: 0.0296 - mse: 0.0109 - val_loss: 0.8898 - val_accuracy: 0.7737 - val_mae: 0.1593 - val_mse: 0.1194 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1014/2000\n",
            "\n",
            "Epoch 1014: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0653 - accuracy: 0.9787 - mae: 0.0300 - mse: 0.0113 - val_loss: 0.8872 - val_accuracy: 0.7703 - val_mae: 0.1596 - val_mse: 0.1194 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1015/2000\n",
            "\n",
            "Epoch 1015: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0681 - accuracy: 0.9771 - mae: 0.0308 - mse: 0.0119 - val_loss: 0.8875 - val_accuracy: 0.7747 - val_mae: 0.1601 - val_mse: 0.1195 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1016/2000\n",
            "\n",
            "Epoch 1016: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0638 - accuracy: 0.9792 - mae: 0.0299 - mse: 0.0110 - val_loss: 0.8835 - val_accuracy: 0.7747 - val_mae: 0.1597 - val_mse: 0.1191 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 1017/2000\n",
            "\n",
            "Epoch 1017: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0655 - accuracy: 0.9791 - mae: 0.0302 - mse: 0.0112 - val_loss: 0.8873 - val_accuracy: 0.7740 - val_mae: 0.1594 - val_mse: 0.1192 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1018/2000\n",
            "\n",
            "Epoch 1018: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0614 - accuracy: 0.9796 - mae: 0.0287 - mse: 0.0105 - val_loss: 0.8900 - val_accuracy: 0.7747 - val_mae: 0.1588 - val_mse: 0.1190 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1019/2000\n",
            "\n",
            "Epoch 1019: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0605 - accuracy: 0.9802 - mae: 0.0285 - mse: 0.0105 - val_loss: 0.8970 - val_accuracy: 0.7753 - val_mae: 0.1584 - val_mse: 0.1190 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1020/2000\n",
            "\n",
            "Epoch 1020: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0665 - accuracy: 0.9778 - mae: 0.0294 - mse: 0.0117 - val_loss: 0.8932 - val_accuracy: 0.7743 - val_mae: 0.1591 - val_mse: 0.1193 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1021/2000\n",
            "\n",
            "Epoch 1021: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0697 - accuracy: 0.9787 - mae: 0.0307 - mse: 0.0119 - val_loss: 0.8882 - val_accuracy: 0.7763 - val_mae: 0.1598 - val_mse: 0.1195 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1022/2000\n",
            "\n",
            "Epoch 1022: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0652 - accuracy: 0.9792 - mae: 0.0298 - mse: 0.0113 - val_loss: 0.8880 - val_accuracy: 0.7730 - val_mae: 0.1599 - val_mse: 0.1195 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1023/2000\n",
            "\n",
            "Epoch 1023: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0612 - accuracy: 0.9805 - mae: 0.0290 - mse: 0.0104 - val_loss: 0.8898 - val_accuracy: 0.7737 - val_mae: 0.1598 - val_mse: 0.1195 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1024/2000\n",
            "\n",
            "Epoch 1024: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0675 - accuracy: 0.9778 - mae: 0.0308 - mse: 0.0118 - val_loss: 0.8919 - val_accuracy: 0.7740 - val_mae: 0.1597 - val_mse: 0.1197 - lr: 1.0000e-04 - 306ms/epoch - 25ms/step\n",
            "Epoch 1025/2000\n",
            "\n",
            "Epoch 1025: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0669 - accuracy: 0.9790 - mae: 0.0304 - mse: 0.0114 - val_loss: 0.8850 - val_accuracy: 0.7730 - val_mae: 0.1598 - val_mse: 0.1196 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1026/2000\n",
            "\n",
            "Epoch 1026: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0666 - accuracy: 0.9788 - mae: 0.0299 - mse: 0.0115 - val_loss: 0.8871 - val_accuracy: 0.7743 - val_mae: 0.1596 - val_mse: 0.1195 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1027/2000\n",
            "\n",
            "Epoch 1027: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0644 - accuracy: 0.9787 - mae: 0.0294 - mse: 0.0111 - val_loss: 0.8820 - val_accuracy: 0.7723 - val_mae: 0.1598 - val_mse: 0.1193 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1028/2000\n",
            "\n",
            "Epoch 1028: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0622 - accuracy: 0.9808 - mae: 0.0288 - mse: 0.0106 - val_loss: 0.8859 - val_accuracy: 0.7710 - val_mae: 0.1600 - val_mse: 0.1197 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1029/2000\n",
            "\n",
            "Epoch 1029: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0701 - accuracy: 0.9762 - mae: 0.0312 - mse: 0.0122 - val_loss: 0.8908 - val_accuracy: 0.7727 - val_mae: 0.1598 - val_mse: 0.1197 - lr: 1.0000e-04 - 321ms/epoch - 27ms/step\n",
            "Epoch 1030/2000\n",
            "\n",
            "Epoch 1030: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0672 - accuracy: 0.9789 - mae: 0.0306 - mse: 0.0116 - val_loss: 0.8848 - val_accuracy: 0.7733 - val_mae: 0.1594 - val_mse: 0.1193 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 1031/2000\n",
            "\n",
            "Epoch 1031: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0653 - accuracy: 0.9777 - mae: 0.0302 - mse: 0.0114 - val_loss: 0.8876 - val_accuracy: 0.7733 - val_mae: 0.1593 - val_mse: 0.1193 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1032/2000\n",
            "\n",
            "Epoch 1032: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0659 - accuracy: 0.9791 - mae: 0.0297 - mse: 0.0113 - val_loss: 0.8860 - val_accuracy: 0.7700 - val_mae: 0.1596 - val_mse: 0.1194 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1033/2000\n",
            "\n",
            "Epoch 1033: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0672 - accuracy: 0.9778 - mae: 0.0305 - mse: 0.0117 - val_loss: 0.8824 - val_accuracy: 0.7733 - val_mae: 0.1595 - val_mse: 0.1192 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1034/2000\n",
            "\n",
            "Epoch 1034: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0631 - accuracy: 0.9802 - mae: 0.0291 - mse: 0.0108 - val_loss: 0.8876 - val_accuracy: 0.7730 - val_mae: 0.1593 - val_mse: 0.1193 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1035/2000\n",
            "\n",
            "Epoch 1035: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0637 - accuracy: 0.9803 - mae: 0.0295 - mse: 0.0108 - val_loss: 0.8884 - val_accuracy: 0.7713 - val_mae: 0.1596 - val_mse: 0.1195 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1036/2000\n",
            "\n",
            "Epoch 1036: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0648 - accuracy: 0.9787 - mae: 0.0297 - mse: 0.0114 - val_loss: 0.8894 - val_accuracy: 0.7723 - val_mae: 0.1598 - val_mse: 0.1196 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1037/2000\n",
            "\n",
            "Epoch 1037: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0674 - accuracy: 0.9778 - mae: 0.0305 - mse: 0.0117 - val_loss: 0.8902 - val_accuracy: 0.7733 - val_mae: 0.1599 - val_mse: 0.1198 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 1038/2000\n",
            "\n",
            "Epoch 1038: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0654 - accuracy: 0.9769 - mae: 0.0303 - mse: 0.0115 - val_loss: 0.8924 - val_accuracy: 0.7720 - val_mae: 0.1597 - val_mse: 0.1198 - lr: 1.0000e-04 - 317ms/epoch - 26ms/step\n",
            "Epoch 1039/2000\n",
            "\n",
            "Epoch 1039: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0596 - accuracy: 0.9815 - mae: 0.0285 - mse: 0.0101 - val_loss: 0.8911 - val_accuracy: 0.7720 - val_mae: 0.1594 - val_mse: 0.1195 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1040/2000\n",
            "\n",
            "Epoch 1040: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0634 - accuracy: 0.9793 - mae: 0.0294 - mse: 0.0109 - val_loss: 0.8975 - val_accuracy: 0.7743 - val_mae: 0.1594 - val_mse: 0.1199 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1041/2000\n",
            "\n",
            "Epoch 1041: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0670 - accuracy: 0.9783 - mae: 0.0298 - mse: 0.0116 - val_loss: 0.8957 - val_accuracy: 0.7723 - val_mae: 0.1597 - val_mse: 0.1200 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1042/2000\n",
            "\n",
            "Epoch 1042: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0635 - accuracy: 0.9802 - mae: 0.0290 - mse: 0.0107 - val_loss: 0.8906 - val_accuracy: 0.7700 - val_mae: 0.1599 - val_mse: 0.1199 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1043/2000\n",
            "\n",
            "Epoch 1043: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0654 - accuracy: 0.9770 - mae: 0.0298 - mse: 0.0116 - val_loss: 0.8990 - val_accuracy: 0.7717 - val_mae: 0.1598 - val_mse: 0.1202 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1044/2000\n",
            "\n",
            "Epoch 1044: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0646 - accuracy: 0.9793 - mae: 0.0292 - mse: 0.0114 - val_loss: 0.8944 - val_accuracy: 0.7693 - val_mae: 0.1595 - val_mse: 0.1198 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1045/2000\n",
            "\n",
            "Epoch 1045: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0636 - accuracy: 0.9794 - mae: 0.0291 - mse: 0.0110 - val_loss: 0.8947 - val_accuracy: 0.7713 - val_mae: 0.1596 - val_mse: 0.1199 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1046/2000\n",
            "\n",
            "Epoch 1046: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0650 - accuracy: 0.9786 - mae: 0.0300 - mse: 0.0114 - val_loss: 0.8925 - val_accuracy: 0.7730 - val_mae: 0.1594 - val_mse: 0.1196 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1047/2000\n",
            "\n",
            "Epoch 1047: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0652 - accuracy: 0.9798 - mae: 0.0292 - mse: 0.0112 - val_loss: 0.8932 - val_accuracy: 0.7727 - val_mae: 0.1593 - val_mse: 0.1195 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1048/2000\n",
            "\n",
            "Epoch 1048: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0607 - accuracy: 0.9808 - mae: 0.0282 - mse: 0.0103 - val_loss: 0.8946 - val_accuracy: 0.7710 - val_mae: 0.1593 - val_mse: 0.1196 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1049/2000\n",
            "\n",
            "Epoch 1049: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0617 - accuracy: 0.9800 - mae: 0.0285 - mse: 0.0106 - val_loss: 0.9049 - val_accuracy: 0.7710 - val_mae: 0.1594 - val_mse: 0.1202 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1050/2000\n",
            "\n",
            "Epoch 1050: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0656 - accuracy: 0.9777 - mae: 0.0293 - mse: 0.0114 - val_loss: 0.8952 - val_accuracy: 0.7723 - val_mae: 0.1592 - val_mse: 0.1197 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1051/2000\n",
            "\n",
            "Epoch 1051: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0628 - accuracy: 0.9787 - mae: 0.0290 - mse: 0.0108 - val_loss: 0.8978 - val_accuracy: 0.7727 - val_mae: 0.1591 - val_mse: 0.1197 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1052/2000\n",
            "\n",
            "Epoch 1052: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0635 - accuracy: 0.9783 - mae: 0.0292 - mse: 0.0111 - val_loss: 0.8981 - val_accuracy: 0.7740 - val_mae: 0.1586 - val_mse: 0.1194 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1053/2000\n",
            "\n",
            "Epoch 1053: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0619 - accuracy: 0.9798 - mae: 0.0287 - mse: 0.0107 - val_loss: 0.9000 - val_accuracy: 0.7743 - val_mae: 0.1588 - val_mse: 0.1195 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1054/2000\n",
            "\n",
            "Epoch 1054: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0627 - accuracy: 0.9805 - mae: 0.0282 - mse: 0.0107 - val_loss: 0.8985 - val_accuracy: 0.7730 - val_mae: 0.1588 - val_mse: 0.1194 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1055/2000\n",
            "\n",
            "Epoch 1055: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0628 - accuracy: 0.9798 - mae: 0.0285 - mse: 0.0109 - val_loss: 0.8996 - val_accuracy: 0.7710 - val_mae: 0.1586 - val_mse: 0.1193 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1056/2000\n",
            "\n",
            "Epoch 1056: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0625 - accuracy: 0.9790 - mae: 0.0284 - mse: 0.0110 - val_loss: 0.9017 - val_accuracy: 0.7713 - val_mae: 0.1585 - val_mse: 0.1194 - lr: 1.0000e-04 - 320ms/epoch - 27ms/step\n",
            "Epoch 1057/2000\n",
            "\n",
            "Epoch 1057: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0572 - accuracy: 0.9822 - mae: 0.0267 - mse: 0.0096 - val_loss: 0.9013 - val_accuracy: 0.7730 - val_mae: 0.1586 - val_mse: 0.1196 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1058/2000\n",
            "\n",
            "Epoch 1058: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0621 - accuracy: 0.9804 - mae: 0.0284 - mse: 0.0108 - val_loss: 0.8987 - val_accuracy: 0.7727 - val_mae: 0.1590 - val_mse: 0.1198 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1059/2000\n",
            "\n",
            "Epoch 1059: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0625 - accuracy: 0.9797 - mae: 0.0287 - mse: 0.0108 - val_loss: 0.8988 - val_accuracy: 0.7740 - val_mae: 0.1590 - val_mse: 0.1197 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1060/2000\n",
            "\n",
            "Epoch 1060: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0615 - accuracy: 0.9791 - mae: 0.0284 - mse: 0.0106 - val_loss: 0.8954 - val_accuracy: 0.7717 - val_mae: 0.1588 - val_mse: 0.1194 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1061/2000\n",
            "\n",
            "Epoch 1061: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0599 - accuracy: 0.9802 - mae: 0.0276 - mse: 0.0104 - val_loss: 0.9030 - val_accuracy: 0.7713 - val_mae: 0.1584 - val_mse: 0.1196 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1062/2000\n",
            "\n",
            "Epoch 1062: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0603 - accuracy: 0.9808 - mae: 0.0283 - mse: 0.0104 - val_loss: 0.8984 - val_accuracy: 0.7707 - val_mae: 0.1585 - val_mse: 0.1194 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1063/2000\n",
            "\n",
            "Epoch 1063: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0645 - accuracy: 0.9783 - mae: 0.0287 - mse: 0.0112 - val_loss: 0.8965 - val_accuracy: 0.7733 - val_mae: 0.1589 - val_mse: 0.1194 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1064/2000\n",
            "\n",
            "Epoch 1064: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0582 - accuracy: 0.9815 - mae: 0.0275 - mse: 0.0099 - val_loss: 0.8994 - val_accuracy: 0.7750 - val_mae: 0.1588 - val_mse: 0.1194 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1065/2000\n",
            "\n",
            "Epoch 1065: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0644 - accuracy: 0.9794 - mae: 0.0286 - mse: 0.0110 - val_loss: 0.9021 - val_accuracy: 0.7707 - val_mae: 0.1592 - val_mse: 0.1199 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1066/2000\n",
            "\n",
            "Epoch 1066: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0622 - accuracy: 0.9794 - mae: 0.0282 - mse: 0.0106 - val_loss: 0.8955 - val_accuracy: 0.7733 - val_mae: 0.1589 - val_mse: 0.1194 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1067/2000\n",
            "\n",
            "Epoch 1067: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0615 - accuracy: 0.9799 - mae: 0.0284 - mse: 0.0108 - val_loss: 0.9073 - val_accuracy: 0.7743 - val_mae: 0.1591 - val_mse: 0.1201 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1068/2000\n",
            "\n",
            "Epoch 1068: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0606 - accuracy: 0.9813 - mae: 0.0279 - mse: 0.0104 - val_loss: 0.8932 - val_accuracy: 0.7733 - val_mae: 0.1590 - val_mse: 0.1193 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1069/2000\n",
            "\n",
            "Epoch 1069: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0641 - accuracy: 0.9787 - mae: 0.0291 - mse: 0.0111 - val_loss: 0.8975 - val_accuracy: 0.7723 - val_mae: 0.1595 - val_mse: 0.1197 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1070/2000\n",
            "\n",
            "Epoch 1070: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0608 - accuracy: 0.9813 - mae: 0.0281 - mse: 0.0103 - val_loss: 0.8915 - val_accuracy: 0.7740 - val_mae: 0.1591 - val_mse: 0.1192 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1071/2000\n",
            "\n",
            "Epoch 1071: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0647 - accuracy: 0.9793 - mae: 0.0287 - mse: 0.0111 - val_loss: 0.8997 - val_accuracy: 0.7750 - val_mae: 0.1593 - val_mse: 0.1196 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1072/2000\n",
            "\n",
            "Epoch 1072: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0605 - accuracy: 0.9799 - mae: 0.0278 - mse: 0.0103 - val_loss: 0.8986 - val_accuracy: 0.7757 - val_mae: 0.1593 - val_mse: 0.1195 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1073/2000\n",
            "\n",
            "Epoch 1073: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0623 - accuracy: 0.9792 - mae: 0.0286 - mse: 0.0110 - val_loss: 0.9015 - val_accuracy: 0.7737 - val_mae: 0.1591 - val_mse: 0.1196 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 1074/2000\n",
            "\n",
            "Epoch 1074: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0612 - accuracy: 0.9816 - mae: 0.0281 - mse: 0.0104 - val_loss: 0.9055 - val_accuracy: 0.7727 - val_mae: 0.1589 - val_mse: 0.1197 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1075/2000\n",
            "\n",
            "Epoch 1075: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0635 - accuracy: 0.9785 - mae: 0.0285 - mse: 0.0112 - val_loss: 0.9022 - val_accuracy: 0.7740 - val_mae: 0.1588 - val_mse: 0.1194 - lr: 1.0000e-04 - 281ms/epoch - 23ms/step\n",
            "Epoch 1076/2000\n",
            "\n",
            "Epoch 1076: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0618 - accuracy: 0.9812 - mae: 0.0276 - mse: 0.0103 - val_loss: 0.9006 - val_accuracy: 0.7760 - val_mae: 0.1589 - val_mse: 0.1195 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1077/2000\n",
            "\n",
            "Epoch 1077: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0641 - accuracy: 0.9802 - mae: 0.0285 - mse: 0.0109 - val_loss: 0.9000 - val_accuracy: 0.7750 - val_mae: 0.1591 - val_mse: 0.1194 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1078/2000\n",
            "\n",
            "Epoch 1078: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0584 - accuracy: 0.9810 - mae: 0.0269 - mse: 0.0100 - val_loss: 0.9030 - val_accuracy: 0.7737 - val_mae: 0.1594 - val_mse: 0.1199 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1079/2000\n",
            "\n",
            "Epoch 1079: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0595 - accuracy: 0.9797 - mae: 0.0276 - mse: 0.0103 - val_loss: 0.9069 - val_accuracy: 0.7717 - val_mae: 0.1596 - val_mse: 0.1203 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1080/2000\n",
            "\n",
            "Epoch 1080: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0600 - accuracy: 0.9812 - mae: 0.0272 - mse: 0.0103 - val_loss: 0.9023 - val_accuracy: 0.7713 - val_mae: 0.1592 - val_mse: 0.1197 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1081/2000\n",
            "\n",
            "Epoch 1081: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0627 - accuracy: 0.9803 - mae: 0.0285 - mse: 0.0108 - val_loss: 0.9066 - val_accuracy: 0.7707 - val_mae: 0.1592 - val_mse: 0.1200 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1082/2000\n",
            "\n",
            "Epoch 1082: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0642 - accuracy: 0.9801 - mae: 0.0286 - mse: 0.0109 - val_loss: 0.9119 - val_accuracy: 0.7717 - val_mae: 0.1593 - val_mse: 0.1204 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1083/2000\n",
            "\n",
            "Epoch 1083: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0622 - accuracy: 0.9797 - mae: 0.0277 - mse: 0.0106 - val_loss: 0.9072 - val_accuracy: 0.7723 - val_mae: 0.1589 - val_mse: 0.1199 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1084/2000\n",
            "\n",
            "Epoch 1084: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0589 - accuracy: 0.9822 - mae: 0.0271 - mse: 0.0099 - val_loss: 0.9059 - val_accuracy: 0.7737 - val_mae: 0.1588 - val_mse: 0.1197 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1085/2000\n",
            "\n",
            "Epoch 1085: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0576 - accuracy: 0.9825 - mae: 0.0265 - mse: 0.0097 - val_loss: 0.9077 - val_accuracy: 0.7750 - val_mae: 0.1589 - val_mse: 0.1199 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1086/2000\n",
            "\n",
            "Epoch 1086: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0640 - accuracy: 0.9782 - mae: 0.0288 - mse: 0.0113 - val_loss: 0.9050 - val_accuracy: 0.7750 - val_mae: 0.1589 - val_mse: 0.1197 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1087/2000\n",
            "\n",
            "Epoch 1087: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0619 - accuracy: 0.9797 - mae: 0.0282 - mse: 0.0107 - val_loss: 0.9066 - val_accuracy: 0.7727 - val_mae: 0.1590 - val_mse: 0.1198 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1088/2000\n",
            "\n",
            "Epoch 1088: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0614 - accuracy: 0.9797 - mae: 0.0273 - mse: 0.0106 - val_loss: 0.9078 - val_accuracy: 0.7713 - val_mae: 0.1589 - val_mse: 0.1198 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1089/2000\n",
            "\n",
            "Epoch 1089: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0614 - accuracy: 0.9805 - mae: 0.0279 - mse: 0.0106 - val_loss: 0.9070 - val_accuracy: 0.7720 - val_mae: 0.1588 - val_mse: 0.1197 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1090/2000\n",
            "\n",
            "Epoch 1090: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0611 - accuracy: 0.9791 - mae: 0.0285 - mse: 0.0107 - val_loss: 0.9006 - val_accuracy: 0.7717 - val_mae: 0.1591 - val_mse: 0.1197 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1091/2000\n",
            "\n",
            "Epoch 1091: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0593 - accuracy: 0.9806 - mae: 0.0275 - mse: 0.0102 - val_loss: 0.9020 - val_accuracy: 0.7713 - val_mae: 0.1590 - val_mse: 0.1198 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1092/2000\n",
            "\n",
            "Epoch 1092: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0567 - accuracy: 0.9827 - mae: 0.0269 - mse: 0.0096 - val_loss: 0.9112 - val_accuracy: 0.7750 - val_mae: 0.1584 - val_mse: 0.1197 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1093/2000\n",
            "\n",
            "Epoch 1093: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0650 - accuracy: 0.9772 - mae: 0.0289 - mse: 0.0116 - val_loss: 0.9060 - val_accuracy: 0.7740 - val_mae: 0.1586 - val_mse: 0.1197 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1094/2000\n",
            "\n",
            "Epoch 1094: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0571 - accuracy: 0.9818 - mae: 0.0267 - mse: 0.0098 - val_loss: 0.9068 - val_accuracy: 0.7717 - val_mae: 0.1588 - val_mse: 0.1199 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1095/2000\n",
            "\n",
            "Epoch 1095: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0607 - accuracy: 0.9795 - mae: 0.0270 - mse: 0.0104 - val_loss: 0.9060 - val_accuracy: 0.7750 - val_mae: 0.1588 - val_mse: 0.1198 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1096/2000\n",
            "\n",
            "Epoch 1096: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0619 - accuracy: 0.9805 - mae: 0.0278 - mse: 0.0108 - val_loss: 0.9033 - val_accuracy: 0.7743 - val_mae: 0.1584 - val_mse: 0.1193 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1097/2000\n",
            "\n",
            "Epoch 1097: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0632 - accuracy: 0.9787 - mae: 0.0288 - mse: 0.0111 - val_loss: 0.9050 - val_accuracy: 0.7760 - val_mae: 0.1585 - val_mse: 0.1194 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1098/2000\n",
            "\n",
            "Epoch 1098: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0593 - accuracy: 0.9811 - mae: 0.0275 - mse: 0.0103 - val_loss: 0.9028 - val_accuracy: 0.7720 - val_mae: 0.1586 - val_mse: 0.1194 - lr: 1.0000e-04 - 283ms/epoch - 24ms/step\n",
            "Epoch 1099/2000\n",
            "\n",
            "Epoch 1099: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0599 - accuracy: 0.9795 - mae: 0.0276 - mse: 0.0104 - val_loss: 0.9097 - val_accuracy: 0.7723 - val_mae: 0.1584 - val_mse: 0.1197 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1100/2000\n",
            "\n",
            "Epoch 1100: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0604 - accuracy: 0.9791 - mae: 0.0273 - mse: 0.0105 - val_loss: 0.9112 - val_accuracy: 0.7717 - val_mae: 0.1586 - val_mse: 0.1198 - lr: 1.0000e-04 - 283ms/epoch - 24ms/step\n",
            "Epoch 1101/2000\n",
            "\n",
            "Epoch 1101: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0611 - accuracy: 0.9804 - mae: 0.0275 - mse: 0.0106 - val_loss: 0.9088 - val_accuracy: 0.7723 - val_mae: 0.1589 - val_mse: 0.1200 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1102/2000\n",
            "\n",
            "Epoch 1102: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0569 - accuracy: 0.9814 - mae: 0.0270 - mse: 0.0098 - val_loss: 0.9045 - val_accuracy: 0.7740 - val_mae: 0.1589 - val_mse: 0.1198 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1103/2000\n",
            "\n",
            "Epoch 1103: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0616 - accuracy: 0.9790 - mae: 0.0283 - mse: 0.0107 - val_loss: 0.9039 - val_accuracy: 0.7730 - val_mae: 0.1591 - val_mse: 0.1199 - lr: 1.0000e-04 - 282ms/epoch - 24ms/step\n",
            "Epoch 1104/2000\n",
            "\n",
            "Epoch 1104: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0594 - accuracy: 0.9802 - mae: 0.0278 - mse: 0.0102 - val_loss: 0.9037 - val_accuracy: 0.7717 - val_mae: 0.1590 - val_mse: 0.1198 - lr: 1.0000e-04 - 283ms/epoch - 24ms/step\n",
            "Epoch 1105/2000\n",
            "\n",
            "Epoch 1105: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0595 - accuracy: 0.9799 - mae: 0.0274 - mse: 0.0103 - val_loss: 0.9034 - val_accuracy: 0.7707 - val_mae: 0.1589 - val_mse: 0.1197 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1106/2000\n",
            "\n",
            "Epoch 1106: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0584 - accuracy: 0.9814 - mae: 0.0269 - mse: 0.0101 - val_loss: 0.9132 - val_accuracy: 0.7713 - val_mae: 0.1591 - val_mse: 0.1203 - lr: 1.0000e-04 - 282ms/epoch - 24ms/step\n",
            "Epoch 1107/2000\n",
            "\n",
            "Epoch 1107: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0591 - accuracy: 0.9807 - mae: 0.0274 - mse: 0.0102 - val_loss: 0.9130 - val_accuracy: 0.7727 - val_mae: 0.1589 - val_mse: 0.1200 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1108/2000\n",
            "\n",
            "Epoch 1108: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0582 - accuracy: 0.9821 - mae: 0.0268 - mse: 0.0101 - val_loss: 0.9176 - val_accuracy: 0.7737 - val_mae: 0.1584 - val_mse: 0.1199 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1109/2000\n",
            "\n",
            "Epoch 1109: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0598 - accuracy: 0.9799 - mae: 0.0272 - mse: 0.0104 - val_loss: 0.9090 - val_accuracy: 0.7733 - val_mae: 0.1582 - val_mse: 0.1195 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1110/2000\n",
            "\n",
            "Epoch 1110: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0563 - accuracy: 0.9819 - mae: 0.0259 - mse: 0.0096 - val_loss: 0.9140 - val_accuracy: 0.7743 - val_mae: 0.1579 - val_mse: 0.1196 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1111/2000\n",
            "\n",
            "Epoch 1111: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0560 - accuracy: 0.9821 - mae: 0.0253 - mse: 0.0096 - val_loss: 0.9150 - val_accuracy: 0.7750 - val_mae: 0.1579 - val_mse: 0.1196 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1112/2000\n",
            "\n",
            "Epoch 1112: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0604 - accuracy: 0.9803 - mae: 0.0270 - mse: 0.0104 - val_loss: 0.9125 - val_accuracy: 0.7723 - val_mae: 0.1582 - val_mse: 0.1196 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1113/2000\n",
            "\n",
            "Epoch 1113: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0629 - accuracy: 0.9769 - mae: 0.0281 - mse: 0.0112 - val_loss: 0.9101 - val_accuracy: 0.7727 - val_mae: 0.1587 - val_mse: 0.1196 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1114/2000\n",
            "\n",
            "Epoch 1114: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0615 - accuracy: 0.9801 - mae: 0.0274 - mse: 0.0104 - val_loss: 0.9064 - val_accuracy: 0.7740 - val_mae: 0.1588 - val_mse: 0.1195 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1115/2000\n",
            "\n",
            "Epoch 1115: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0585 - accuracy: 0.9821 - mae: 0.0272 - mse: 0.0100 - val_loss: 0.9119 - val_accuracy: 0.7723 - val_mae: 0.1587 - val_mse: 0.1197 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1116/2000\n",
            "\n",
            "Epoch 1116: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0568 - accuracy: 0.9818 - mae: 0.0263 - mse: 0.0097 - val_loss: 0.9141 - val_accuracy: 0.7750 - val_mae: 0.1583 - val_mse: 0.1196 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1117/2000\n",
            "\n",
            "Epoch 1117: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0600 - accuracy: 0.9801 - mae: 0.0273 - mse: 0.0105 - val_loss: 0.9129 - val_accuracy: 0.7750 - val_mae: 0.1586 - val_mse: 0.1198 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1118/2000\n",
            "\n",
            "Epoch 1118: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0583 - accuracy: 0.9819 - mae: 0.0267 - mse: 0.0098 - val_loss: 0.9125 - val_accuracy: 0.7743 - val_mae: 0.1587 - val_mse: 0.1198 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1119/2000\n",
            "\n",
            "Epoch 1119: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0590 - accuracy: 0.9820 - mae: 0.0268 - mse: 0.0101 - val_loss: 0.9142 - val_accuracy: 0.7753 - val_mae: 0.1584 - val_mse: 0.1199 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1120/2000\n",
            "\n",
            "Epoch 1120: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0597 - accuracy: 0.9808 - mae: 0.0266 - mse: 0.0103 - val_loss: 0.9192 - val_accuracy: 0.7773 - val_mae: 0.1583 - val_mse: 0.1199 - lr: 1.0000e-04 - 321ms/epoch - 27ms/step\n",
            "Epoch 1121/2000\n",
            "\n",
            "Epoch 1121: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0592 - accuracy: 0.9814 - mae: 0.0263 - mse: 0.0099 - val_loss: 0.9190 - val_accuracy: 0.7750 - val_mae: 0.1588 - val_mse: 0.1200 - lr: 1.0000e-04 - 318ms/epoch - 27ms/step\n",
            "Epoch 1122/2000\n",
            "\n",
            "Epoch 1122: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0592 - accuracy: 0.9808 - mae: 0.0277 - mse: 0.0103 - val_loss: 0.9097 - val_accuracy: 0.7737 - val_mae: 0.1590 - val_mse: 0.1200 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1123/2000\n",
            "\n",
            "Epoch 1123: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0581 - accuracy: 0.9794 - mae: 0.0273 - mse: 0.0103 - val_loss: 0.9199 - val_accuracy: 0.7743 - val_mae: 0.1588 - val_mse: 0.1203 - lr: 1.0000e-04 - 311ms/epoch - 26ms/step\n",
            "Epoch 1124/2000\n",
            "\n",
            "Epoch 1124: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0578 - accuracy: 0.9820 - mae: 0.0266 - mse: 0.0099 - val_loss: 0.9193 - val_accuracy: 0.7750 - val_mae: 0.1586 - val_mse: 0.1201 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1125/2000\n",
            "\n",
            "Epoch 1125: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0594 - accuracy: 0.9810 - mae: 0.0264 - mse: 0.0101 - val_loss: 0.9251 - val_accuracy: 0.7737 - val_mae: 0.1585 - val_mse: 0.1202 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 1126/2000\n",
            "\n",
            "Epoch 1126: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0575 - accuracy: 0.9818 - mae: 0.0268 - mse: 0.0099 - val_loss: 0.9143 - val_accuracy: 0.7747 - val_mae: 0.1584 - val_mse: 0.1197 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1127/2000\n",
            "\n",
            "Epoch 1127: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0594 - accuracy: 0.9796 - mae: 0.0270 - mse: 0.0104 - val_loss: 0.9206 - val_accuracy: 0.7740 - val_mae: 0.1586 - val_mse: 0.1201 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1128/2000\n",
            "\n",
            "Epoch 1128: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0555 - accuracy: 0.9836 - mae: 0.0253 - mse: 0.0094 - val_loss: 0.9215 - val_accuracy: 0.7723 - val_mae: 0.1583 - val_mse: 0.1200 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1129/2000\n",
            "\n",
            "Epoch 1129: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0545 - accuracy: 0.9827 - mae: 0.0254 - mse: 0.0091 - val_loss: 0.9224 - val_accuracy: 0.7743 - val_mae: 0.1584 - val_mse: 0.1201 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1130/2000\n",
            "\n",
            "Epoch 1130: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0575 - accuracy: 0.9799 - mae: 0.0261 - mse: 0.0100 - val_loss: 0.9206 - val_accuracy: 0.7760 - val_mae: 0.1587 - val_mse: 0.1202 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1131/2000\n",
            "\n",
            "Epoch 1131: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0551 - accuracy: 0.9827 - mae: 0.0256 - mse: 0.0093 - val_loss: 0.9214 - val_accuracy: 0.7747 - val_mae: 0.1587 - val_mse: 0.1203 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 1132/2000\n",
            "\n",
            "Epoch 1132: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0568 - accuracy: 0.9822 - mae: 0.0259 - mse: 0.0097 - val_loss: 0.9223 - val_accuracy: 0.7770 - val_mae: 0.1585 - val_mse: 0.1201 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1133/2000\n",
            "\n",
            "Epoch 1133: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0594 - accuracy: 0.9809 - mae: 0.0265 - mse: 0.0101 - val_loss: 0.9242 - val_accuracy: 0.7767 - val_mae: 0.1581 - val_mse: 0.1199 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1134/2000\n",
            "\n",
            "Epoch 1134: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0594 - accuracy: 0.9807 - mae: 0.0265 - mse: 0.0103 - val_loss: 0.9244 - val_accuracy: 0.7750 - val_mae: 0.1585 - val_mse: 0.1202 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1135/2000\n",
            "\n",
            "Epoch 1135: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0555 - accuracy: 0.9818 - mae: 0.0258 - mse: 0.0094 - val_loss: 0.9229 - val_accuracy: 0.7757 - val_mae: 0.1585 - val_mse: 0.1202 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1136/2000\n",
            "\n",
            "Epoch 1136: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0584 - accuracy: 0.9810 - mae: 0.0262 - mse: 0.0101 - val_loss: 0.9213 - val_accuracy: 0.7760 - val_mae: 0.1583 - val_mse: 0.1201 - lr: 1.0000e-04 - 282ms/epoch - 24ms/step\n",
            "Epoch 1137/2000\n",
            "\n",
            "Epoch 1137: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0539 - accuracy: 0.9844 - mae: 0.0252 - mse: 0.0091 - val_loss: 0.9243 - val_accuracy: 0.7753 - val_mae: 0.1584 - val_mse: 0.1202 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1138/2000\n",
            "\n",
            "Epoch 1138: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0553 - accuracy: 0.9836 - mae: 0.0259 - mse: 0.0095 - val_loss: 0.9276 - val_accuracy: 0.7757 - val_mae: 0.1584 - val_mse: 0.1203 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1139/2000\n",
            "\n",
            "Epoch 1139: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0539 - accuracy: 0.9826 - mae: 0.0250 - mse: 0.0093 - val_loss: 0.9256 - val_accuracy: 0.7757 - val_mae: 0.1585 - val_mse: 0.1205 - lr: 1.0000e-04 - 283ms/epoch - 24ms/step\n",
            "Epoch 1140/2000\n",
            "\n",
            "Epoch 1140: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0583 - accuracy: 0.9808 - mae: 0.0261 - mse: 0.0101 - val_loss: 0.9273 - val_accuracy: 0.7747 - val_mae: 0.1587 - val_mse: 0.1207 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1141/2000\n",
            "\n",
            "Epoch 1141: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0567 - accuracy: 0.9822 - mae: 0.0257 - mse: 0.0098 - val_loss: 0.9241 - val_accuracy: 0.7740 - val_mae: 0.1586 - val_mse: 0.1203 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1142/2000\n",
            "\n",
            "Epoch 1142: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0574 - accuracy: 0.9822 - mae: 0.0261 - mse: 0.0098 - val_loss: 0.9228 - val_accuracy: 0.7727 - val_mae: 0.1586 - val_mse: 0.1202 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1143/2000\n",
            "\n",
            "Epoch 1143: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0545 - accuracy: 0.9824 - mae: 0.0253 - mse: 0.0094 - val_loss: 0.9243 - val_accuracy: 0.7747 - val_mae: 0.1587 - val_mse: 0.1203 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1144/2000\n",
            "\n",
            "Epoch 1144: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0561 - accuracy: 0.9808 - mae: 0.0266 - mse: 0.0097 - val_loss: 0.9323 - val_accuracy: 0.7737 - val_mae: 0.1591 - val_mse: 0.1209 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1145/2000\n",
            "\n",
            "Epoch 1145: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0586 - accuracy: 0.9819 - mae: 0.0261 - mse: 0.0100 - val_loss: 0.9339 - val_accuracy: 0.7750 - val_mae: 0.1587 - val_mse: 0.1208 - lr: 1.0000e-04 - 313ms/epoch - 26ms/step\n",
            "Epoch 1146/2000\n",
            "\n",
            "Epoch 1146: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0599 - accuracy: 0.9787 - mae: 0.0268 - mse: 0.0106 - val_loss: 0.9295 - val_accuracy: 0.7753 - val_mae: 0.1586 - val_mse: 0.1205 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1147/2000\n",
            "\n",
            "Epoch 1147: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0552 - accuracy: 0.9813 - mae: 0.0256 - mse: 0.0097 - val_loss: 0.9257 - val_accuracy: 0.7733 - val_mae: 0.1589 - val_mse: 0.1206 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1148/2000\n",
            "\n",
            "Epoch 1148: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0568 - accuracy: 0.9815 - mae: 0.0262 - mse: 0.0098 - val_loss: 0.9248 - val_accuracy: 0.7730 - val_mae: 0.1590 - val_mse: 0.1207 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1149/2000\n",
            "\n",
            "Epoch 1149: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0545 - accuracy: 0.9830 - mae: 0.0248 - mse: 0.0093 - val_loss: 0.9266 - val_accuracy: 0.7720 - val_mae: 0.1589 - val_mse: 0.1207 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1150/2000\n",
            "\n",
            "Epoch 1150: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0574 - accuracy: 0.9817 - mae: 0.0256 - mse: 0.0098 - val_loss: 0.9221 - val_accuracy: 0.7720 - val_mae: 0.1589 - val_mse: 0.1206 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1151/2000\n",
            "\n",
            "Epoch 1151: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0567 - accuracy: 0.9818 - mae: 0.0258 - mse: 0.0097 - val_loss: 0.9285 - val_accuracy: 0.7727 - val_mae: 0.1588 - val_mse: 0.1206 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1152/2000\n",
            "\n",
            "Epoch 1152: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0541 - accuracy: 0.9832 - mae: 0.0252 - mse: 0.0092 - val_loss: 0.9279 - val_accuracy: 0.7727 - val_mae: 0.1587 - val_mse: 0.1205 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1153/2000\n",
            "\n",
            "Epoch 1153: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0576 - accuracy: 0.9809 - mae: 0.0263 - mse: 0.0101 - val_loss: 0.9282 - val_accuracy: 0.7753 - val_mae: 0.1587 - val_mse: 0.1207 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1154/2000\n",
            "\n",
            "Epoch 1154: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0553 - accuracy: 0.9818 - mae: 0.0252 - mse: 0.0096 - val_loss: 0.9204 - val_accuracy: 0.7747 - val_mae: 0.1590 - val_mse: 0.1207 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1155/2000\n",
            "\n",
            "Epoch 1155: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0575 - accuracy: 0.9818 - mae: 0.0261 - mse: 0.0099 - val_loss: 0.9200 - val_accuracy: 0.7737 - val_mae: 0.1589 - val_mse: 0.1205 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1156/2000\n",
            "\n",
            "Epoch 1156: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0572 - accuracy: 0.9825 - mae: 0.0263 - mse: 0.0098 - val_loss: 0.9151 - val_accuracy: 0.7743 - val_mae: 0.1587 - val_mse: 0.1201 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1157/2000\n",
            "\n",
            "Epoch 1157: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0585 - accuracy: 0.9808 - mae: 0.0265 - mse: 0.0102 - val_loss: 0.9257 - val_accuracy: 0.7723 - val_mae: 0.1584 - val_mse: 0.1204 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1158/2000\n",
            "\n",
            "Epoch 1158: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0601 - accuracy: 0.9803 - mae: 0.0270 - mse: 0.0105 - val_loss: 0.9242 - val_accuracy: 0.7733 - val_mae: 0.1585 - val_mse: 0.1204 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1159/2000\n",
            "\n",
            "Epoch 1159: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0570 - accuracy: 0.9816 - mae: 0.0259 - mse: 0.0098 - val_loss: 0.9250 - val_accuracy: 0.7733 - val_mae: 0.1588 - val_mse: 0.1205 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1160/2000\n",
            "\n",
            "Epoch 1160: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0571 - accuracy: 0.9826 - mae: 0.0256 - mse: 0.0096 - val_loss: 0.9268 - val_accuracy: 0.7757 - val_mae: 0.1587 - val_mse: 0.1207 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1161/2000\n",
            "\n",
            "Epoch 1161: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0591 - accuracy: 0.9815 - mae: 0.0263 - mse: 0.0102 - val_loss: 0.9315 - val_accuracy: 0.7760 - val_mae: 0.1587 - val_mse: 0.1209 - lr: 1.0000e-04 - 316ms/epoch - 26ms/step\n",
            "Epoch 1162/2000\n",
            "\n",
            "Epoch 1162: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0539 - accuracy: 0.9834 - mae: 0.0247 - mse: 0.0091 - val_loss: 0.9304 - val_accuracy: 0.7743 - val_mae: 0.1586 - val_mse: 0.1206 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1163/2000\n",
            "\n",
            "Epoch 1163: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0512 - accuracy: 0.9843 - mae: 0.0240 - mse: 0.0087 - val_loss: 0.9342 - val_accuracy: 0.7740 - val_mae: 0.1585 - val_mse: 0.1207 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1164/2000\n",
            "\n",
            "Epoch 1164: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0567 - accuracy: 0.9814 - mae: 0.0256 - mse: 0.0099 - val_loss: 0.9321 - val_accuracy: 0.7747 - val_mae: 0.1583 - val_mse: 0.1206 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1165/2000\n",
            "\n",
            "Epoch 1165: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0550 - accuracy: 0.9827 - mae: 0.0254 - mse: 0.0094 - val_loss: 0.9329 - val_accuracy: 0.7740 - val_mae: 0.1585 - val_mse: 0.1206 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1166/2000\n",
            "\n",
            "Epoch 1166: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0556 - accuracy: 0.9818 - mae: 0.0253 - mse: 0.0094 - val_loss: 0.9322 - val_accuracy: 0.7733 - val_mae: 0.1588 - val_mse: 0.1209 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1167/2000\n",
            "\n",
            "Epoch 1167: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0599 - accuracy: 0.9817 - mae: 0.0260 - mse: 0.0100 - val_loss: 0.9292 - val_accuracy: 0.7693 - val_mae: 0.1594 - val_mse: 0.1212 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1168/2000\n",
            "\n",
            "Epoch 1168: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0569 - accuracy: 0.9820 - mae: 0.0256 - mse: 0.0097 - val_loss: 0.9290 - val_accuracy: 0.7723 - val_mae: 0.1595 - val_mse: 0.1212 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1169/2000\n",
            "\n",
            "Epoch 1169: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0547 - accuracy: 0.9833 - mae: 0.0254 - mse: 0.0093 - val_loss: 0.9293 - val_accuracy: 0.7723 - val_mae: 0.1593 - val_mse: 0.1211 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1170/2000\n",
            "\n",
            "Epoch 1170: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0575 - accuracy: 0.9810 - mae: 0.0258 - mse: 0.0101 - val_loss: 0.9389 - val_accuracy: 0.7750 - val_mae: 0.1590 - val_mse: 0.1215 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1171/2000\n",
            "\n",
            "Epoch 1171: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0520 - accuracy: 0.9842 - mae: 0.0242 - mse: 0.0088 - val_loss: 0.9354 - val_accuracy: 0.7750 - val_mae: 0.1588 - val_mse: 0.1213 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1172/2000\n",
            "\n",
            "Epoch 1172: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0548 - accuracy: 0.9824 - mae: 0.0249 - mse: 0.0093 - val_loss: 0.9423 - val_accuracy: 0.7730 - val_mae: 0.1590 - val_mse: 0.1216 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1173/2000\n",
            "\n",
            "Epoch 1173: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0550 - accuracy: 0.9825 - mae: 0.0251 - mse: 0.0095 - val_loss: 0.9351 - val_accuracy: 0.7723 - val_mae: 0.1586 - val_mse: 0.1210 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1174/2000\n",
            "\n",
            "Epoch 1174: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0553 - accuracy: 0.9803 - mae: 0.0254 - mse: 0.0097 - val_loss: 0.9310 - val_accuracy: 0.7720 - val_mae: 0.1585 - val_mse: 0.1207 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 1175/2000\n",
            "\n",
            "Epoch 1175: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0565 - accuracy: 0.9819 - mae: 0.0255 - mse: 0.0097 - val_loss: 0.9383 - val_accuracy: 0.7727 - val_mae: 0.1586 - val_mse: 0.1212 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1176/2000\n",
            "\n",
            "Epoch 1176: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0543 - accuracy: 0.9835 - mae: 0.0244 - mse: 0.0090 - val_loss: 0.9376 - val_accuracy: 0.7723 - val_mae: 0.1592 - val_mse: 0.1215 - lr: 1.0000e-04 - 317ms/epoch - 26ms/step\n",
            "Epoch 1177/2000\n",
            "\n",
            "Epoch 1177: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0546 - accuracy: 0.9827 - mae: 0.0252 - mse: 0.0093 - val_loss: 0.9344 - val_accuracy: 0.7713 - val_mae: 0.1591 - val_mse: 0.1214 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1178/2000\n",
            "\n",
            "Epoch 1178: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0520 - accuracy: 0.9829 - mae: 0.0242 - mse: 0.0089 - val_loss: 0.9340 - val_accuracy: 0.7713 - val_mae: 0.1588 - val_mse: 0.1212 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1179/2000\n",
            "\n",
            "Epoch 1179: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0550 - accuracy: 0.9824 - mae: 0.0251 - mse: 0.0094 - val_loss: 0.9337 - val_accuracy: 0.7717 - val_mae: 0.1587 - val_mse: 0.1211 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1180/2000\n",
            "\n",
            "Epoch 1180: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0554 - accuracy: 0.9822 - mae: 0.0255 - mse: 0.0095 - val_loss: 0.9427 - val_accuracy: 0.7730 - val_mae: 0.1580 - val_mse: 0.1209 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1181/2000\n",
            "\n",
            "Epoch 1181: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0552 - accuracy: 0.9813 - mae: 0.0247 - mse: 0.0095 - val_loss: 0.9401 - val_accuracy: 0.7737 - val_mae: 0.1580 - val_mse: 0.1209 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1182/2000\n",
            "\n",
            "Epoch 1182: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0571 - accuracy: 0.9814 - mae: 0.0256 - mse: 0.0098 - val_loss: 0.9348 - val_accuracy: 0.7737 - val_mae: 0.1582 - val_mse: 0.1207 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1183/2000\n",
            "\n",
            "Epoch 1183: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0552 - accuracy: 0.9818 - mae: 0.0248 - mse: 0.0094 - val_loss: 0.9347 - val_accuracy: 0.7730 - val_mae: 0.1583 - val_mse: 0.1207 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1184/2000\n",
            "\n",
            "Epoch 1184: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0549 - accuracy: 0.9822 - mae: 0.0251 - mse: 0.0096 - val_loss: 0.9293 - val_accuracy: 0.7730 - val_mae: 0.1584 - val_mse: 0.1206 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1185/2000\n",
            "\n",
            "Epoch 1185: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0547 - accuracy: 0.9822 - mae: 0.0247 - mse: 0.0094 - val_loss: 0.9360 - val_accuracy: 0.7733 - val_mae: 0.1581 - val_mse: 0.1207 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1186/2000\n",
            "\n",
            "Epoch 1186: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0535 - accuracy: 0.9834 - mae: 0.0240 - mse: 0.0091 - val_loss: 0.9321 - val_accuracy: 0.7723 - val_mae: 0.1583 - val_mse: 0.1206 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1187/2000\n",
            "\n",
            "Epoch 1187: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0535 - accuracy: 0.9828 - mae: 0.0243 - mse: 0.0091 - val_loss: 0.9333 - val_accuracy: 0.7723 - val_mae: 0.1584 - val_mse: 0.1208 - lr: 1.0000e-04 - 306ms/epoch - 25ms/step\n",
            "Epoch 1188/2000\n",
            "\n",
            "Epoch 1188: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0592 - accuracy: 0.9806 - mae: 0.0258 - mse: 0.0103 - val_loss: 0.9340 - val_accuracy: 0.7717 - val_mae: 0.1588 - val_mse: 0.1210 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1189/2000\n",
            "\n",
            "Epoch 1189: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0547 - accuracy: 0.9834 - mae: 0.0251 - mse: 0.0094 - val_loss: 0.9313 - val_accuracy: 0.7707 - val_mae: 0.1589 - val_mse: 0.1209 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1190/2000\n",
            "\n",
            "Epoch 1190: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0536 - accuracy: 0.9832 - mae: 0.0247 - mse: 0.0091 - val_loss: 0.9359 - val_accuracy: 0.7723 - val_mae: 0.1584 - val_mse: 0.1208 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1191/2000\n",
            "\n",
            "Epoch 1191: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0508 - accuracy: 0.9832 - mae: 0.0236 - mse: 0.0087 - val_loss: 0.9394 - val_accuracy: 0.7750 - val_mae: 0.1579 - val_mse: 0.1205 - lr: 1.0000e-04 - 311ms/epoch - 26ms/step\n",
            "Epoch 1192/2000\n",
            "\n",
            "Epoch 1192: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0538 - accuracy: 0.9813 - mae: 0.0248 - mse: 0.0093 - val_loss: 0.9390 - val_accuracy: 0.7727 - val_mae: 0.1584 - val_mse: 0.1208 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1193/2000\n",
            "\n",
            "Epoch 1193: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0574 - accuracy: 0.9810 - mae: 0.0249 - mse: 0.0100 - val_loss: 0.9341 - val_accuracy: 0.7720 - val_mae: 0.1588 - val_mse: 0.1210 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1194/2000\n",
            "\n",
            "Epoch 1194: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0541 - accuracy: 0.9826 - mae: 0.0250 - mse: 0.0092 - val_loss: 0.9330 - val_accuracy: 0.7707 - val_mae: 0.1591 - val_mse: 0.1211 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1195/2000\n",
            "\n",
            "Epoch 1195: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0523 - accuracy: 0.9838 - mae: 0.0239 - mse: 0.0088 - val_loss: 0.9317 - val_accuracy: 0.7713 - val_mae: 0.1589 - val_mse: 0.1210 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1196/2000\n",
            "\n",
            "Epoch 1196: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0520 - accuracy: 0.9837 - mae: 0.0240 - mse: 0.0088 - val_loss: 0.9345 - val_accuracy: 0.7703 - val_mae: 0.1589 - val_mse: 0.1211 - lr: 1.0000e-04 - 306ms/epoch - 25ms/step\n",
            "Epoch 1197/2000\n",
            "\n",
            "Epoch 1197: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0516 - accuracy: 0.9841 - mae: 0.0239 - mse: 0.0087 - val_loss: 0.9430 - val_accuracy: 0.7737 - val_mae: 0.1582 - val_mse: 0.1209 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1198/2000\n",
            "\n",
            "Epoch 1198: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0510 - accuracy: 0.9830 - mae: 0.0235 - mse: 0.0087 - val_loss: 0.9450 - val_accuracy: 0.7750 - val_mae: 0.1577 - val_mse: 0.1205 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 1199/2000\n",
            "\n",
            "Epoch 1199: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0567 - accuracy: 0.9804 - mae: 0.0256 - mse: 0.0102 - val_loss: 0.9423 - val_accuracy: 0.7763 - val_mae: 0.1578 - val_mse: 0.1205 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1200/2000\n",
            "\n",
            "Epoch 1200: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0494 - accuracy: 0.9852 - mae: 0.0229 - mse: 0.0083 - val_loss: 0.9425 - val_accuracy: 0.7720 - val_mae: 0.1582 - val_mse: 0.1209 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1201/2000\n",
            "\n",
            "Epoch 1201: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0533 - accuracy: 0.9818 - mae: 0.0247 - mse: 0.0094 - val_loss: 0.9443 - val_accuracy: 0.7720 - val_mae: 0.1585 - val_mse: 0.1212 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1202/2000\n",
            "\n",
            "Epoch 1202: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0493 - accuracy: 0.9857 - mae: 0.0225 - mse: 0.0083 - val_loss: 0.9457 - val_accuracy: 0.7717 - val_mae: 0.1585 - val_mse: 0.1214 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1203/2000\n",
            "\n",
            "Epoch 1203: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0508 - accuracy: 0.9836 - mae: 0.0238 - mse: 0.0089 - val_loss: 0.9484 - val_accuracy: 0.7710 - val_mae: 0.1586 - val_mse: 0.1215 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1204/2000\n",
            "\n",
            "Epoch 1204: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0540 - accuracy: 0.9825 - mae: 0.0243 - mse: 0.0093 - val_loss: 0.9501 - val_accuracy: 0.7733 - val_mae: 0.1586 - val_mse: 0.1216 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1205/2000\n",
            "\n",
            "Epoch 1205: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0542 - accuracy: 0.9809 - mae: 0.0246 - mse: 0.0095 - val_loss: 0.9445 - val_accuracy: 0.7727 - val_mae: 0.1588 - val_mse: 0.1215 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1206/2000\n",
            "\n",
            "Epoch 1206: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0552 - accuracy: 0.9813 - mae: 0.0249 - mse: 0.0097 - val_loss: 0.9420 - val_accuracy: 0.7720 - val_mae: 0.1587 - val_mse: 0.1214 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1207/2000\n",
            "\n",
            "Epoch 1207: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0542 - accuracy: 0.9826 - mae: 0.0241 - mse: 0.0094 - val_loss: 0.9412 - val_accuracy: 0.7713 - val_mae: 0.1589 - val_mse: 0.1217 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1208/2000\n",
            "\n",
            "Epoch 1208: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0534 - accuracy: 0.9828 - mae: 0.0243 - mse: 0.0093 - val_loss: 0.9377 - val_accuracy: 0.7710 - val_mae: 0.1588 - val_mse: 0.1214 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1209/2000\n",
            "\n",
            "Epoch 1209: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0526 - accuracy: 0.9830 - mae: 0.0242 - mse: 0.0090 - val_loss: 0.9450 - val_accuracy: 0.7723 - val_mae: 0.1582 - val_mse: 0.1212 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1210/2000\n",
            "\n",
            "Epoch 1210: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0530 - accuracy: 0.9828 - mae: 0.0240 - mse: 0.0090 - val_loss: 0.9431 - val_accuracy: 0.7727 - val_mae: 0.1581 - val_mse: 0.1211 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1211/2000\n",
            "\n",
            "Epoch 1211: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0536 - accuracy: 0.9842 - mae: 0.0237 - mse: 0.0091 - val_loss: 0.9418 - val_accuracy: 0.7723 - val_mae: 0.1578 - val_mse: 0.1208 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1212/2000\n",
            "\n",
            "Epoch 1212: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0523 - accuracy: 0.9827 - mae: 0.0243 - mse: 0.0090 - val_loss: 0.9462 - val_accuracy: 0.7727 - val_mae: 0.1577 - val_mse: 0.1208 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 1213/2000\n",
            "\n",
            "Epoch 1213: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0509 - accuracy: 0.9842 - mae: 0.0237 - mse: 0.0088 - val_loss: 0.9454 - val_accuracy: 0.7750 - val_mae: 0.1575 - val_mse: 0.1206 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1214/2000\n",
            "\n",
            "Epoch 1214: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0532 - accuracy: 0.9818 - mae: 0.0244 - mse: 0.0093 - val_loss: 0.9492 - val_accuracy: 0.7750 - val_mae: 0.1581 - val_mse: 0.1212 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1215/2000\n",
            "\n",
            "Epoch 1215: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0527 - accuracy: 0.9823 - mae: 0.0237 - mse: 0.0092 - val_loss: 0.9439 - val_accuracy: 0.7727 - val_mae: 0.1585 - val_mse: 0.1213 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1216/2000\n",
            "\n",
            "Epoch 1216: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0553 - accuracy: 0.9802 - mae: 0.0253 - mse: 0.0099 - val_loss: 0.9410 - val_accuracy: 0.7763 - val_mae: 0.1584 - val_mse: 0.1213 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1217/2000\n",
            "\n",
            "Epoch 1217: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0501 - accuracy: 0.9840 - mae: 0.0232 - mse: 0.0086 - val_loss: 0.9435 - val_accuracy: 0.7730 - val_mae: 0.1583 - val_mse: 0.1213 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1218/2000\n",
            "\n",
            "Epoch 1218: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0516 - accuracy: 0.9835 - mae: 0.0239 - mse: 0.0088 - val_loss: 0.9472 - val_accuracy: 0.7730 - val_mae: 0.1582 - val_mse: 0.1215 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1219/2000\n",
            "\n",
            "Epoch 1219: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0514 - accuracy: 0.9844 - mae: 0.0233 - mse: 0.0087 - val_loss: 0.9532 - val_accuracy: 0.7737 - val_mae: 0.1584 - val_mse: 0.1218 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1220/2000\n",
            "\n",
            "Epoch 1220: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0524 - accuracy: 0.9833 - mae: 0.0240 - mse: 0.0090 - val_loss: 0.9488 - val_accuracy: 0.7710 - val_mae: 0.1583 - val_mse: 0.1214 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1221/2000\n",
            "\n",
            "Epoch 1221: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0529 - accuracy: 0.9828 - mae: 0.0241 - mse: 0.0092 - val_loss: 0.9452 - val_accuracy: 0.7720 - val_mae: 0.1582 - val_mse: 0.1212 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1222/2000\n",
            "\n",
            "Epoch 1222: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0539 - accuracy: 0.9828 - mae: 0.0245 - mse: 0.0093 - val_loss: 0.9490 - val_accuracy: 0.7733 - val_mae: 0.1583 - val_mse: 0.1214 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1223/2000\n",
            "\n",
            "Epoch 1223: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0577 - accuracy: 0.9818 - mae: 0.0255 - mse: 0.0100 - val_loss: 0.9470 - val_accuracy: 0.7743 - val_mae: 0.1583 - val_mse: 0.1213 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1224/2000\n",
            "\n",
            "Epoch 1224: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0503 - accuracy: 0.9840 - mae: 0.0228 - mse: 0.0085 - val_loss: 0.9469 - val_accuracy: 0.7737 - val_mae: 0.1584 - val_mse: 0.1213 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1225/2000\n",
            "\n",
            "Epoch 1225: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0539 - accuracy: 0.9835 - mae: 0.0242 - mse: 0.0091 - val_loss: 0.9450 - val_accuracy: 0.7713 - val_mae: 0.1585 - val_mse: 0.1212 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1226/2000\n",
            "\n",
            "Epoch 1226: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0513 - accuracy: 0.9852 - mae: 0.0237 - mse: 0.0086 - val_loss: 0.9466 - val_accuracy: 0.7703 - val_mae: 0.1581 - val_mse: 0.1211 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1227/2000\n",
            "\n",
            "Epoch 1227: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0523 - accuracy: 0.9833 - mae: 0.0239 - mse: 0.0090 - val_loss: 0.9498 - val_accuracy: 0.7750 - val_mae: 0.1579 - val_mse: 0.1210 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1228/2000\n",
            "\n",
            "Epoch 1228: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0503 - accuracy: 0.9839 - mae: 0.0235 - mse: 0.0087 - val_loss: 0.9477 - val_accuracy: 0.7717 - val_mae: 0.1581 - val_mse: 0.1212 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1229/2000\n",
            "\n",
            "Epoch 1229: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0554 - accuracy: 0.9815 - mae: 0.0248 - mse: 0.0097 - val_loss: 0.9482 - val_accuracy: 0.7743 - val_mae: 0.1578 - val_mse: 0.1209 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1230/2000\n",
            "\n",
            "Epoch 1230: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0514 - accuracy: 0.9828 - mae: 0.0236 - mse: 0.0089 - val_loss: 0.9459 - val_accuracy: 0.7720 - val_mae: 0.1576 - val_mse: 0.1209 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1231/2000\n",
            "\n",
            "Epoch 1231: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0519 - accuracy: 0.9818 - mae: 0.0235 - mse: 0.0090 - val_loss: 0.9493 - val_accuracy: 0.7730 - val_mae: 0.1575 - val_mse: 0.1208 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1232/2000\n",
            "\n",
            "Epoch 1232: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0492 - accuracy: 0.9845 - mae: 0.0226 - mse: 0.0083 - val_loss: 0.9556 - val_accuracy: 0.7727 - val_mae: 0.1575 - val_mse: 0.1211 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1233/2000\n",
            "\n",
            "Epoch 1233: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0489 - accuracy: 0.9847 - mae: 0.0228 - mse: 0.0084 - val_loss: 0.9561 - val_accuracy: 0.7737 - val_mae: 0.1574 - val_mse: 0.1212 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1234/2000\n",
            "\n",
            "Epoch 1234: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0493 - accuracy: 0.9827 - mae: 0.0227 - mse: 0.0086 - val_loss: 0.9635 - val_accuracy: 0.7737 - val_mae: 0.1574 - val_mse: 0.1214 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1235/2000\n",
            "\n",
            "Epoch 1235: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0504 - accuracy: 0.9831 - mae: 0.0226 - mse: 0.0087 - val_loss: 0.9592 - val_accuracy: 0.7723 - val_mae: 0.1577 - val_mse: 0.1216 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1236/2000\n",
            "\n",
            "Epoch 1236: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0532 - accuracy: 0.9823 - mae: 0.0242 - mse: 0.0093 - val_loss: 0.9543 - val_accuracy: 0.7713 - val_mae: 0.1582 - val_mse: 0.1216 - lr: 1.0000e-04 - 316ms/epoch - 26ms/step\n",
            "Epoch 1237/2000\n",
            "\n",
            "Epoch 1237: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0493 - accuracy: 0.9851 - mae: 0.0230 - mse: 0.0085 - val_loss: 0.9503 - val_accuracy: 0.7750 - val_mae: 0.1578 - val_mse: 0.1212 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1238/2000\n",
            "\n",
            "Epoch 1238: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0557 - accuracy: 0.9834 - mae: 0.0241 - mse: 0.0094 - val_loss: 0.9465 - val_accuracy: 0.7717 - val_mae: 0.1581 - val_mse: 0.1211 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1239/2000\n",
            "\n",
            "Epoch 1239: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0497 - accuracy: 0.9836 - mae: 0.0225 - mse: 0.0085 - val_loss: 0.9477 - val_accuracy: 0.7703 - val_mae: 0.1590 - val_mse: 0.1218 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1240/2000\n",
            "\n",
            "Epoch 1240: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0500 - accuracy: 0.9841 - mae: 0.0228 - mse: 0.0083 - val_loss: 0.9427 - val_accuracy: 0.7703 - val_mae: 0.1587 - val_mse: 0.1213 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1241/2000\n",
            "\n",
            "Epoch 1241: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0511 - accuracy: 0.9838 - mae: 0.0232 - mse: 0.0088 - val_loss: 0.9574 - val_accuracy: 0.7747 - val_mae: 0.1581 - val_mse: 0.1216 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1242/2000\n",
            "\n",
            "Epoch 1242: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0524 - accuracy: 0.9832 - mae: 0.0236 - mse: 0.0089 - val_loss: 0.9560 - val_accuracy: 0.7750 - val_mae: 0.1580 - val_mse: 0.1213 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1243/2000\n",
            "\n",
            "Epoch 1243: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0508 - accuracy: 0.9837 - mae: 0.0231 - mse: 0.0087 - val_loss: 0.9520 - val_accuracy: 0.7737 - val_mae: 0.1583 - val_mse: 0.1214 - lr: 1.0000e-04 - 313ms/epoch - 26ms/step\n",
            "Epoch 1244/2000\n",
            "\n",
            "Epoch 1244: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0489 - accuracy: 0.9839 - mae: 0.0227 - mse: 0.0084 - val_loss: 0.9526 - val_accuracy: 0.7733 - val_mae: 0.1581 - val_mse: 0.1213 - lr: 1.0000e-04 - 315ms/epoch - 26ms/step\n",
            "Epoch 1245/2000\n",
            "\n",
            "Epoch 1245: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0497 - accuracy: 0.9833 - mae: 0.0231 - mse: 0.0087 - val_loss: 0.9575 - val_accuracy: 0.7703 - val_mae: 0.1576 - val_mse: 0.1211 - lr: 1.0000e-04 - 318ms/epoch - 26ms/step\n",
            "Epoch 1246/2000\n",
            "\n",
            "Epoch 1246: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0512 - accuracy: 0.9840 - mae: 0.0231 - mse: 0.0089 - val_loss: 0.9599 - val_accuracy: 0.7727 - val_mae: 0.1578 - val_mse: 0.1214 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1247/2000\n",
            "\n",
            "Epoch 1247: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0548 - accuracy: 0.9806 - mae: 0.0246 - mse: 0.0096 - val_loss: 0.9506 - val_accuracy: 0.7740 - val_mae: 0.1578 - val_mse: 0.1210 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1248/2000\n",
            "\n",
            "Epoch 1248: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0531 - accuracy: 0.9842 - mae: 0.0238 - mse: 0.0090 - val_loss: 0.9492 - val_accuracy: 0.7723 - val_mae: 0.1581 - val_mse: 0.1211 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1249/2000\n",
            "\n",
            "Epoch 1249: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0529 - accuracy: 0.9823 - mae: 0.0234 - mse: 0.0090 - val_loss: 0.9535 - val_accuracy: 0.7740 - val_mae: 0.1582 - val_mse: 0.1213 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1250/2000\n",
            "\n",
            "Epoch 1250: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0526 - accuracy: 0.9818 - mae: 0.0239 - mse: 0.0091 - val_loss: 0.9498 - val_accuracy: 0.7727 - val_mae: 0.1580 - val_mse: 0.1211 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 1251/2000\n",
            "\n",
            "Epoch 1251: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0511 - accuracy: 0.9834 - mae: 0.0230 - mse: 0.0088 - val_loss: 0.9552 - val_accuracy: 0.7733 - val_mae: 0.1576 - val_mse: 0.1211 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1252/2000\n",
            "\n",
            "Epoch 1252: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0492 - accuracy: 0.9845 - mae: 0.0225 - mse: 0.0084 - val_loss: 0.9523 - val_accuracy: 0.7753 - val_mae: 0.1576 - val_mse: 0.1210 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1253/2000\n",
            "\n",
            "Epoch 1253: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0526 - accuracy: 0.9822 - mae: 0.0236 - mse: 0.0093 - val_loss: 0.9530 - val_accuracy: 0.7747 - val_mae: 0.1575 - val_mse: 0.1209 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1254/2000\n",
            "\n",
            "Epoch 1254: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0537 - accuracy: 0.9821 - mae: 0.0240 - mse: 0.0093 - val_loss: 0.9532 - val_accuracy: 0.7730 - val_mae: 0.1580 - val_mse: 0.1212 - lr: 1.0000e-04 - 318ms/epoch - 26ms/step\n",
            "Epoch 1255/2000\n",
            "\n",
            "Epoch 1255: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0509 - accuracy: 0.9830 - mae: 0.0234 - mse: 0.0087 - val_loss: 0.9448 - val_accuracy: 0.7730 - val_mae: 0.1582 - val_mse: 0.1210 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1256/2000\n",
            "\n",
            "Epoch 1256: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0506 - accuracy: 0.9833 - mae: 0.0234 - mse: 0.0087 - val_loss: 0.9532 - val_accuracy: 0.7733 - val_mae: 0.1579 - val_mse: 0.1212 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1257/2000\n",
            "\n",
            "Epoch 1257: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0506 - accuracy: 0.9825 - mae: 0.0230 - mse: 0.0088 - val_loss: 0.9546 - val_accuracy: 0.7757 - val_mae: 0.1573 - val_mse: 0.1209 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1258/2000\n",
            "\n",
            "Epoch 1258: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0527 - accuracy: 0.9833 - mae: 0.0230 - mse: 0.0089 - val_loss: 0.9617 - val_accuracy: 0.7750 - val_mae: 0.1575 - val_mse: 0.1213 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1259/2000\n",
            "\n",
            "Epoch 1259: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0444 - accuracy: 0.9865 - mae: 0.0215 - mse: 0.0075 - val_loss: 0.9598 - val_accuracy: 0.7737 - val_mae: 0.1573 - val_mse: 0.1211 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1260/2000\n",
            "\n",
            "Epoch 1260: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0494 - accuracy: 0.9842 - mae: 0.0220 - mse: 0.0084 - val_loss: 0.9565 - val_accuracy: 0.7740 - val_mae: 0.1573 - val_mse: 0.1210 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1261/2000\n",
            "\n",
            "Epoch 1261: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0505 - accuracy: 0.9837 - mae: 0.0229 - mse: 0.0088 - val_loss: 0.9560 - val_accuracy: 0.7740 - val_mae: 0.1574 - val_mse: 0.1210 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1262/2000\n",
            "\n",
            "Epoch 1262: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0491 - accuracy: 0.9848 - mae: 0.0221 - mse: 0.0083 - val_loss: 0.9612 - val_accuracy: 0.7747 - val_mae: 0.1571 - val_mse: 0.1210 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1263/2000\n",
            "\n",
            "Epoch 1263: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0484 - accuracy: 0.9857 - mae: 0.0219 - mse: 0.0080 - val_loss: 0.9597 - val_accuracy: 0.7767 - val_mae: 0.1570 - val_mse: 0.1209 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1264/2000\n",
            "\n",
            "Epoch 1264: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0510 - accuracy: 0.9842 - mae: 0.0234 - mse: 0.0088 - val_loss: 0.9525 - val_accuracy: 0.7740 - val_mae: 0.1571 - val_mse: 0.1207 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1265/2000\n",
            "\n",
            "Epoch 1265: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0490 - accuracy: 0.9842 - mae: 0.0228 - mse: 0.0085 - val_loss: 0.9514 - val_accuracy: 0.7723 - val_mae: 0.1573 - val_mse: 0.1206 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1266/2000\n",
            "\n",
            "Epoch 1266: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0500 - accuracy: 0.9837 - mae: 0.0227 - mse: 0.0086 - val_loss: 0.9526 - val_accuracy: 0.7743 - val_mae: 0.1573 - val_mse: 0.1207 - lr: 1.0000e-04 - 283ms/epoch - 24ms/step\n",
            "Epoch 1267/2000\n",
            "\n",
            "Epoch 1267: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0511 - accuracy: 0.9846 - mae: 0.0229 - mse: 0.0087 - val_loss: 0.9557 - val_accuracy: 0.7713 - val_mae: 0.1573 - val_mse: 0.1209 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1268/2000\n",
            "\n",
            "Epoch 1268: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0512 - accuracy: 0.9833 - mae: 0.0235 - mse: 0.0090 - val_loss: 0.9523 - val_accuracy: 0.7727 - val_mae: 0.1570 - val_mse: 0.1205 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1269/2000\n",
            "\n",
            "Epoch 1269: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0496 - accuracy: 0.9843 - mae: 0.0221 - mse: 0.0084 - val_loss: 0.9611 - val_accuracy: 0.7747 - val_mae: 0.1570 - val_mse: 0.1208 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1270/2000\n",
            "\n",
            "Epoch 1270: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0483 - accuracy: 0.9853 - mae: 0.0224 - mse: 0.0082 - val_loss: 0.9635 - val_accuracy: 0.7730 - val_mae: 0.1572 - val_mse: 0.1212 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1271/2000\n",
            "\n",
            "Epoch 1271: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0517 - accuracy: 0.9824 - mae: 0.0233 - mse: 0.0091 - val_loss: 0.9580 - val_accuracy: 0.7757 - val_mae: 0.1573 - val_mse: 0.1211 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1272/2000\n",
            "\n",
            "Epoch 1272: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0537 - accuracy: 0.9821 - mae: 0.0241 - mse: 0.0094 - val_loss: 0.9562 - val_accuracy: 0.7720 - val_mae: 0.1577 - val_mse: 0.1213 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1273/2000\n",
            "\n",
            "Epoch 1273: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0490 - accuracy: 0.9834 - mae: 0.0224 - mse: 0.0083 - val_loss: 0.9577 - val_accuracy: 0.7740 - val_mae: 0.1575 - val_mse: 0.1212 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1274/2000\n",
            "\n",
            "Epoch 1274: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0503 - accuracy: 0.9837 - mae: 0.0229 - mse: 0.0086 - val_loss: 0.9579 - val_accuracy: 0.7733 - val_mae: 0.1571 - val_mse: 0.1209 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1275/2000\n",
            "\n",
            "Epoch 1275: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0468 - accuracy: 0.9846 - mae: 0.0221 - mse: 0.0082 - val_loss: 0.9683 - val_accuracy: 0.7740 - val_mae: 0.1570 - val_mse: 0.1212 - lr: 1.0000e-04 - 306ms/epoch - 25ms/step\n",
            "Epoch 1276/2000\n",
            "\n",
            "Epoch 1276: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0501 - accuracy: 0.9837 - mae: 0.0225 - mse: 0.0087 - val_loss: 0.9684 - val_accuracy: 0.7757 - val_mae: 0.1566 - val_mse: 0.1209 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1277/2000\n",
            "\n",
            "Epoch 1277: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0491 - accuracy: 0.9846 - mae: 0.0223 - mse: 0.0085 - val_loss: 0.9648 - val_accuracy: 0.7743 - val_mae: 0.1572 - val_mse: 0.1212 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1278/2000\n",
            "\n",
            "Epoch 1278: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0542 - accuracy: 0.9815 - mae: 0.0234 - mse: 0.0094 - val_loss: 0.9563 - val_accuracy: 0.7707 - val_mae: 0.1578 - val_mse: 0.1214 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1279/2000\n",
            "\n",
            "Epoch 1279: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0547 - accuracy: 0.9818 - mae: 0.0238 - mse: 0.0097 - val_loss: 0.9530 - val_accuracy: 0.7727 - val_mae: 0.1582 - val_mse: 0.1215 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1280/2000\n",
            "\n",
            "Epoch 1280: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0491 - accuracy: 0.9845 - mae: 0.0228 - mse: 0.0083 - val_loss: 0.9569 - val_accuracy: 0.7757 - val_mae: 0.1580 - val_mse: 0.1214 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1281/2000\n",
            "\n",
            "Epoch 1281: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0501 - accuracy: 0.9838 - mae: 0.0230 - mse: 0.0087 - val_loss: 0.9517 - val_accuracy: 0.7747 - val_mae: 0.1579 - val_mse: 0.1213 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1282/2000\n",
            "\n",
            "Epoch 1282: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0520 - accuracy: 0.9836 - mae: 0.0234 - mse: 0.0088 - val_loss: 0.9507 - val_accuracy: 0.7730 - val_mae: 0.1576 - val_mse: 0.1210 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1283/2000\n",
            "\n",
            "Epoch 1283: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0500 - accuracy: 0.9843 - mae: 0.0225 - mse: 0.0086 - val_loss: 0.9568 - val_accuracy: 0.7737 - val_mae: 0.1576 - val_mse: 0.1212 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1284/2000\n",
            "\n",
            "Epoch 1284: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0458 - accuracy: 0.9845 - mae: 0.0222 - mse: 0.0079 - val_loss: 0.9572 - val_accuracy: 0.7720 - val_mae: 0.1580 - val_mse: 0.1217 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1285/2000\n",
            "\n",
            "Epoch 1285: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0495 - accuracy: 0.9839 - mae: 0.0225 - mse: 0.0085 - val_loss: 0.9601 - val_accuracy: 0.7717 - val_mae: 0.1579 - val_mse: 0.1218 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1286/2000\n",
            "\n",
            "Epoch 1286: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0516 - accuracy: 0.9830 - mae: 0.0232 - mse: 0.0090 - val_loss: 0.9628 - val_accuracy: 0.7720 - val_mae: 0.1574 - val_mse: 0.1215 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1287/2000\n",
            "\n",
            "Epoch 1287: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0455 - accuracy: 0.9855 - mae: 0.0211 - mse: 0.0077 - val_loss: 0.9644 - val_accuracy: 0.7743 - val_mae: 0.1569 - val_mse: 0.1211 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1288/2000\n",
            "\n",
            "Epoch 1288: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0480 - accuracy: 0.9841 - mae: 0.0222 - mse: 0.0082 - val_loss: 0.9623 - val_accuracy: 0.7727 - val_mae: 0.1570 - val_mse: 0.1211 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1289/2000\n",
            "\n",
            "Epoch 1289: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0489 - accuracy: 0.9849 - mae: 0.0223 - mse: 0.0083 - val_loss: 0.9609 - val_accuracy: 0.7750 - val_mae: 0.1572 - val_mse: 0.1211 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1290/2000\n",
            "\n",
            "Epoch 1290: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0509 - accuracy: 0.9833 - mae: 0.0229 - mse: 0.0088 - val_loss: 0.9704 - val_accuracy: 0.7730 - val_mae: 0.1577 - val_mse: 0.1218 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1291/2000\n",
            "\n",
            "Epoch 1291: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0527 - accuracy: 0.9834 - mae: 0.0227 - mse: 0.0089 - val_loss: 0.9573 - val_accuracy: 0.7730 - val_mae: 0.1578 - val_mse: 0.1212 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1292/2000\n",
            "\n",
            "Epoch 1292: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0521 - accuracy: 0.9826 - mae: 0.0237 - mse: 0.0091 - val_loss: 0.9597 - val_accuracy: 0.7737 - val_mae: 0.1579 - val_mse: 0.1213 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1293/2000\n",
            "\n",
            "Epoch 1293: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0497 - accuracy: 0.9830 - mae: 0.0231 - mse: 0.0087 - val_loss: 0.9641 - val_accuracy: 0.7757 - val_mae: 0.1575 - val_mse: 0.1213 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1294/2000\n",
            "\n",
            "Epoch 1294: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0482 - accuracy: 0.9858 - mae: 0.0221 - mse: 0.0081 - val_loss: 0.9705 - val_accuracy: 0.7757 - val_mae: 0.1571 - val_mse: 0.1213 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1295/2000\n",
            "\n",
            "Epoch 1295: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0463 - accuracy: 0.9849 - mae: 0.0215 - mse: 0.0080 - val_loss: 0.9652 - val_accuracy: 0.7747 - val_mae: 0.1573 - val_mse: 0.1214 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1296/2000\n",
            "\n",
            "Epoch 1296: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0494 - accuracy: 0.9843 - mae: 0.0223 - mse: 0.0086 - val_loss: 0.9645 - val_accuracy: 0.7747 - val_mae: 0.1576 - val_mse: 0.1213 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1297/2000\n",
            "\n",
            "Epoch 1297: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0506 - accuracy: 0.9828 - mae: 0.0227 - mse: 0.0090 - val_loss: 0.9669 - val_accuracy: 0.7733 - val_mae: 0.1574 - val_mse: 0.1215 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1298/2000\n",
            "\n",
            "Epoch 1298: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0487 - accuracy: 0.9838 - mae: 0.0219 - mse: 0.0083 - val_loss: 0.9660 - val_accuracy: 0.7743 - val_mae: 0.1576 - val_mse: 0.1217 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1299/2000\n",
            "\n",
            "Epoch 1299: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0535 - accuracy: 0.9825 - mae: 0.0230 - mse: 0.0093 - val_loss: 0.9607 - val_accuracy: 0.7760 - val_mae: 0.1577 - val_mse: 0.1213 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1300/2000\n",
            "\n",
            "Epoch 1300: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0471 - accuracy: 0.9856 - mae: 0.0219 - mse: 0.0080 - val_loss: 0.9590 - val_accuracy: 0.7750 - val_mae: 0.1574 - val_mse: 0.1211 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1301/2000\n",
            "\n",
            "Epoch 1301: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0493 - accuracy: 0.9852 - mae: 0.0220 - mse: 0.0084 - val_loss: 0.9626 - val_accuracy: 0.7743 - val_mae: 0.1574 - val_mse: 0.1214 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1302/2000\n",
            "\n",
            "Epoch 1302: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0493 - accuracy: 0.9841 - mae: 0.0220 - mse: 0.0085 - val_loss: 0.9632 - val_accuracy: 0.7720 - val_mae: 0.1575 - val_mse: 0.1215 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1303/2000\n",
            "\n",
            "Epoch 1303: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0476 - accuracy: 0.9854 - mae: 0.0213 - mse: 0.0081 - val_loss: 0.9666 - val_accuracy: 0.7750 - val_mae: 0.1573 - val_mse: 0.1213 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1304/2000\n",
            "\n",
            "Epoch 1304: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0488 - accuracy: 0.9847 - mae: 0.0222 - mse: 0.0082 - val_loss: 0.9636 - val_accuracy: 0.7727 - val_mae: 0.1574 - val_mse: 0.1211 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1305/2000\n",
            "\n",
            "Epoch 1305: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0496 - accuracy: 0.9835 - mae: 0.0226 - mse: 0.0086 - val_loss: 0.9625 - val_accuracy: 0.7717 - val_mae: 0.1578 - val_mse: 0.1214 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1306/2000\n",
            "\n",
            "Epoch 1306: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0439 - accuracy: 0.9859 - mae: 0.0207 - mse: 0.0073 - val_loss: 0.9674 - val_accuracy: 0.7743 - val_mae: 0.1575 - val_mse: 0.1214 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1307/2000\n",
            "\n",
            "Epoch 1307: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0471 - accuracy: 0.9857 - mae: 0.0214 - mse: 0.0080 - val_loss: 0.9642 - val_accuracy: 0.7733 - val_mae: 0.1572 - val_mse: 0.1213 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1308/2000\n",
            "\n",
            "Epoch 1308: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0481 - accuracy: 0.9851 - mae: 0.0220 - mse: 0.0081 - val_loss: 0.9679 - val_accuracy: 0.7743 - val_mae: 0.1572 - val_mse: 0.1214 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1309/2000\n",
            "\n",
            "Epoch 1309: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0490 - accuracy: 0.9840 - mae: 0.0223 - mse: 0.0084 - val_loss: 0.9668 - val_accuracy: 0.7720 - val_mae: 0.1572 - val_mse: 0.1213 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1310/2000\n",
            "\n",
            "Epoch 1310: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0480 - accuracy: 0.9845 - mae: 0.0217 - mse: 0.0083 - val_loss: 0.9660 - val_accuracy: 0.7723 - val_mae: 0.1574 - val_mse: 0.1213 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1311/2000\n",
            "\n",
            "Epoch 1311: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0460 - accuracy: 0.9850 - mae: 0.0210 - mse: 0.0080 - val_loss: 0.9633 - val_accuracy: 0.7727 - val_mae: 0.1577 - val_mse: 0.1214 - lr: 1.0000e-04 - 332ms/epoch - 28ms/step\n",
            "Epoch 1312/2000\n",
            "\n",
            "Epoch 1312: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0483 - accuracy: 0.9845 - mae: 0.0217 - mse: 0.0082 - val_loss: 0.9654 - val_accuracy: 0.7737 - val_mae: 0.1577 - val_mse: 0.1215 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1313/2000\n",
            "\n",
            "Epoch 1313: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0450 - accuracy: 0.9851 - mae: 0.0207 - mse: 0.0077 - val_loss: 0.9698 - val_accuracy: 0.7723 - val_mae: 0.1575 - val_mse: 0.1216 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1314/2000\n",
            "\n",
            "Epoch 1314: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0483 - accuracy: 0.9847 - mae: 0.0220 - mse: 0.0082 - val_loss: 0.9672 - val_accuracy: 0.7747 - val_mae: 0.1572 - val_mse: 0.1212 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1315/2000\n",
            "\n",
            "Epoch 1315: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0498 - accuracy: 0.9835 - mae: 0.0224 - mse: 0.0086 - val_loss: 0.9721 - val_accuracy: 0.7757 - val_mae: 0.1570 - val_mse: 0.1213 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1316/2000\n",
            "\n",
            "Epoch 1316: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0480 - accuracy: 0.9841 - mae: 0.0215 - mse: 0.0082 - val_loss: 0.9658 - val_accuracy: 0.7753 - val_mae: 0.1574 - val_mse: 0.1215 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1317/2000\n",
            "\n",
            "Epoch 1317: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0489 - accuracy: 0.9829 - mae: 0.0224 - mse: 0.0086 - val_loss: 0.9695 - val_accuracy: 0.7740 - val_mae: 0.1576 - val_mse: 0.1216 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1318/2000\n",
            "\n",
            "Epoch 1318: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0508 - accuracy: 0.9841 - mae: 0.0226 - mse: 0.0088 - val_loss: 0.9673 - val_accuracy: 0.7720 - val_mae: 0.1577 - val_mse: 0.1216 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1319/2000\n",
            "\n",
            "Epoch 1319: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0488 - accuracy: 0.9837 - mae: 0.0221 - mse: 0.0085 - val_loss: 0.9671 - val_accuracy: 0.7730 - val_mae: 0.1577 - val_mse: 0.1216 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1320/2000\n",
            "\n",
            "Epoch 1320: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0490 - accuracy: 0.9836 - mae: 0.0216 - mse: 0.0084 - val_loss: 0.9710 - val_accuracy: 0.7717 - val_mae: 0.1579 - val_mse: 0.1221 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1321/2000\n",
            "\n",
            "Epoch 1321: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0474 - accuracy: 0.9845 - mae: 0.0215 - mse: 0.0082 - val_loss: 0.9646 - val_accuracy: 0.7730 - val_mae: 0.1580 - val_mse: 0.1219 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1322/2000\n",
            "\n",
            "Epoch 1322: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0466 - accuracy: 0.9838 - mae: 0.0218 - mse: 0.0080 - val_loss: 0.9687 - val_accuracy: 0.7717 - val_mae: 0.1581 - val_mse: 0.1220 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1323/2000\n",
            "\n",
            "Epoch 1323: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0519 - accuracy: 0.9839 - mae: 0.0222 - mse: 0.0088 - val_loss: 0.9699 - val_accuracy: 0.7717 - val_mae: 0.1580 - val_mse: 0.1221 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1324/2000\n",
            "\n",
            "Epoch 1324: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0486 - accuracy: 0.9836 - mae: 0.0223 - mse: 0.0085 - val_loss: 0.9708 - val_accuracy: 0.7747 - val_mae: 0.1579 - val_mse: 0.1221 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1325/2000\n",
            "\n",
            "Epoch 1325: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0439 - accuracy: 0.9863 - mae: 0.0204 - mse: 0.0074 - val_loss: 0.9718 - val_accuracy: 0.7747 - val_mae: 0.1576 - val_mse: 0.1219 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1326/2000\n",
            "\n",
            "Epoch 1326: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0443 - accuracy: 0.9856 - mae: 0.0206 - mse: 0.0076 - val_loss: 0.9711 - val_accuracy: 0.7743 - val_mae: 0.1571 - val_mse: 0.1214 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1327/2000\n",
            "\n",
            "Epoch 1327: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0474 - accuracy: 0.9850 - mae: 0.0212 - mse: 0.0081 - val_loss: 0.9719 - val_accuracy: 0.7737 - val_mae: 0.1575 - val_mse: 0.1217 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1328/2000\n",
            "\n",
            "Epoch 1328: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0488 - accuracy: 0.9848 - mae: 0.0216 - mse: 0.0082 - val_loss: 0.9671 - val_accuracy: 0.7737 - val_mae: 0.1576 - val_mse: 0.1216 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1329/2000\n",
            "\n",
            "Epoch 1329: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0497 - accuracy: 0.9837 - mae: 0.0220 - mse: 0.0087 - val_loss: 0.9616 - val_accuracy: 0.7747 - val_mae: 0.1578 - val_mse: 0.1215 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1330/2000\n",
            "\n",
            "Epoch 1330: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0447 - accuracy: 0.9862 - mae: 0.0209 - mse: 0.0075 - val_loss: 0.9697 - val_accuracy: 0.7740 - val_mae: 0.1577 - val_mse: 0.1216 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1331/2000\n",
            "\n",
            "Epoch 1331: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0440 - accuracy: 0.9857 - mae: 0.0204 - mse: 0.0076 - val_loss: 0.9715 - val_accuracy: 0.7753 - val_mae: 0.1573 - val_mse: 0.1215 - lr: 1.0000e-04 - 318ms/epoch - 26ms/step\n",
            "Epoch 1332/2000\n",
            "\n",
            "Epoch 1332: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0521 - accuracy: 0.9822 - mae: 0.0227 - mse: 0.0091 - val_loss: 0.9689 - val_accuracy: 0.7760 - val_mae: 0.1574 - val_mse: 0.1215 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1333/2000\n",
            "\n",
            "Epoch 1333: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0453 - accuracy: 0.9863 - mae: 0.0205 - mse: 0.0078 - val_loss: 0.9713 - val_accuracy: 0.7733 - val_mae: 0.1577 - val_mse: 0.1217 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1334/2000\n",
            "\n",
            "Epoch 1334: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0483 - accuracy: 0.9841 - mae: 0.0220 - mse: 0.0085 - val_loss: 0.9698 - val_accuracy: 0.7750 - val_mae: 0.1575 - val_mse: 0.1215 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1335/2000\n",
            "\n",
            "Epoch 1335: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0499 - accuracy: 0.9832 - mae: 0.0221 - mse: 0.0088 - val_loss: 0.9814 - val_accuracy: 0.7730 - val_mae: 0.1574 - val_mse: 0.1218 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1336/2000\n",
            "\n",
            "Epoch 1336: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0435 - accuracy: 0.9858 - mae: 0.0203 - mse: 0.0075 - val_loss: 0.9783 - val_accuracy: 0.7743 - val_mae: 0.1570 - val_mse: 0.1215 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1337/2000\n",
            "\n",
            "Epoch 1337: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0423 - accuracy: 0.9878 - mae: 0.0196 - mse: 0.0069 - val_loss: 0.9787 - val_accuracy: 0.7710 - val_mae: 0.1573 - val_mse: 0.1217 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1338/2000\n",
            "\n",
            "Epoch 1338: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0444 - accuracy: 0.9862 - mae: 0.0200 - mse: 0.0075 - val_loss: 0.9824 - val_accuracy: 0.7720 - val_mae: 0.1572 - val_mse: 0.1219 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1339/2000\n",
            "\n",
            "Epoch 1339: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0487 - accuracy: 0.9834 - mae: 0.0218 - mse: 0.0085 - val_loss: 0.9772 - val_accuracy: 0.7707 - val_mae: 0.1574 - val_mse: 0.1219 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1340/2000\n",
            "\n",
            "Epoch 1340: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0461 - accuracy: 0.9857 - mae: 0.0206 - mse: 0.0078 - val_loss: 0.9760 - val_accuracy: 0.7733 - val_mae: 0.1574 - val_mse: 0.1218 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1341/2000\n",
            "\n",
            "Epoch 1341: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0487 - accuracy: 0.9846 - mae: 0.0220 - mse: 0.0084 - val_loss: 0.9738 - val_accuracy: 0.7733 - val_mae: 0.1577 - val_mse: 0.1219 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1342/2000\n",
            "\n",
            "Epoch 1342: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0468 - accuracy: 0.9851 - mae: 0.0212 - mse: 0.0080 - val_loss: 0.9699 - val_accuracy: 0.7693 - val_mae: 0.1581 - val_mse: 0.1220 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 1343/2000\n",
            "\n",
            "Epoch 1343: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0456 - accuracy: 0.9861 - mae: 0.0213 - mse: 0.0077 - val_loss: 0.9718 - val_accuracy: 0.7697 - val_mae: 0.1578 - val_mse: 0.1220 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1344/2000\n",
            "\n",
            "Epoch 1344: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0482 - accuracy: 0.9838 - mae: 0.0219 - mse: 0.0085 - val_loss: 0.9747 - val_accuracy: 0.7710 - val_mae: 0.1577 - val_mse: 0.1219 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1345/2000\n",
            "\n",
            "Epoch 1345: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0488 - accuracy: 0.9843 - mae: 0.0215 - mse: 0.0083 - val_loss: 0.9722 - val_accuracy: 0.7697 - val_mae: 0.1573 - val_mse: 0.1214 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1346/2000\n",
            "\n",
            "Epoch 1346: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0471 - accuracy: 0.9857 - mae: 0.0214 - mse: 0.0080 - val_loss: 0.9770 - val_accuracy: 0.7710 - val_mae: 0.1577 - val_mse: 0.1219 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1347/2000\n",
            "\n",
            "Epoch 1347: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0488 - accuracy: 0.9851 - mae: 0.0219 - mse: 0.0084 - val_loss: 0.9718 - val_accuracy: 0.7697 - val_mae: 0.1575 - val_mse: 0.1215 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1348/2000\n",
            "\n",
            "Epoch 1348: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0454 - accuracy: 0.9860 - mae: 0.0204 - mse: 0.0076 - val_loss: 0.9793 - val_accuracy: 0.7727 - val_mae: 0.1569 - val_mse: 0.1215 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1349/2000\n",
            "\n",
            "Epoch 1349: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0459 - accuracy: 0.9855 - mae: 0.0207 - mse: 0.0078 - val_loss: 0.9761 - val_accuracy: 0.7717 - val_mae: 0.1568 - val_mse: 0.1213 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1350/2000\n",
            "\n",
            "Epoch 1350: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0444 - accuracy: 0.9852 - mae: 0.0207 - mse: 0.0077 - val_loss: 0.9772 - val_accuracy: 0.7717 - val_mae: 0.1567 - val_mse: 0.1214 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1351/2000\n",
            "\n",
            "Epoch 1351: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0455 - accuracy: 0.9860 - mae: 0.0205 - mse: 0.0078 - val_loss: 0.9809 - val_accuracy: 0.7710 - val_mae: 0.1571 - val_mse: 0.1219 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1352/2000\n",
            "\n",
            "Epoch 1352: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0466 - accuracy: 0.9852 - mae: 0.0208 - mse: 0.0080 - val_loss: 0.9780 - val_accuracy: 0.7737 - val_mae: 0.1570 - val_mse: 0.1216 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1353/2000\n",
            "\n",
            "Epoch 1353: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0461 - accuracy: 0.9843 - mae: 0.0211 - mse: 0.0080 - val_loss: 0.9827 - val_accuracy: 0.7743 - val_mae: 0.1570 - val_mse: 0.1217 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1354/2000\n",
            "\n",
            "Epoch 1354: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0454 - accuracy: 0.9861 - mae: 0.0204 - mse: 0.0077 - val_loss: 0.9792 - val_accuracy: 0.7743 - val_mae: 0.1571 - val_mse: 0.1217 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1355/2000\n",
            "\n",
            "Epoch 1355: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0434 - accuracy: 0.9859 - mae: 0.0198 - mse: 0.0074 - val_loss: 0.9812 - val_accuracy: 0.7753 - val_mae: 0.1570 - val_mse: 0.1217 - lr: 1.0000e-04 - 306ms/epoch - 25ms/step\n",
            "Epoch 1356/2000\n",
            "\n",
            "Epoch 1356: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0463 - accuracy: 0.9832 - mae: 0.0213 - mse: 0.0083 - val_loss: 0.9782 - val_accuracy: 0.7750 - val_mae: 0.1570 - val_mse: 0.1214 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1357/2000\n",
            "\n",
            "Epoch 1357: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0430 - accuracy: 0.9863 - mae: 0.0205 - mse: 0.0073 - val_loss: 0.9771 - val_accuracy: 0.7733 - val_mae: 0.1571 - val_mse: 0.1215 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1358/2000\n",
            "\n",
            "Epoch 1358: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0459 - accuracy: 0.9852 - mae: 0.0207 - mse: 0.0077 - val_loss: 0.9783 - val_accuracy: 0.7727 - val_mae: 0.1571 - val_mse: 0.1217 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1359/2000\n",
            "\n",
            "Epoch 1359: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0481 - accuracy: 0.9848 - mae: 0.0208 - mse: 0.0080 - val_loss: 0.9787 - val_accuracy: 0.7753 - val_mae: 0.1573 - val_mse: 0.1217 - lr: 1.0000e-04 - 318ms/epoch - 27ms/step\n",
            "Epoch 1360/2000\n",
            "\n",
            "Epoch 1360: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0426 - accuracy: 0.9875 - mae: 0.0201 - mse: 0.0071 - val_loss: 0.9808 - val_accuracy: 0.7743 - val_mae: 0.1573 - val_mse: 0.1220 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1361/2000\n",
            "\n",
            "Epoch 1361: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0460 - accuracy: 0.9847 - mae: 0.0208 - mse: 0.0080 - val_loss: 0.9784 - val_accuracy: 0.7737 - val_mae: 0.1573 - val_mse: 0.1219 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1362/2000\n",
            "\n",
            "Epoch 1362: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0462 - accuracy: 0.9844 - mae: 0.0209 - mse: 0.0079 - val_loss: 0.9873 - val_accuracy: 0.7737 - val_mae: 0.1569 - val_mse: 0.1220 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1363/2000\n",
            "\n",
            "Epoch 1363: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0432 - accuracy: 0.9858 - mae: 0.0202 - mse: 0.0075 - val_loss: 0.9845 - val_accuracy: 0.7740 - val_mae: 0.1567 - val_mse: 0.1218 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1364/2000\n",
            "\n",
            "Epoch 1364: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0458 - accuracy: 0.9858 - mae: 0.0206 - mse: 0.0077 - val_loss: 0.9868 - val_accuracy: 0.7733 - val_mae: 0.1570 - val_mse: 0.1220 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1365/2000\n",
            "\n",
            "Epoch 1365: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0488 - accuracy: 0.9828 - mae: 0.0216 - mse: 0.0086 - val_loss: 0.9808 - val_accuracy: 0.7737 - val_mae: 0.1571 - val_mse: 0.1216 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1366/2000\n",
            "\n",
            "Epoch 1366: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0454 - accuracy: 0.9858 - mae: 0.0208 - mse: 0.0077 - val_loss: 0.9798 - val_accuracy: 0.7727 - val_mae: 0.1573 - val_mse: 0.1218 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1367/2000\n",
            "\n",
            "Epoch 1367: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0464 - accuracy: 0.9843 - mae: 0.0207 - mse: 0.0079 - val_loss: 0.9844 - val_accuracy: 0.7743 - val_mae: 0.1572 - val_mse: 0.1220 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 1368/2000\n",
            "\n",
            "Epoch 1368: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0501 - accuracy: 0.9829 - mae: 0.0222 - mse: 0.0089 - val_loss: 0.9846 - val_accuracy: 0.7727 - val_mae: 0.1568 - val_mse: 0.1217 - lr: 1.0000e-04 - 314ms/epoch - 26ms/step\n",
            "Epoch 1369/2000\n",
            "\n",
            "Epoch 1369: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0463 - accuracy: 0.9858 - mae: 0.0208 - mse: 0.0079 - val_loss: 0.9865 - val_accuracy: 0.7717 - val_mae: 0.1572 - val_mse: 0.1220 - lr: 1.0000e-04 - 306ms/epoch - 25ms/step\n",
            "Epoch 1370/2000\n",
            "\n",
            "Epoch 1370: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0430 - accuracy: 0.9854 - mae: 0.0200 - mse: 0.0074 - val_loss: 0.9872 - val_accuracy: 0.7730 - val_mae: 0.1570 - val_mse: 0.1219 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1371/2000\n",
            "\n",
            "Epoch 1371: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0470 - accuracy: 0.9854 - mae: 0.0209 - mse: 0.0080 - val_loss: 0.9897 - val_accuracy: 0.7753 - val_mae: 0.1568 - val_mse: 0.1220 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1372/2000\n",
            "\n",
            "Epoch 1372: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0443 - accuracy: 0.9870 - mae: 0.0206 - mse: 0.0076 - val_loss: 0.9860 - val_accuracy: 0.7753 - val_mae: 0.1569 - val_mse: 0.1219 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 1373/2000\n",
            "\n",
            "Epoch 1373: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0444 - accuracy: 0.9872 - mae: 0.0201 - mse: 0.0073 - val_loss: 0.9924 - val_accuracy: 0.7727 - val_mae: 0.1572 - val_mse: 0.1225 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1374/2000\n",
            "\n",
            "Epoch 1374: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0450 - accuracy: 0.9858 - mae: 0.0207 - mse: 0.0077 - val_loss: 0.9921 - val_accuracy: 0.7723 - val_mae: 0.1573 - val_mse: 0.1224 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1375/2000\n",
            "\n",
            "Epoch 1375: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0479 - accuracy: 0.9841 - mae: 0.0212 - mse: 0.0082 - val_loss: 0.9864 - val_accuracy: 0.7750 - val_mae: 0.1577 - val_mse: 0.1224 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1376/2000\n",
            "\n",
            "Epoch 1376: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0445 - accuracy: 0.9855 - mae: 0.0204 - mse: 0.0076 - val_loss: 0.9815 - val_accuracy: 0.7747 - val_mae: 0.1580 - val_mse: 0.1224 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1377/2000\n",
            "\n",
            "Epoch 1377: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0445 - accuracy: 0.9859 - mae: 0.0208 - mse: 0.0076 - val_loss: 0.9802 - val_accuracy: 0.7740 - val_mae: 0.1581 - val_mse: 0.1224 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1378/2000\n",
            "\n",
            "Epoch 1378: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0461 - accuracy: 0.9848 - mae: 0.0204 - mse: 0.0078 - val_loss: 0.9819 - val_accuracy: 0.7727 - val_mae: 0.1578 - val_mse: 0.1224 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1379/2000\n",
            "\n",
            "Epoch 1379: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0446 - accuracy: 0.9858 - mae: 0.0206 - mse: 0.0076 - val_loss: 0.9879 - val_accuracy: 0.7747 - val_mae: 0.1570 - val_mse: 0.1220 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1380/2000\n",
            "\n",
            "Epoch 1380: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0451 - accuracy: 0.9843 - mae: 0.0205 - mse: 0.0080 - val_loss: 0.9861 - val_accuracy: 0.7757 - val_mae: 0.1565 - val_mse: 0.1216 - lr: 1.0000e-04 - 315ms/epoch - 26ms/step\n",
            "Epoch 1381/2000\n",
            "\n",
            "Epoch 1381: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0419 - accuracy: 0.9868 - mae: 0.0195 - mse: 0.0070 - val_loss: 0.9906 - val_accuracy: 0.7730 - val_mae: 0.1567 - val_mse: 0.1217 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1382/2000\n",
            "\n",
            "Epoch 1382: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0467 - accuracy: 0.9849 - mae: 0.0210 - mse: 0.0080 - val_loss: 0.9861 - val_accuracy: 0.7713 - val_mae: 0.1569 - val_mse: 0.1217 - lr: 1.0000e-04 - 317ms/epoch - 26ms/step\n",
            "Epoch 1383/2000\n",
            "\n",
            "Epoch 1383: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0423 - accuracy: 0.9874 - mae: 0.0199 - mse: 0.0072 - val_loss: 0.9879 - val_accuracy: 0.7710 - val_mae: 0.1572 - val_mse: 0.1222 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1384/2000\n",
            "\n",
            "Epoch 1384: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0419 - accuracy: 0.9877 - mae: 0.0193 - mse: 0.0069 - val_loss: 0.9900 - val_accuracy: 0.7717 - val_mae: 0.1574 - val_mse: 0.1226 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1385/2000\n",
            "\n",
            "Epoch 1385: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0458 - accuracy: 0.9861 - mae: 0.0203 - mse: 0.0077 - val_loss: 0.9871 - val_accuracy: 0.7743 - val_mae: 0.1577 - val_mse: 0.1227 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1386/2000\n",
            "\n",
            "Epoch 1386: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0438 - accuracy: 0.9852 - mae: 0.0200 - mse: 0.0075 - val_loss: 0.9939 - val_accuracy: 0.7727 - val_mae: 0.1577 - val_mse: 0.1227 - lr: 1.0000e-04 - 318ms/epoch - 26ms/step\n",
            "Epoch 1387/2000\n",
            "\n",
            "Epoch 1387: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0447 - accuracy: 0.9860 - mae: 0.0199 - mse: 0.0077 - val_loss: 0.9838 - val_accuracy: 0.7740 - val_mae: 0.1578 - val_mse: 0.1226 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1388/2000\n",
            "\n",
            "Epoch 1388: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0475 - accuracy: 0.9857 - mae: 0.0206 - mse: 0.0080 - val_loss: 0.9870 - val_accuracy: 0.7740 - val_mae: 0.1574 - val_mse: 0.1224 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1389/2000\n",
            "\n",
            "Epoch 1389: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0441 - accuracy: 0.9872 - mae: 0.0199 - mse: 0.0073 - val_loss: 0.9916 - val_accuracy: 0.7730 - val_mae: 0.1568 - val_mse: 0.1222 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1390/2000\n",
            "\n",
            "Epoch 1390: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0476 - accuracy: 0.9848 - mae: 0.0203 - mse: 0.0080 - val_loss: 0.9886 - val_accuracy: 0.7737 - val_mae: 0.1565 - val_mse: 0.1218 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1391/2000\n",
            "\n",
            "Epoch 1391: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0455 - accuracy: 0.9853 - mae: 0.0202 - mse: 0.0078 - val_loss: 0.9861 - val_accuracy: 0.7710 - val_mae: 0.1570 - val_mse: 0.1221 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1392/2000\n",
            "\n",
            "Epoch 1392: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0449 - accuracy: 0.9859 - mae: 0.0209 - mse: 0.0077 - val_loss: 0.9850 - val_accuracy: 0.7733 - val_mae: 0.1571 - val_mse: 0.1222 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1393/2000\n",
            "\n",
            "Epoch 1393: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0466 - accuracy: 0.9848 - mae: 0.0209 - mse: 0.0080 - val_loss: 0.9836 - val_accuracy: 0.7757 - val_mae: 0.1568 - val_mse: 0.1219 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1394/2000\n",
            "\n",
            "Epoch 1394: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0425 - accuracy: 0.9858 - mae: 0.0196 - mse: 0.0074 - val_loss: 0.9869 - val_accuracy: 0.7733 - val_mae: 0.1567 - val_mse: 0.1218 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1395/2000\n",
            "\n",
            "Epoch 1395: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0436 - accuracy: 0.9855 - mae: 0.0202 - mse: 0.0076 - val_loss: 0.9859 - val_accuracy: 0.7753 - val_mae: 0.1566 - val_mse: 0.1218 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1396/2000\n",
            "\n",
            "Epoch 1396: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0466 - accuracy: 0.9846 - mae: 0.0206 - mse: 0.0080 - val_loss: 0.9951 - val_accuracy: 0.7730 - val_mae: 0.1571 - val_mse: 0.1225 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1397/2000\n",
            "\n",
            "Epoch 1397: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0476 - accuracy: 0.9845 - mae: 0.0211 - mse: 0.0081 - val_loss: 0.9864 - val_accuracy: 0.7733 - val_mae: 0.1572 - val_mse: 0.1222 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1398/2000\n",
            "\n",
            "Epoch 1398: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0458 - accuracy: 0.9854 - mae: 0.0208 - mse: 0.0078 - val_loss: 0.9903 - val_accuracy: 0.7747 - val_mae: 0.1568 - val_mse: 0.1218 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1399/2000\n",
            "\n",
            "Epoch 1399: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0425 - accuracy: 0.9854 - mae: 0.0199 - mse: 0.0073 - val_loss: 0.9945 - val_accuracy: 0.7727 - val_mae: 0.1570 - val_mse: 0.1223 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1400/2000\n",
            "\n",
            "Epoch 1400: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0437 - accuracy: 0.9868 - mae: 0.0202 - mse: 0.0073 - val_loss: 0.9926 - val_accuracy: 0.7723 - val_mae: 0.1567 - val_mse: 0.1219 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1401/2000\n",
            "\n",
            "Epoch 1401: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0394 - accuracy: 0.9869 - mae: 0.0184 - mse: 0.0067 - val_loss: 0.9928 - val_accuracy: 0.7747 - val_mae: 0.1565 - val_mse: 0.1218 - lr: 1.0000e-04 - 314ms/epoch - 26ms/step\n",
            "Epoch 1402/2000\n",
            "\n",
            "Epoch 1402: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0457 - accuracy: 0.9840 - mae: 0.0203 - mse: 0.0080 - val_loss: 1.0007 - val_accuracy: 0.7740 - val_mae: 0.1567 - val_mse: 0.1221 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1403/2000\n",
            "\n",
            "Epoch 1403: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0484 - accuracy: 0.9846 - mae: 0.0208 - mse: 0.0082 - val_loss: 0.9914 - val_accuracy: 0.7750 - val_mae: 0.1569 - val_mse: 0.1220 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1404/2000\n",
            "\n",
            "Epoch 1404: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0428 - accuracy: 0.9866 - mae: 0.0200 - mse: 0.0074 - val_loss: 0.9894 - val_accuracy: 0.7750 - val_mae: 0.1567 - val_mse: 0.1216 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1405/2000\n",
            "\n",
            "Epoch 1405: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0440 - accuracy: 0.9858 - mae: 0.0200 - mse: 0.0077 - val_loss: 0.9857 - val_accuracy: 0.7737 - val_mae: 0.1566 - val_mse: 0.1216 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1406/2000\n",
            "\n",
            "Epoch 1406: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0430 - accuracy: 0.9868 - mae: 0.0198 - mse: 0.0074 - val_loss: 0.9966 - val_accuracy: 0.7753 - val_mae: 0.1563 - val_mse: 0.1216 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 1407/2000\n",
            "\n",
            "Epoch 1407: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0395 - accuracy: 0.9877 - mae: 0.0189 - mse: 0.0068 - val_loss: 0.9930 - val_accuracy: 0.7760 - val_mae: 0.1563 - val_mse: 0.1216 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1408/2000\n",
            "\n",
            "Epoch 1408: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0449 - accuracy: 0.9854 - mae: 0.0200 - mse: 0.0077 - val_loss: 0.9944 - val_accuracy: 0.7743 - val_mae: 0.1566 - val_mse: 0.1220 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1409/2000\n",
            "\n",
            "Epoch 1409: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0436 - accuracy: 0.9864 - mae: 0.0200 - mse: 0.0074 - val_loss: 0.9915 - val_accuracy: 0.7747 - val_mae: 0.1567 - val_mse: 0.1219 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1410/2000\n",
            "\n",
            "Epoch 1410: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0466 - accuracy: 0.9840 - mae: 0.0207 - mse: 0.0081 - val_loss: 0.9917 - val_accuracy: 0.7757 - val_mae: 0.1566 - val_mse: 0.1219 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1411/2000\n",
            "\n",
            "Epoch 1411: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0432 - accuracy: 0.9860 - mae: 0.0195 - mse: 0.0073 - val_loss: 0.9873 - val_accuracy: 0.7750 - val_mae: 0.1569 - val_mse: 0.1219 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1412/2000\n",
            "\n",
            "Epoch 1412: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0426 - accuracy: 0.9869 - mae: 0.0194 - mse: 0.0072 - val_loss: 0.9882 - val_accuracy: 0.7730 - val_mae: 0.1571 - val_mse: 0.1220 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1413/2000\n",
            "\n",
            "Epoch 1413: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0446 - accuracy: 0.9850 - mae: 0.0204 - mse: 0.0077 - val_loss: 0.9865 - val_accuracy: 0.7760 - val_mae: 0.1570 - val_mse: 0.1217 - lr: 1.0000e-04 - 311ms/epoch - 26ms/step\n",
            "Epoch 1414/2000\n",
            "\n",
            "Epoch 1414: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0419 - accuracy: 0.9864 - mae: 0.0197 - mse: 0.0071 - val_loss: 0.9909 - val_accuracy: 0.7733 - val_mae: 0.1567 - val_mse: 0.1217 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1415/2000\n",
            "\n",
            "Epoch 1415: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0439 - accuracy: 0.9861 - mae: 0.0200 - mse: 0.0076 - val_loss: 0.9885 - val_accuracy: 0.7753 - val_mae: 0.1568 - val_mse: 0.1217 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1416/2000\n",
            "\n",
            "Epoch 1416: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0425 - accuracy: 0.9867 - mae: 0.0193 - mse: 0.0072 - val_loss: 0.9913 - val_accuracy: 0.7727 - val_mae: 0.1568 - val_mse: 0.1217 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1417/2000\n",
            "\n",
            "Epoch 1417: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0444 - accuracy: 0.9853 - mae: 0.0199 - mse: 0.0076 - val_loss: 0.9914 - val_accuracy: 0.7723 - val_mae: 0.1567 - val_mse: 0.1216 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1418/2000\n",
            "\n",
            "Epoch 1418: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0478 - accuracy: 0.9838 - mae: 0.0211 - mse: 0.0083 - val_loss: 0.9929 - val_accuracy: 0.7753 - val_mae: 0.1568 - val_mse: 0.1218 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1419/2000\n",
            "\n",
            "Epoch 1419: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0449 - accuracy: 0.9857 - mae: 0.0200 - mse: 0.0076 - val_loss: 0.9934 - val_accuracy: 0.7727 - val_mae: 0.1567 - val_mse: 0.1219 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1420/2000\n",
            "\n",
            "Epoch 1420: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0441 - accuracy: 0.9854 - mae: 0.0196 - mse: 0.0076 - val_loss: 0.9936 - val_accuracy: 0.7723 - val_mae: 0.1572 - val_mse: 0.1222 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1421/2000\n",
            "\n",
            "Epoch 1421: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0394 - accuracy: 0.9880 - mae: 0.0189 - mse: 0.0067 - val_loss: 0.9928 - val_accuracy: 0.7730 - val_mae: 0.1573 - val_mse: 0.1222 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1422/2000\n",
            "\n",
            "Epoch 1422: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0476 - accuracy: 0.9838 - mae: 0.0208 - mse: 0.0083 - val_loss: 0.9913 - val_accuracy: 0.7723 - val_mae: 0.1574 - val_mse: 0.1223 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1423/2000\n",
            "\n",
            "Epoch 1423: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0437 - accuracy: 0.9860 - mae: 0.0196 - mse: 0.0076 - val_loss: 0.9923 - val_accuracy: 0.7727 - val_mae: 0.1574 - val_mse: 0.1224 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1424/2000\n",
            "\n",
            "Epoch 1424: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0436 - accuracy: 0.9866 - mae: 0.0194 - mse: 0.0073 - val_loss: 0.9940 - val_accuracy: 0.7737 - val_mae: 0.1571 - val_mse: 0.1222 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 1425/2000\n",
            "\n",
            "Epoch 1425: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0460 - accuracy: 0.9868 - mae: 0.0201 - mse: 0.0077 - val_loss: 0.9930 - val_accuracy: 0.7730 - val_mae: 0.1571 - val_mse: 0.1222 - lr: 1.0000e-04 - 311ms/epoch - 26ms/step\n",
            "Epoch 1426/2000\n",
            "\n",
            "Epoch 1426: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0441 - accuracy: 0.9857 - mae: 0.0196 - mse: 0.0075 - val_loss: 0.9951 - val_accuracy: 0.7723 - val_mae: 0.1577 - val_mse: 0.1225 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1427/2000\n",
            "\n",
            "Epoch 1427: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0435 - accuracy: 0.9856 - mae: 0.0199 - mse: 0.0075 - val_loss: 0.9867 - val_accuracy: 0.7733 - val_mae: 0.1576 - val_mse: 0.1223 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1428/2000\n",
            "\n",
            "Epoch 1428: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0393 - accuracy: 0.9877 - mae: 0.0188 - mse: 0.0066 - val_loss: 0.9904 - val_accuracy: 0.7740 - val_mae: 0.1575 - val_mse: 0.1224 - lr: 1.0000e-04 - 324ms/epoch - 27ms/step\n",
            "Epoch 1429/2000\n",
            "\n",
            "Epoch 1429: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0447 - accuracy: 0.9861 - mae: 0.0202 - mse: 0.0076 - val_loss: 0.9970 - val_accuracy: 0.7723 - val_mae: 0.1577 - val_mse: 0.1226 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1430/2000\n",
            "\n",
            "Epoch 1430: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0428 - accuracy: 0.9855 - mae: 0.0199 - mse: 0.0074 - val_loss: 0.9927 - val_accuracy: 0.7727 - val_mae: 0.1576 - val_mse: 0.1224 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1431/2000\n",
            "\n",
            "Epoch 1431: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0415 - accuracy: 0.9866 - mae: 0.0192 - mse: 0.0071 - val_loss: 0.9929 - val_accuracy: 0.7737 - val_mae: 0.1574 - val_mse: 0.1222 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1432/2000\n",
            "\n",
            "Epoch 1432: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0427 - accuracy: 0.9867 - mae: 0.0193 - mse: 0.0072 - val_loss: 1.0009 - val_accuracy: 0.7723 - val_mae: 0.1571 - val_mse: 0.1224 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1433/2000\n",
            "\n",
            "Epoch 1433: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0423 - accuracy: 0.9853 - mae: 0.0192 - mse: 0.0073 - val_loss: 1.0020 - val_accuracy: 0.7743 - val_mae: 0.1564 - val_mse: 0.1221 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1434/2000\n",
            "\n",
            "Epoch 1434: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0437 - accuracy: 0.9846 - mae: 0.0196 - mse: 0.0077 - val_loss: 1.0012 - val_accuracy: 0.7733 - val_mae: 0.1567 - val_mse: 0.1223 - lr: 1.0000e-04 - 319ms/epoch - 27ms/step\n",
            "Epoch 1435/2000\n",
            "\n",
            "Epoch 1435: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0443 - accuracy: 0.9846 - mae: 0.0199 - mse: 0.0078 - val_loss: 0.9974 - val_accuracy: 0.7747 - val_mae: 0.1566 - val_mse: 0.1219 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1436/2000\n",
            "\n",
            "Epoch 1436: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0418 - accuracy: 0.9868 - mae: 0.0193 - mse: 0.0071 - val_loss: 0.9958 - val_accuracy: 0.7730 - val_mae: 0.1569 - val_mse: 0.1221 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1437/2000\n",
            "\n",
            "Epoch 1437: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0397 - accuracy: 0.9873 - mae: 0.0187 - mse: 0.0068 - val_loss: 1.0003 - val_accuracy: 0.7727 - val_mae: 0.1569 - val_mse: 0.1222 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1438/2000\n",
            "\n",
            "Epoch 1438: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0442 - accuracy: 0.9864 - mae: 0.0195 - mse: 0.0074 - val_loss: 0.9973 - val_accuracy: 0.7737 - val_mae: 0.1570 - val_mse: 0.1222 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1439/2000\n",
            "\n",
            "Epoch 1439: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0440 - accuracy: 0.9849 - mae: 0.0197 - mse: 0.0078 - val_loss: 0.9957 - val_accuracy: 0.7727 - val_mae: 0.1574 - val_mse: 0.1222 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1440/2000\n",
            "\n",
            "Epoch 1440: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0422 - accuracy: 0.9859 - mae: 0.0194 - mse: 0.0073 - val_loss: 0.9947 - val_accuracy: 0.7743 - val_mae: 0.1575 - val_mse: 0.1223 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1441/2000\n",
            "\n",
            "Epoch 1441: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0425 - accuracy: 0.9856 - mae: 0.0190 - mse: 0.0074 - val_loss: 0.9979 - val_accuracy: 0.7737 - val_mae: 0.1572 - val_mse: 0.1223 - lr: 1.0000e-04 - 306ms/epoch - 25ms/step\n",
            "Epoch 1442/2000\n",
            "\n",
            "Epoch 1442: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0400 - accuracy: 0.9875 - mae: 0.0184 - mse: 0.0067 - val_loss: 1.0070 - val_accuracy: 0.7737 - val_mae: 0.1573 - val_mse: 0.1227 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1443/2000\n",
            "\n",
            "Epoch 1443: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0391 - accuracy: 0.9874 - mae: 0.0189 - mse: 0.0066 - val_loss: 1.0028 - val_accuracy: 0.7713 - val_mae: 0.1574 - val_mse: 0.1228 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1444/2000\n",
            "\n",
            "Epoch 1444: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0419 - accuracy: 0.9864 - mae: 0.0191 - mse: 0.0071 - val_loss: 1.0053 - val_accuracy: 0.7740 - val_mae: 0.1576 - val_mse: 0.1230 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1445/2000\n",
            "\n",
            "Epoch 1445: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0385 - accuracy: 0.9877 - mae: 0.0181 - mse: 0.0065 - val_loss: 1.0138 - val_accuracy: 0.7723 - val_mae: 0.1572 - val_mse: 0.1229 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1446/2000\n",
            "\n",
            "Epoch 1446: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0438 - accuracy: 0.9880 - mae: 0.0195 - mse: 0.0073 - val_loss: 1.0031 - val_accuracy: 0.7753 - val_mae: 0.1570 - val_mse: 0.1225 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1447/2000\n",
            "\n",
            "Epoch 1447: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0440 - accuracy: 0.9856 - mae: 0.0197 - mse: 0.0077 - val_loss: 1.0012 - val_accuracy: 0.7730 - val_mae: 0.1572 - val_mse: 0.1225 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1448/2000\n",
            "\n",
            "Epoch 1448: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0429 - accuracy: 0.9852 - mae: 0.0196 - mse: 0.0075 - val_loss: 1.0104 - val_accuracy: 0.7740 - val_mae: 0.1570 - val_mse: 0.1226 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1449/2000\n",
            "\n",
            "Epoch 1449: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0438 - accuracy: 0.9849 - mae: 0.0196 - mse: 0.0078 - val_loss: 1.0043 - val_accuracy: 0.7733 - val_mae: 0.1570 - val_mse: 0.1223 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1450/2000\n",
            "\n",
            "Epoch 1450: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0452 - accuracy: 0.9847 - mae: 0.0201 - mse: 0.0079 - val_loss: 1.0033 - val_accuracy: 0.7727 - val_mae: 0.1576 - val_mse: 0.1227 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1451/2000\n",
            "\n",
            "Epoch 1451: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0440 - accuracy: 0.9856 - mae: 0.0200 - mse: 0.0076 - val_loss: 0.9988 - val_accuracy: 0.7713 - val_mae: 0.1578 - val_mse: 0.1227 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1452/2000\n",
            "\n",
            "Epoch 1452: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0416 - accuracy: 0.9862 - mae: 0.0193 - mse: 0.0071 - val_loss: 1.0078 - val_accuracy: 0.7717 - val_mae: 0.1576 - val_mse: 0.1231 - lr: 1.0000e-04 - 279ms/epoch - 23ms/step\n",
            "Epoch 1453/2000\n",
            "\n",
            "Epoch 1453: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0421 - accuracy: 0.9862 - mae: 0.0189 - mse: 0.0072 - val_loss: 1.0029 - val_accuracy: 0.7707 - val_mae: 0.1573 - val_mse: 0.1227 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1454/2000\n",
            "\n",
            "Epoch 1454: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0434 - accuracy: 0.9860 - mae: 0.0197 - mse: 0.0075 - val_loss: 0.9978 - val_accuracy: 0.7743 - val_mae: 0.1574 - val_mse: 0.1225 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1455/2000\n",
            "\n",
            "Epoch 1455: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0390 - accuracy: 0.9877 - mae: 0.0188 - mse: 0.0067 - val_loss: 0.9963 - val_accuracy: 0.7743 - val_mae: 0.1571 - val_mse: 0.1223 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1456/2000\n",
            "\n",
            "Epoch 1456: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0417 - accuracy: 0.9864 - mae: 0.0188 - mse: 0.0072 - val_loss: 1.0132 - val_accuracy: 0.7713 - val_mae: 0.1571 - val_mse: 0.1228 - lr: 1.0000e-04 - 283ms/epoch - 24ms/step\n",
            "Epoch 1457/2000\n",
            "\n",
            "Epoch 1457: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0427 - accuracy: 0.9840 - mae: 0.0192 - mse: 0.0076 - val_loss: 1.0111 - val_accuracy: 0.7707 - val_mae: 0.1568 - val_mse: 0.1226 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1458/2000\n",
            "\n",
            "Epoch 1458: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0431 - accuracy: 0.9854 - mae: 0.0195 - mse: 0.0076 - val_loss: 1.0172 - val_accuracy: 0.7737 - val_mae: 0.1563 - val_mse: 0.1224 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1459/2000\n",
            "\n",
            "Epoch 1459: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0408 - accuracy: 0.9872 - mae: 0.0185 - mse: 0.0069 - val_loss: 1.0134 - val_accuracy: 0.7743 - val_mae: 0.1564 - val_mse: 0.1225 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1460/2000\n",
            "\n",
            "Epoch 1460: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0454 - accuracy: 0.9843 - mae: 0.0195 - mse: 0.0079 - val_loss: 1.0118 - val_accuracy: 0.7707 - val_mae: 0.1569 - val_mse: 0.1227 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1461/2000\n",
            "\n",
            "Epoch 1461: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0422 - accuracy: 0.9864 - mae: 0.0194 - mse: 0.0071 - val_loss: 1.0047 - val_accuracy: 0.7727 - val_mae: 0.1568 - val_mse: 0.1225 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1462/2000\n",
            "\n",
            "Epoch 1462: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0447 - accuracy: 0.9850 - mae: 0.0197 - mse: 0.0078 - val_loss: 1.0102 - val_accuracy: 0.7710 - val_mae: 0.1572 - val_mse: 0.1229 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1463/2000\n",
            "\n",
            "Epoch 1463: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0435 - accuracy: 0.9862 - mae: 0.0195 - mse: 0.0074 - val_loss: 1.0092 - val_accuracy: 0.7700 - val_mae: 0.1576 - val_mse: 0.1230 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1464/2000\n",
            "\n",
            "Epoch 1464: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0429 - accuracy: 0.9868 - mae: 0.0193 - mse: 0.0073 - val_loss: 1.0096 - val_accuracy: 0.7717 - val_mae: 0.1575 - val_mse: 0.1229 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1465/2000\n",
            "\n",
            "Epoch 1465: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0421 - accuracy: 0.9859 - mae: 0.0190 - mse: 0.0072 - val_loss: 1.0049 - val_accuracy: 0.7720 - val_mae: 0.1570 - val_mse: 0.1224 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1466/2000\n",
            "\n",
            "Epoch 1466: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0438 - accuracy: 0.9863 - mae: 0.0195 - mse: 0.0074 - val_loss: 1.0042 - val_accuracy: 0.7730 - val_mae: 0.1568 - val_mse: 0.1222 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1467/2000\n",
            "\n",
            "Epoch 1467: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0449 - accuracy: 0.9857 - mae: 0.0199 - mse: 0.0077 - val_loss: 1.0022 - val_accuracy: 0.7710 - val_mae: 0.1570 - val_mse: 0.1222 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1468/2000\n",
            "\n",
            "Epoch 1468: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0410 - accuracy: 0.9857 - mae: 0.0188 - mse: 0.0072 - val_loss: 1.0095 - val_accuracy: 0.7747 - val_mae: 0.1572 - val_mse: 0.1227 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1469/2000\n",
            "\n",
            "Epoch 1469: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0404 - accuracy: 0.9876 - mae: 0.0189 - mse: 0.0068 - val_loss: 1.0023 - val_accuracy: 0.7717 - val_mae: 0.1573 - val_mse: 0.1227 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1470/2000\n",
            "\n",
            "Epoch 1470: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0423 - accuracy: 0.9856 - mae: 0.0185 - mse: 0.0073 - val_loss: 1.0030 - val_accuracy: 0.7713 - val_mae: 0.1573 - val_mse: 0.1227 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1471/2000\n",
            "\n",
            "Epoch 1471: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0438 - accuracy: 0.9853 - mae: 0.0199 - mse: 0.0075 - val_loss: 1.0025 - val_accuracy: 0.7727 - val_mae: 0.1572 - val_mse: 0.1226 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1472/2000\n",
            "\n",
            "Epoch 1472: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0418 - accuracy: 0.9866 - mae: 0.0189 - mse: 0.0072 - val_loss: 1.0052 - val_accuracy: 0.7743 - val_mae: 0.1572 - val_mse: 0.1228 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1473/2000\n",
            "\n",
            "Epoch 1473: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0421 - accuracy: 0.9861 - mae: 0.0188 - mse: 0.0073 - val_loss: 1.0098 - val_accuracy: 0.7730 - val_mae: 0.1572 - val_mse: 0.1228 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1474/2000\n",
            "\n",
            "Epoch 1474: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0379 - accuracy: 0.9893 - mae: 0.0175 - mse: 0.0061 - val_loss: 1.0104 - val_accuracy: 0.7727 - val_mae: 0.1568 - val_mse: 0.1226 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1475/2000\n",
            "\n",
            "Epoch 1475: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0412 - accuracy: 0.9876 - mae: 0.0183 - mse: 0.0070 - val_loss: 1.0119 - val_accuracy: 0.7720 - val_mae: 0.1563 - val_mse: 0.1223 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1476/2000\n",
            "\n",
            "Epoch 1476: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0375 - accuracy: 0.9885 - mae: 0.0174 - mse: 0.0062 - val_loss: 1.0127 - val_accuracy: 0.7727 - val_mae: 0.1563 - val_mse: 0.1223 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1477/2000\n",
            "\n",
            "Epoch 1477: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0442 - accuracy: 0.9852 - mae: 0.0191 - mse: 0.0076 - val_loss: 1.0099 - val_accuracy: 0.7750 - val_mae: 0.1564 - val_mse: 0.1222 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1478/2000\n",
            "\n",
            "Epoch 1478: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0411 - accuracy: 0.9869 - mae: 0.0188 - mse: 0.0070 - val_loss: 1.0106 - val_accuracy: 0.7733 - val_mae: 0.1568 - val_mse: 0.1224 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1479/2000\n",
            "\n",
            "Epoch 1479: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0413 - accuracy: 0.9869 - mae: 0.0187 - mse: 0.0070 - val_loss: 1.0134 - val_accuracy: 0.7727 - val_mae: 0.1571 - val_mse: 0.1227 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1480/2000\n",
            "\n",
            "Epoch 1480: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0446 - accuracy: 0.9842 - mae: 0.0200 - mse: 0.0079 - val_loss: 1.0157 - val_accuracy: 0.7740 - val_mae: 0.1572 - val_mse: 0.1231 - lr: 1.0000e-04 - 321ms/epoch - 27ms/step\n",
            "Epoch 1481/2000\n",
            "\n",
            "Epoch 1481: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0415 - accuracy: 0.9857 - mae: 0.0189 - mse: 0.0072 - val_loss: 1.0100 - val_accuracy: 0.7747 - val_mae: 0.1573 - val_mse: 0.1228 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1482/2000\n",
            "\n",
            "Epoch 1482: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0402 - accuracy: 0.9877 - mae: 0.0183 - mse: 0.0069 - val_loss: 1.0088 - val_accuracy: 0.7740 - val_mae: 0.1571 - val_mse: 0.1225 - lr: 1.0000e-04 - 341ms/epoch - 28ms/step\n",
            "Epoch 1483/2000\n",
            "\n",
            "Epoch 1483: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0398 - accuracy: 0.9884 - mae: 0.0180 - mse: 0.0066 - val_loss: 1.0110 - val_accuracy: 0.7750 - val_mae: 0.1570 - val_mse: 0.1225 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1484/2000\n",
            "\n",
            "Epoch 1484: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0432 - accuracy: 0.9852 - mae: 0.0192 - mse: 0.0076 - val_loss: 1.0082 - val_accuracy: 0.7720 - val_mae: 0.1569 - val_mse: 0.1226 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1485/2000\n",
            "\n",
            "Epoch 1485: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0400 - accuracy: 0.9872 - mae: 0.0184 - mse: 0.0068 - val_loss: 1.0103 - val_accuracy: 0.7720 - val_mae: 0.1574 - val_mse: 0.1229 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1486/2000\n",
            "\n",
            "Epoch 1486: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0414 - accuracy: 0.9862 - mae: 0.0190 - mse: 0.0071 - val_loss: 1.0072 - val_accuracy: 0.7733 - val_mae: 0.1572 - val_mse: 0.1228 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1487/2000\n",
            "\n",
            "Epoch 1487: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0442 - accuracy: 0.9860 - mae: 0.0194 - mse: 0.0075 - val_loss: 1.0090 - val_accuracy: 0.7710 - val_mae: 0.1570 - val_mse: 0.1226 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1488/2000\n",
            "\n",
            "Epoch 1488: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0394 - accuracy: 0.9881 - mae: 0.0181 - mse: 0.0066 - val_loss: 1.0119 - val_accuracy: 0.7713 - val_mae: 0.1568 - val_mse: 0.1225 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1489/2000\n",
            "\n",
            "Epoch 1489: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0392 - accuracy: 0.9874 - mae: 0.0183 - mse: 0.0067 - val_loss: 1.0042 - val_accuracy: 0.7730 - val_mae: 0.1568 - val_mse: 0.1224 - lr: 1.0000e-04 - 311ms/epoch - 26ms/step\n",
            "Epoch 1490/2000\n",
            "\n",
            "Epoch 1490: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0425 - accuracy: 0.9857 - mae: 0.0190 - mse: 0.0074 - val_loss: 1.0151 - val_accuracy: 0.7737 - val_mae: 0.1567 - val_mse: 0.1225 - lr: 1.0000e-04 - 314ms/epoch - 26ms/step\n",
            "Epoch 1491/2000\n",
            "\n",
            "Epoch 1491: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0414 - accuracy: 0.9870 - mae: 0.0188 - mse: 0.0071 - val_loss: 1.0117 - val_accuracy: 0.7740 - val_mae: 0.1563 - val_mse: 0.1223 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 1492/2000\n",
            "\n",
            "Epoch 1492: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0419 - accuracy: 0.9864 - mae: 0.0190 - mse: 0.0072 - val_loss: 1.0168 - val_accuracy: 0.7727 - val_mae: 0.1566 - val_mse: 0.1225 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1493/2000\n",
            "\n",
            "Epoch 1493: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0397 - accuracy: 0.9880 - mae: 0.0182 - mse: 0.0067 - val_loss: 1.0134 - val_accuracy: 0.7707 - val_mae: 0.1570 - val_mse: 0.1228 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1494/2000\n",
            "\n",
            "Epoch 1494: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0416 - accuracy: 0.9874 - mae: 0.0187 - mse: 0.0071 - val_loss: 1.0070 - val_accuracy: 0.7743 - val_mae: 0.1568 - val_mse: 0.1226 - lr: 1.0000e-04 - 314ms/epoch - 26ms/step\n",
            "Epoch 1495/2000\n",
            "\n",
            "Epoch 1495: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0447 - accuracy: 0.9837 - mae: 0.0199 - mse: 0.0081 - val_loss: 1.0022 - val_accuracy: 0.7733 - val_mae: 0.1570 - val_mse: 0.1225 - lr: 1.0000e-04 - 316ms/epoch - 26ms/step\n",
            "Epoch 1496/2000\n",
            "\n",
            "Epoch 1496: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0400 - accuracy: 0.9878 - mae: 0.0186 - mse: 0.0068 - val_loss: 1.0084 - val_accuracy: 0.7723 - val_mae: 0.1571 - val_mse: 0.1227 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1497/2000\n",
            "\n",
            "Epoch 1497: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0431 - accuracy: 0.9861 - mae: 0.0187 - mse: 0.0073 - val_loss: 1.0115 - val_accuracy: 0.7717 - val_mae: 0.1574 - val_mse: 0.1229 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1498/2000\n",
            "\n",
            "Epoch 1498: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0390 - accuracy: 0.9879 - mae: 0.0187 - mse: 0.0067 - val_loss: 1.0047 - val_accuracy: 0.7707 - val_mae: 0.1572 - val_mse: 0.1227 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1499/2000\n",
            "\n",
            "Epoch 1499: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0408 - accuracy: 0.9865 - mae: 0.0185 - mse: 0.0070 - val_loss: 1.0099 - val_accuracy: 0.7740 - val_mae: 0.1570 - val_mse: 0.1228 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1500/2000\n",
            "\n",
            "Epoch 1500: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0399 - accuracy: 0.9871 - mae: 0.0183 - mse: 0.0068 - val_loss: 1.0131 - val_accuracy: 0.7717 - val_mae: 0.1570 - val_mse: 0.1228 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1501/2000\n",
            "\n",
            "Epoch 1501: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0417 - accuracy: 0.9864 - mae: 0.0188 - mse: 0.0071 - val_loss: 1.0117 - val_accuracy: 0.7737 - val_mae: 0.1570 - val_mse: 0.1227 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1502/2000\n",
            "\n",
            "Epoch 1502: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0393 - accuracy: 0.9878 - mae: 0.0177 - mse: 0.0067 - val_loss: 1.0131 - val_accuracy: 0.7720 - val_mae: 0.1575 - val_mse: 0.1231 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1503/2000\n",
            "\n",
            "Epoch 1503: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0424 - accuracy: 0.9862 - mae: 0.0191 - mse: 0.0074 - val_loss: 1.0126 - val_accuracy: 0.7710 - val_mae: 0.1575 - val_mse: 0.1232 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1504/2000\n",
            "\n",
            "Epoch 1504: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0392 - accuracy: 0.9865 - mae: 0.0181 - mse: 0.0068 - val_loss: 1.0139 - val_accuracy: 0.7723 - val_mae: 0.1571 - val_mse: 0.1230 - lr: 1.0000e-04 - 319ms/epoch - 27ms/step\n",
            "Epoch 1505/2000\n",
            "\n",
            "Epoch 1505: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0399 - accuracy: 0.9873 - mae: 0.0182 - mse: 0.0068 - val_loss: 1.0169 - val_accuracy: 0.7730 - val_mae: 0.1572 - val_mse: 0.1231 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1506/2000\n",
            "\n",
            "Epoch 1506: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0373 - accuracy: 0.9881 - mae: 0.0175 - mse: 0.0064 - val_loss: 1.0154 - val_accuracy: 0.7720 - val_mae: 0.1572 - val_mse: 0.1232 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1507/2000\n",
            "\n",
            "Epoch 1507: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0418 - accuracy: 0.9864 - mae: 0.0178 - mse: 0.0070 - val_loss: 1.0151 - val_accuracy: 0.7733 - val_mae: 0.1566 - val_mse: 0.1226 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1508/2000\n",
            "\n",
            "Epoch 1508: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0406 - accuracy: 0.9872 - mae: 0.0181 - mse: 0.0069 - val_loss: 1.0178 - val_accuracy: 0.7713 - val_mae: 0.1569 - val_mse: 0.1227 - lr: 1.0000e-04 - 317ms/epoch - 26ms/step\n",
            "Epoch 1509/2000\n",
            "\n",
            "Epoch 1509: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0398 - accuracy: 0.9875 - mae: 0.0182 - mse: 0.0069 - val_loss: 1.0088 - val_accuracy: 0.7743 - val_mae: 0.1570 - val_mse: 0.1227 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1510/2000\n",
            "\n",
            "Epoch 1510: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0431 - accuracy: 0.9879 - mae: 0.0188 - mse: 0.0071 - val_loss: 1.0129 - val_accuracy: 0.7737 - val_mae: 0.1571 - val_mse: 0.1226 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1511/2000\n",
            "\n",
            "Epoch 1511: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0416 - accuracy: 0.9865 - mae: 0.0184 - mse: 0.0071 - val_loss: 1.0115 - val_accuracy: 0.7730 - val_mae: 0.1570 - val_mse: 0.1224 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1512/2000\n",
            "\n",
            "Epoch 1512: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0427 - accuracy: 0.9852 - mae: 0.0188 - mse: 0.0074 - val_loss: 1.0148 - val_accuracy: 0.7720 - val_mae: 0.1568 - val_mse: 0.1226 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1513/2000\n",
            "\n",
            "Epoch 1513: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0381 - accuracy: 0.9879 - mae: 0.0178 - mse: 0.0065 - val_loss: 1.0125 - val_accuracy: 0.7717 - val_mae: 0.1565 - val_mse: 0.1224 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1514/2000\n",
            "\n",
            "Epoch 1514: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0367 - accuracy: 0.9879 - mae: 0.0175 - mse: 0.0064 - val_loss: 1.0193 - val_accuracy: 0.7763 - val_mae: 0.1564 - val_mse: 0.1227 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1515/2000\n",
            "\n",
            "Epoch 1515: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0366 - accuracy: 0.9891 - mae: 0.0170 - mse: 0.0062 - val_loss: 1.0272 - val_accuracy: 0.7747 - val_mae: 0.1560 - val_mse: 0.1226 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1516/2000\n",
            "\n",
            "Epoch 1516: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0402 - accuracy: 0.9877 - mae: 0.0178 - mse: 0.0069 - val_loss: 1.0227 - val_accuracy: 0.7757 - val_mae: 0.1564 - val_mse: 0.1226 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1517/2000\n",
            "\n",
            "Epoch 1517: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0407 - accuracy: 0.9865 - mae: 0.0186 - mse: 0.0071 - val_loss: 1.0237 - val_accuracy: 0.7730 - val_mae: 0.1565 - val_mse: 0.1225 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1518/2000\n",
            "\n",
            "Epoch 1518: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0383 - accuracy: 0.9880 - mae: 0.0179 - mse: 0.0066 - val_loss: 1.0217 - val_accuracy: 0.7730 - val_mae: 0.1567 - val_mse: 0.1227 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1519/2000\n",
            "\n",
            "Epoch 1519: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0415 - accuracy: 0.9861 - mae: 0.0187 - mse: 0.0071 - val_loss: 1.0260 - val_accuracy: 0.7723 - val_mae: 0.1573 - val_mse: 0.1234 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1520/2000\n",
            "\n",
            "Epoch 1520: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0402 - accuracy: 0.9873 - mae: 0.0178 - mse: 0.0068 - val_loss: 1.0243 - val_accuracy: 0.7717 - val_mae: 0.1571 - val_mse: 0.1230 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1521/2000\n",
            "\n",
            "Epoch 1521: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0398 - accuracy: 0.9872 - mae: 0.0178 - mse: 0.0067 - val_loss: 1.0258 - val_accuracy: 0.7703 - val_mae: 0.1568 - val_mse: 0.1229 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1522/2000\n",
            "\n",
            "Epoch 1522: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0402 - accuracy: 0.9870 - mae: 0.0180 - mse: 0.0068 - val_loss: 1.0236 - val_accuracy: 0.7707 - val_mae: 0.1567 - val_mse: 0.1229 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1523/2000\n",
            "\n",
            "Epoch 1523: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0398 - accuracy: 0.9876 - mae: 0.0177 - mse: 0.0068 - val_loss: 1.0237 - val_accuracy: 0.7713 - val_mae: 0.1570 - val_mse: 0.1230 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1524/2000\n",
            "\n",
            "Epoch 1524: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0403 - accuracy: 0.9869 - mae: 0.0182 - mse: 0.0069 - val_loss: 1.0218 - val_accuracy: 0.7733 - val_mae: 0.1571 - val_mse: 0.1231 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1525/2000\n",
            "\n",
            "Epoch 1525: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0391 - accuracy: 0.9877 - mae: 0.0178 - mse: 0.0068 - val_loss: 1.0234 - val_accuracy: 0.7723 - val_mae: 0.1568 - val_mse: 0.1229 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1526/2000\n",
            "\n",
            "Epoch 1526: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0390 - accuracy: 0.9876 - mae: 0.0183 - mse: 0.0067 - val_loss: 1.0250 - val_accuracy: 0.7727 - val_mae: 0.1568 - val_mse: 0.1230 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1527/2000\n",
            "\n",
            "Epoch 1527: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0410 - accuracy: 0.9865 - mae: 0.0185 - mse: 0.0071 - val_loss: 1.0223 - val_accuracy: 0.7730 - val_mae: 0.1571 - val_mse: 0.1230 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1528/2000\n",
            "\n",
            "Epoch 1528: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0410 - accuracy: 0.9866 - mae: 0.0185 - mse: 0.0072 - val_loss: 1.0251 - val_accuracy: 0.7723 - val_mae: 0.1578 - val_mse: 0.1237 - lr: 1.0000e-04 - 313ms/epoch - 26ms/step\n",
            "Epoch 1529/2000\n",
            "\n",
            "Epoch 1529: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0393 - accuracy: 0.9869 - mae: 0.0178 - mse: 0.0067 - val_loss: 1.0261 - val_accuracy: 0.7690 - val_mae: 0.1578 - val_mse: 0.1240 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1530/2000\n",
            "\n",
            "Epoch 1530: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0404 - accuracy: 0.9873 - mae: 0.0182 - mse: 0.0069 - val_loss: 1.0225 - val_accuracy: 0.7707 - val_mae: 0.1577 - val_mse: 0.1239 - lr: 1.0000e-04 - 283ms/epoch - 24ms/step\n",
            "Epoch 1531/2000\n",
            "\n",
            "Epoch 1531: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0423 - accuracy: 0.9858 - mae: 0.0181 - mse: 0.0074 - val_loss: 1.0256 - val_accuracy: 0.7700 - val_mae: 0.1577 - val_mse: 0.1239 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1532/2000\n",
            "\n",
            "Epoch 1532: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0388 - accuracy: 0.9874 - mae: 0.0178 - mse: 0.0067 - val_loss: 1.0259 - val_accuracy: 0.7710 - val_mae: 0.1574 - val_mse: 0.1235 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1533/2000\n",
            "\n",
            "Epoch 1533: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0373 - accuracy: 0.9880 - mae: 0.0172 - mse: 0.0063 - val_loss: 1.0145 - val_accuracy: 0.7733 - val_mae: 0.1569 - val_mse: 0.1229 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1534/2000\n",
            "\n",
            "Epoch 1534: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0394 - accuracy: 0.9876 - mae: 0.0178 - mse: 0.0067 - val_loss: 1.0218 - val_accuracy: 0.7730 - val_mae: 0.1565 - val_mse: 0.1228 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1535/2000\n",
            "\n",
            "Epoch 1535: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0432 - accuracy: 0.9862 - mae: 0.0190 - mse: 0.0075 - val_loss: 1.0135 - val_accuracy: 0.7717 - val_mae: 0.1565 - val_mse: 0.1225 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1536/2000\n",
            "\n",
            "Epoch 1536: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0411 - accuracy: 0.9873 - mae: 0.0184 - mse: 0.0070 - val_loss: 1.0166 - val_accuracy: 0.7717 - val_mae: 0.1564 - val_mse: 0.1225 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1537/2000\n",
            "\n",
            "Epoch 1537: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0379 - accuracy: 0.9869 - mae: 0.0177 - mse: 0.0065 - val_loss: 1.0252 - val_accuracy: 0.7723 - val_mae: 0.1562 - val_mse: 0.1227 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1538/2000\n",
            "\n",
            "Epoch 1538: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0427 - accuracy: 0.9857 - mae: 0.0189 - mse: 0.0077 - val_loss: 1.0261 - val_accuracy: 0.7733 - val_mae: 0.1560 - val_mse: 0.1225 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1539/2000\n",
            "\n",
            "Epoch 1539: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0425 - accuracy: 0.9872 - mae: 0.0185 - mse: 0.0071 - val_loss: 1.0214 - val_accuracy: 0.7733 - val_mae: 0.1562 - val_mse: 0.1224 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1540/2000\n",
            "\n",
            "Epoch 1540: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0405 - accuracy: 0.9875 - mae: 0.0184 - mse: 0.0071 - val_loss: 1.0245 - val_accuracy: 0.7737 - val_mae: 0.1567 - val_mse: 0.1229 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1541/2000\n",
            "\n",
            "Epoch 1541: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0385 - accuracy: 0.9881 - mae: 0.0172 - mse: 0.0065 - val_loss: 1.0174 - val_accuracy: 0.7730 - val_mae: 0.1568 - val_mse: 0.1229 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1542/2000\n",
            "\n",
            "Epoch 1542: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0379 - accuracy: 0.9878 - mae: 0.0173 - mse: 0.0065 - val_loss: 1.0197 - val_accuracy: 0.7757 - val_mae: 0.1567 - val_mse: 0.1229 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 1543/2000\n",
            "\n",
            "Epoch 1543: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0404 - accuracy: 0.9861 - mae: 0.0183 - mse: 0.0070 - val_loss: 1.0248 - val_accuracy: 0.7737 - val_mae: 0.1565 - val_mse: 0.1230 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1544/2000\n",
            "\n",
            "Epoch 1544: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0375 - accuracy: 0.9873 - mae: 0.0178 - mse: 0.0064 - val_loss: 1.0311 - val_accuracy: 0.7740 - val_mae: 0.1563 - val_mse: 0.1229 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1545/2000\n",
            "\n",
            "Epoch 1545: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0389 - accuracy: 0.9868 - mae: 0.0180 - mse: 0.0067 - val_loss: 1.0278 - val_accuracy: 0.7750 - val_mae: 0.1561 - val_mse: 0.1226 - lr: 1.0000e-04 - 282ms/epoch - 24ms/step\n",
            "Epoch 1546/2000\n",
            "\n",
            "Epoch 1546: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0371 - accuracy: 0.9876 - mae: 0.0175 - mse: 0.0064 - val_loss: 1.0383 - val_accuracy: 0.7767 - val_mae: 0.1557 - val_mse: 0.1226 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1547/2000\n",
            "\n",
            "Epoch 1547: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0389 - accuracy: 0.9874 - mae: 0.0175 - mse: 0.0067 - val_loss: 1.0374 - val_accuracy: 0.7743 - val_mae: 0.1557 - val_mse: 0.1226 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1548/2000\n",
            "\n",
            "Epoch 1548: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0364 - accuracy: 0.9885 - mae: 0.0169 - mse: 0.0061 - val_loss: 1.0337 - val_accuracy: 0.7757 - val_mae: 0.1558 - val_mse: 0.1227 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1549/2000\n",
            "\n",
            "Epoch 1549: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0416 - accuracy: 0.9862 - mae: 0.0179 - mse: 0.0073 - val_loss: 1.0288 - val_accuracy: 0.7750 - val_mae: 0.1560 - val_mse: 0.1225 - lr: 1.0000e-04 - 311ms/epoch - 26ms/step\n",
            "Epoch 1550/2000\n",
            "\n",
            "Epoch 1550: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0390 - accuracy: 0.9875 - mae: 0.0175 - mse: 0.0066 - val_loss: 1.0291 - val_accuracy: 0.7737 - val_mae: 0.1560 - val_mse: 0.1225 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1551/2000\n",
            "\n",
            "Epoch 1551: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0383 - accuracy: 0.9873 - mae: 0.0180 - mse: 0.0066 - val_loss: 1.0283 - val_accuracy: 0.7727 - val_mae: 0.1560 - val_mse: 0.1227 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1552/2000\n",
            "\n",
            "Epoch 1552: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0363 - accuracy: 0.9886 - mae: 0.0173 - mse: 0.0063 - val_loss: 1.0299 - val_accuracy: 0.7747 - val_mae: 0.1562 - val_mse: 0.1228 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1553/2000\n",
            "\n",
            "Epoch 1553: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0383 - accuracy: 0.9884 - mae: 0.0175 - mse: 0.0065 - val_loss: 1.0304 - val_accuracy: 0.7723 - val_mae: 0.1564 - val_mse: 0.1228 - lr: 1.0000e-04 - 314ms/epoch - 26ms/step\n",
            "Epoch 1554/2000\n",
            "\n",
            "Epoch 1554: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0382 - accuracy: 0.9875 - mae: 0.0174 - mse: 0.0066 - val_loss: 1.0333 - val_accuracy: 0.7743 - val_mae: 0.1563 - val_mse: 0.1228 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1555/2000\n",
            "\n",
            "Epoch 1555: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0377 - accuracy: 0.9883 - mae: 0.0172 - mse: 0.0065 - val_loss: 1.0276 - val_accuracy: 0.7740 - val_mae: 0.1565 - val_mse: 0.1228 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1556/2000\n",
            "\n",
            "Epoch 1556: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0398 - accuracy: 0.9878 - mae: 0.0175 - mse: 0.0068 - val_loss: 1.0247 - val_accuracy: 0.7727 - val_mae: 0.1568 - val_mse: 0.1229 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1557/2000\n",
            "\n",
            "Epoch 1557: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0387 - accuracy: 0.9878 - mae: 0.0175 - mse: 0.0065 - val_loss: 1.0246 - val_accuracy: 0.7713 - val_mae: 0.1570 - val_mse: 0.1231 - lr: 1.0000e-04 - 313ms/epoch - 26ms/step\n",
            "Epoch 1558/2000\n",
            "\n",
            "Epoch 1558: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0388 - accuracy: 0.9865 - mae: 0.0177 - mse: 0.0067 - val_loss: 1.0298 - val_accuracy: 0.7717 - val_mae: 0.1571 - val_mse: 0.1235 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1559/2000\n",
            "\n",
            "Epoch 1559: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0375 - accuracy: 0.9874 - mae: 0.0176 - mse: 0.0066 - val_loss: 1.0336 - val_accuracy: 0.7733 - val_mae: 0.1568 - val_mse: 0.1234 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1560/2000\n",
            "\n",
            "Epoch 1560: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0348 - accuracy: 0.9891 - mae: 0.0164 - mse: 0.0059 - val_loss: 1.0385 - val_accuracy: 0.7763 - val_mae: 0.1563 - val_mse: 0.1232 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1561/2000\n",
            "\n",
            "Epoch 1561: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0392 - accuracy: 0.9871 - mae: 0.0175 - mse: 0.0068 - val_loss: 1.0397 - val_accuracy: 0.7770 - val_mae: 0.1560 - val_mse: 0.1230 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1562/2000\n",
            "\n",
            "Epoch 1562: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0384 - accuracy: 0.9869 - mae: 0.0177 - mse: 0.0066 - val_loss: 1.0378 - val_accuracy: 0.7760 - val_mae: 0.1562 - val_mse: 0.1231 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1563/2000\n",
            "\n",
            "Epoch 1563: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0387 - accuracy: 0.9868 - mae: 0.0174 - mse: 0.0068 - val_loss: 1.0392 - val_accuracy: 0.7740 - val_mae: 0.1562 - val_mse: 0.1229 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1564/2000\n",
            "\n",
            "Epoch 1564: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0395 - accuracy: 0.9882 - mae: 0.0170 - mse: 0.0067 - val_loss: 1.0315 - val_accuracy: 0.7737 - val_mae: 0.1560 - val_mse: 0.1226 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1565/2000\n",
            "\n",
            "Epoch 1565: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0358 - accuracy: 0.9887 - mae: 0.0166 - mse: 0.0060 - val_loss: 1.0254 - val_accuracy: 0.7720 - val_mae: 0.1565 - val_mse: 0.1228 - lr: 1.0000e-04 - 315ms/epoch - 26ms/step\n",
            "Epoch 1566/2000\n",
            "\n",
            "Epoch 1566: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0359 - accuracy: 0.9886 - mae: 0.0172 - mse: 0.0061 - val_loss: 1.0273 - val_accuracy: 0.7730 - val_mae: 0.1562 - val_mse: 0.1227 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1567/2000\n",
            "\n",
            "Epoch 1567: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0387 - accuracy: 0.9858 - mae: 0.0173 - mse: 0.0068 - val_loss: 1.0306 - val_accuracy: 0.7737 - val_mae: 0.1562 - val_mse: 0.1227 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1568/2000\n",
            "\n",
            "Epoch 1568: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0394 - accuracy: 0.9888 - mae: 0.0171 - mse: 0.0065 - val_loss: 1.0258 - val_accuracy: 0.7750 - val_mae: 0.1563 - val_mse: 0.1227 - lr: 1.0000e-04 - 319ms/epoch - 27ms/step\n",
            "Epoch 1569/2000\n",
            "\n",
            "Epoch 1569: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0383 - accuracy: 0.9877 - mae: 0.0174 - mse: 0.0065 - val_loss: 1.0245 - val_accuracy: 0.7727 - val_mae: 0.1563 - val_mse: 0.1227 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1570/2000\n",
            "\n",
            "Epoch 1570: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0419 - accuracy: 0.9867 - mae: 0.0183 - mse: 0.0073 - val_loss: 1.0270 - val_accuracy: 0.7733 - val_mae: 0.1565 - val_mse: 0.1228 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1571/2000\n",
            "\n",
            "Epoch 1571: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0381 - accuracy: 0.9875 - mae: 0.0174 - mse: 0.0066 - val_loss: 1.0228 - val_accuracy: 0.7753 - val_mae: 0.1562 - val_mse: 0.1226 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 1572/2000\n",
            "\n",
            "Epoch 1572: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0404 - accuracy: 0.9872 - mae: 0.0176 - mse: 0.0068 - val_loss: 1.0288 - val_accuracy: 0.7720 - val_mae: 0.1567 - val_mse: 0.1230 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1573/2000\n",
            "\n",
            "Epoch 1573: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0387 - accuracy: 0.9861 - mae: 0.0180 - mse: 0.0068 - val_loss: 1.0277 - val_accuracy: 0.7723 - val_mae: 0.1565 - val_mse: 0.1228 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1574/2000\n",
            "\n",
            "Epoch 1574: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0368 - accuracy: 0.9884 - mae: 0.0172 - mse: 0.0063 - val_loss: 1.0318 - val_accuracy: 0.7737 - val_mae: 0.1560 - val_mse: 0.1226 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1575/2000\n",
            "\n",
            "Epoch 1575: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0378 - accuracy: 0.9878 - mae: 0.0172 - mse: 0.0065 - val_loss: 1.0319 - val_accuracy: 0.7723 - val_mae: 0.1558 - val_mse: 0.1227 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1576/2000\n",
            "\n",
            "Epoch 1576: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0390 - accuracy: 0.9878 - mae: 0.0175 - mse: 0.0067 - val_loss: 1.0343 - val_accuracy: 0.7733 - val_mae: 0.1558 - val_mse: 0.1226 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1577/2000\n",
            "\n",
            "Epoch 1577: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0375 - accuracy: 0.9880 - mae: 0.0164 - mse: 0.0062 - val_loss: 1.0360 - val_accuracy: 0.7747 - val_mae: 0.1562 - val_mse: 0.1229 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1578/2000\n",
            "\n",
            "Epoch 1578: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0401 - accuracy: 0.9865 - mae: 0.0175 - mse: 0.0070 - val_loss: 1.0321 - val_accuracy: 0.7733 - val_mae: 0.1564 - val_mse: 0.1228 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1579/2000\n",
            "\n",
            "Epoch 1579: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0402 - accuracy: 0.9869 - mae: 0.0179 - mse: 0.0070 - val_loss: 1.0273 - val_accuracy: 0.7723 - val_mae: 0.1567 - val_mse: 0.1230 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1580/2000\n",
            "\n",
            "Epoch 1580: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0402 - accuracy: 0.9865 - mae: 0.0182 - mse: 0.0070 - val_loss: 1.0242 - val_accuracy: 0.7743 - val_mae: 0.1566 - val_mse: 0.1228 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1581/2000\n",
            "\n",
            "Epoch 1581: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0382 - accuracy: 0.9877 - mae: 0.0171 - mse: 0.0065 - val_loss: 1.0255 - val_accuracy: 0.7743 - val_mae: 0.1565 - val_mse: 0.1228 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1582/2000\n",
            "\n",
            "Epoch 1582: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0348 - accuracy: 0.9887 - mae: 0.0168 - mse: 0.0059 - val_loss: 1.0286 - val_accuracy: 0.7733 - val_mae: 0.1560 - val_mse: 0.1225 - lr: 1.0000e-04 - 313ms/epoch - 26ms/step\n",
            "Epoch 1583/2000\n",
            "\n",
            "Epoch 1583: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0360 - accuracy: 0.9880 - mae: 0.0168 - mse: 0.0061 - val_loss: 1.0347 - val_accuracy: 0.7740 - val_mae: 0.1557 - val_mse: 0.1225 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1584/2000\n",
            "\n",
            "Epoch 1584: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0386 - accuracy: 0.9883 - mae: 0.0172 - mse: 0.0065 - val_loss: 1.0319 - val_accuracy: 0.7713 - val_mae: 0.1557 - val_mse: 0.1225 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1585/2000\n",
            "\n",
            "Epoch 1585: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0385 - accuracy: 0.9873 - mae: 0.0176 - mse: 0.0067 - val_loss: 1.0311 - val_accuracy: 0.7727 - val_mae: 0.1558 - val_mse: 0.1225 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1586/2000\n",
            "\n",
            "Epoch 1586: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0416 - accuracy: 0.9862 - mae: 0.0180 - mse: 0.0072 - val_loss: 1.0325 - val_accuracy: 0.7717 - val_mae: 0.1560 - val_mse: 0.1228 - lr: 1.0000e-04 - 317ms/epoch - 26ms/step\n",
            "Epoch 1587/2000\n",
            "\n",
            "Epoch 1587: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0402 - accuracy: 0.9862 - mae: 0.0177 - mse: 0.0070 - val_loss: 1.0328 - val_accuracy: 0.7730 - val_mae: 0.1561 - val_mse: 0.1228 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1588/2000\n",
            "\n",
            "Epoch 1588: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0331 - accuracy: 0.9892 - mae: 0.0159 - mse: 0.0055 - val_loss: 1.0297 - val_accuracy: 0.7717 - val_mae: 0.1562 - val_mse: 0.1229 - lr: 1.0000e-04 - 312ms/epoch - 26ms/step\n",
            "Epoch 1589/2000\n",
            "\n",
            "Epoch 1589: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0357 - accuracy: 0.9882 - mae: 0.0165 - mse: 0.0062 - val_loss: 1.0357 - val_accuracy: 0.7720 - val_mae: 0.1559 - val_mse: 0.1228 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1590/2000\n",
            "\n",
            "Epoch 1590: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0382 - accuracy: 0.9872 - mae: 0.0170 - mse: 0.0066 - val_loss: 1.0337 - val_accuracy: 0.7733 - val_mae: 0.1558 - val_mse: 0.1227 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1591/2000\n",
            "\n",
            "Epoch 1591: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0366 - accuracy: 0.9878 - mae: 0.0167 - mse: 0.0064 - val_loss: 1.0331 - val_accuracy: 0.7740 - val_mae: 0.1555 - val_mse: 0.1224 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1592/2000\n",
            "\n",
            "Epoch 1592: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0363 - accuracy: 0.9871 - mae: 0.0166 - mse: 0.0064 - val_loss: 1.0418 - val_accuracy: 0.7743 - val_mae: 0.1555 - val_mse: 0.1227 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1593/2000\n",
            "\n",
            "Epoch 1593: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0362 - accuracy: 0.9873 - mae: 0.0171 - mse: 0.0063 - val_loss: 1.0342 - val_accuracy: 0.7743 - val_mae: 0.1554 - val_mse: 0.1223 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1594/2000\n",
            "\n",
            "Epoch 1594: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0413 - accuracy: 0.9875 - mae: 0.0174 - mse: 0.0070 - val_loss: 1.0318 - val_accuracy: 0.7737 - val_mae: 0.1561 - val_mse: 0.1228 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1595/2000\n",
            "\n",
            "Epoch 1595: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0363 - accuracy: 0.9883 - mae: 0.0168 - mse: 0.0061 - val_loss: 1.0356 - val_accuracy: 0.7747 - val_mae: 0.1558 - val_mse: 0.1227 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1596/2000\n",
            "\n",
            "Epoch 1596: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0399 - accuracy: 0.9863 - mae: 0.0177 - mse: 0.0070 - val_loss: 1.0398 - val_accuracy: 0.7747 - val_mae: 0.1557 - val_mse: 0.1228 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1597/2000\n",
            "\n",
            "Epoch 1597: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0370 - accuracy: 0.9885 - mae: 0.0171 - mse: 0.0063 - val_loss: 1.0400 - val_accuracy: 0.7733 - val_mae: 0.1562 - val_mse: 0.1232 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1598/2000\n",
            "\n",
            "Epoch 1598: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0404 - accuracy: 0.9862 - mae: 0.0176 - mse: 0.0069 - val_loss: 1.0325 - val_accuracy: 0.7727 - val_mae: 0.1562 - val_mse: 0.1230 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1599/2000\n",
            "\n",
            "Epoch 1599: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0401 - accuracy: 0.9854 - mae: 0.0176 - mse: 0.0070 - val_loss: 1.0365 - val_accuracy: 0.7700 - val_mae: 0.1563 - val_mse: 0.1230 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1600/2000\n",
            "\n",
            "Epoch 1600: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0366 - accuracy: 0.9888 - mae: 0.0171 - mse: 0.0063 - val_loss: 1.0293 - val_accuracy: 0.7717 - val_mae: 0.1560 - val_mse: 0.1226 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1601/2000\n",
            "\n",
            "Epoch 1601: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0381 - accuracy: 0.9873 - mae: 0.0170 - mse: 0.0065 - val_loss: 1.0293 - val_accuracy: 0.7730 - val_mae: 0.1559 - val_mse: 0.1225 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1602/2000\n",
            "\n",
            "Epoch 1602: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0391 - accuracy: 0.9877 - mae: 0.0178 - mse: 0.0067 - val_loss: 1.0369 - val_accuracy: 0.7713 - val_mae: 0.1561 - val_mse: 0.1229 - lr: 1.0000e-04 - 318ms/epoch - 27ms/step\n",
            "Epoch 1603/2000\n",
            "\n",
            "Epoch 1603: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0358 - accuracy: 0.9895 - mae: 0.0167 - mse: 0.0062 - val_loss: 1.0384 - val_accuracy: 0.7713 - val_mae: 0.1560 - val_mse: 0.1229 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1604/2000\n",
            "\n",
            "Epoch 1604: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0364 - accuracy: 0.9883 - mae: 0.0166 - mse: 0.0063 - val_loss: 1.0404 - val_accuracy: 0.7740 - val_mae: 0.1558 - val_mse: 0.1227 - lr: 1.0000e-04 - 280ms/epoch - 23ms/step\n",
            "Epoch 1605/2000\n",
            "\n",
            "Epoch 1605: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0379 - accuracy: 0.9880 - mae: 0.0164 - mse: 0.0065 - val_loss: 1.0435 - val_accuracy: 0.7740 - val_mae: 0.1558 - val_mse: 0.1229 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1606/2000\n",
            "\n",
            "Epoch 1606: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0390 - accuracy: 0.9865 - mae: 0.0169 - mse: 0.0068 - val_loss: 1.0398 - val_accuracy: 0.7740 - val_mae: 0.1556 - val_mse: 0.1228 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1607/2000\n",
            "\n",
            "Epoch 1607: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0391 - accuracy: 0.9871 - mae: 0.0175 - mse: 0.0067 - val_loss: 1.0393 - val_accuracy: 0.7743 - val_mae: 0.1555 - val_mse: 0.1226 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1608/2000\n",
            "\n",
            "Epoch 1608: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0406 - accuracy: 0.9864 - mae: 0.0178 - mse: 0.0072 - val_loss: 1.0419 - val_accuracy: 0.7750 - val_mae: 0.1556 - val_mse: 0.1225 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1609/2000\n",
            "\n",
            "Epoch 1609: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0365 - accuracy: 0.9883 - mae: 0.0165 - mse: 0.0062 - val_loss: 1.0318 - val_accuracy: 0.7743 - val_mae: 0.1555 - val_mse: 0.1223 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1610/2000\n",
            "\n",
            "Epoch 1610: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0377 - accuracy: 0.9872 - mae: 0.0173 - mse: 0.0065 - val_loss: 1.0334 - val_accuracy: 0.7740 - val_mae: 0.1557 - val_mse: 0.1224 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1611/2000\n",
            "\n",
            "Epoch 1611: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0357 - accuracy: 0.9877 - mae: 0.0163 - mse: 0.0061 - val_loss: 1.0413 - val_accuracy: 0.7763 - val_mae: 0.1554 - val_mse: 0.1224 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1612/2000\n",
            "\n",
            "Epoch 1612: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0382 - accuracy: 0.9868 - mae: 0.0174 - mse: 0.0066 - val_loss: 1.0414 - val_accuracy: 0.7750 - val_mae: 0.1555 - val_mse: 0.1226 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1613/2000\n",
            "\n",
            "Epoch 1613: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0351 - accuracy: 0.9900 - mae: 0.0160 - mse: 0.0058 - val_loss: 1.0475 - val_accuracy: 0.7740 - val_mae: 0.1553 - val_mse: 0.1225 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1614/2000\n",
            "\n",
            "Epoch 1614: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0417 - accuracy: 0.9847 - mae: 0.0175 - mse: 0.0076 - val_loss: 1.0406 - val_accuracy: 0.7710 - val_mae: 0.1560 - val_mse: 0.1228 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1615/2000\n",
            "\n",
            "Epoch 1615: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0364 - accuracy: 0.9881 - mae: 0.0163 - mse: 0.0060 - val_loss: 1.0295 - val_accuracy: 0.7727 - val_mae: 0.1564 - val_mse: 0.1228 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1616/2000\n",
            "\n",
            "Epoch 1616: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0379 - accuracy: 0.9882 - mae: 0.0167 - mse: 0.0064 - val_loss: 1.0227 - val_accuracy: 0.7730 - val_mae: 0.1567 - val_mse: 0.1227 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1617/2000\n",
            "\n",
            "Epoch 1617: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0399 - accuracy: 0.9868 - mae: 0.0178 - mse: 0.0069 - val_loss: 1.0326 - val_accuracy: 0.7723 - val_mae: 0.1564 - val_mse: 0.1227 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1618/2000\n",
            "\n",
            "Epoch 1618: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0353 - accuracy: 0.9889 - mae: 0.0162 - mse: 0.0059 - val_loss: 1.0315 - val_accuracy: 0.7757 - val_mae: 0.1560 - val_mse: 0.1224 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1619/2000\n",
            "\n",
            "Epoch 1619: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0349 - accuracy: 0.9895 - mae: 0.0163 - mse: 0.0058 - val_loss: 1.0378 - val_accuracy: 0.7757 - val_mae: 0.1558 - val_mse: 0.1224 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1620/2000\n",
            "\n",
            "Epoch 1620: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0364 - accuracy: 0.9873 - mae: 0.0165 - mse: 0.0065 - val_loss: 1.0459 - val_accuracy: 0.7750 - val_mae: 0.1557 - val_mse: 0.1227 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1621/2000\n",
            "\n",
            "Epoch 1621: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0385 - accuracy: 0.9878 - mae: 0.0167 - mse: 0.0065 - val_loss: 1.0450 - val_accuracy: 0.7760 - val_mae: 0.1554 - val_mse: 0.1224 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1622/2000\n",
            "\n",
            "Epoch 1622: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0365 - accuracy: 0.9882 - mae: 0.0167 - mse: 0.0062 - val_loss: 1.0423 - val_accuracy: 0.7753 - val_mae: 0.1558 - val_mse: 0.1228 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1623/2000\n",
            "\n",
            "Epoch 1623: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0378 - accuracy: 0.9870 - mae: 0.0168 - mse: 0.0066 - val_loss: 1.0351 - val_accuracy: 0.7767 - val_mae: 0.1560 - val_mse: 0.1228 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1624/2000\n",
            "\n",
            "Epoch 1624: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0342 - accuracy: 0.9883 - mae: 0.0161 - mse: 0.0059 - val_loss: 1.0416 - val_accuracy: 0.7723 - val_mae: 0.1561 - val_mse: 0.1230 - lr: 1.0000e-04 - 317ms/epoch - 26ms/step\n",
            "Epoch 1625/2000\n",
            "\n",
            "Epoch 1625: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0388 - accuracy: 0.9868 - mae: 0.0172 - mse: 0.0067 - val_loss: 1.0394 - val_accuracy: 0.7740 - val_mae: 0.1558 - val_mse: 0.1227 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1626/2000\n",
            "\n",
            "Epoch 1626: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0353 - accuracy: 0.9885 - mae: 0.0159 - mse: 0.0059 - val_loss: 1.0437 - val_accuracy: 0.7753 - val_mae: 0.1557 - val_mse: 0.1227 - lr: 1.0000e-04 - 316ms/epoch - 26ms/step\n",
            "Epoch 1627/2000\n",
            "\n",
            "Epoch 1627: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0408 - accuracy: 0.9864 - mae: 0.0174 - mse: 0.0071 - val_loss: 1.0326 - val_accuracy: 0.7740 - val_mae: 0.1564 - val_mse: 0.1230 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1628/2000\n",
            "\n",
            "Epoch 1628: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0393 - accuracy: 0.9871 - mae: 0.0177 - mse: 0.0068 - val_loss: 1.0339 - val_accuracy: 0.7753 - val_mae: 0.1564 - val_mse: 0.1230 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1629/2000\n",
            "\n",
            "Epoch 1629: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0347 - accuracy: 0.9893 - mae: 0.0163 - mse: 0.0059 - val_loss: 1.0352 - val_accuracy: 0.7740 - val_mae: 0.1562 - val_mse: 0.1229 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1630/2000\n",
            "\n",
            "Epoch 1630: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0370 - accuracy: 0.9886 - mae: 0.0171 - mse: 0.0063 - val_loss: 1.0300 - val_accuracy: 0.7733 - val_mae: 0.1565 - val_mse: 0.1229 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1631/2000\n",
            "\n",
            "Epoch 1631: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0334 - accuracy: 0.9889 - mae: 0.0159 - mse: 0.0057 - val_loss: 1.0374 - val_accuracy: 0.7727 - val_mae: 0.1567 - val_mse: 0.1230 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1632/2000\n",
            "\n",
            "Epoch 1632: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0341 - accuracy: 0.9887 - mae: 0.0158 - mse: 0.0057 - val_loss: 1.0382 - val_accuracy: 0.7750 - val_mae: 0.1564 - val_mse: 0.1228 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1633/2000\n",
            "\n",
            "Epoch 1633: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0348 - accuracy: 0.9887 - mae: 0.0163 - mse: 0.0059 - val_loss: 1.0450 - val_accuracy: 0.7767 - val_mae: 0.1561 - val_mse: 0.1230 - lr: 1.0000e-04 - 326ms/epoch - 27ms/step\n",
            "Epoch 1634/2000\n",
            "\n",
            "Epoch 1634: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0342 - accuracy: 0.9893 - mae: 0.0161 - mse: 0.0058 - val_loss: 1.0482 - val_accuracy: 0.7770 - val_mae: 0.1556 - val_mse: 0.1228 - lr: 1.0000e-04 - 320ms/epoch - 27ms/step\n",
            "Epoch 1635/2000\n",
            "\n",
            "Epoch 1635: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0349 - accuracy: 0.9884 - mae: 0.0155 - mse: 0.0059 - val_loss: 1.0507 - val_accuracy: 0.7770 - val_mae: 0.1555 - val_mse: 0.1228 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1636/2000\n",
            "\n",
            "Epoch 1636: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0359 - accuracy: 0.9874 - mae: 0.0166 - mse: 0.0063 - val_loss: 1.0555 - val_accuracy: 0.7777 - val_mae: 0.1558 - val_mse: 0.1232 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1637/2000\n",
            "\n",
            "Epoch 1637: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0386 - accuracy: 0.9880 - mae: 0.0171 - mse: 0.0066 - val_loss: 1.0497 - val_accuracy: 0.7773 - val_mae: 0.1555 - val_mse: 0.1228 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1638/2000\n",
            "\n",
            "Epoch 1638: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0367 - accuracy: 0.9874 - mae: 0.0163 - mse: 0.0064 - val_loss: 1.0569 - val_accuracy: 0.7770 - val_mae: 0.1556 - val_mse: 0.1229 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1639/2000\n",
            "\n",
            "Epoch 1639: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0375 - accuracy: 0.9887 - mae: 0.0169 - mse: 0.0064 - val_loss: 1.0481 - val_accuracy: 0.7760 - val_mae: 0.1553 - val_mse: 0.1225 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1640/2000\n",
            "\n",
            "Epoch 1640: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0352 - accuracy: 0.9893 - mae: 0.0159 - mse: 0.0060 - val_loss: 1.0483 - val_accuracy: 0.7783 - val_mae: 0.1557 - val_mse: 0.1230 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1641/2000\n",
            "\n",
            "Epoch 1641: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0350 - accuracy: 0.9882 - mae: 0.0158 - mse: 0.0060 - val_loss: 1.0487 - val_accuracy: 0.7777 - val_mae: 0.1556 - val_mse: 0.1229 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1642/2000\n",
            "\n",
            "Epoch 1642: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0393 - accuracy: 0.9862 - mae: 0.0172 - mse: 0.0068 - val_loss: 1.0434 - val_accuracy: 0.7763 - val_mae: 0.1555 - val_mse: 0.1225 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1643/2000\n",
            "\n",
            "Epoch 1643: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0375 - accuracy: 0.9874 - mae: 0.0165 - mse: 0.0064 - val_loss: 1.0487 - val_accuracy: 0.7767 - val_mae: 0.1559 - val_mse: 0.1229 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1644/2000\n",
            "\n",
            "Epoch 1644: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0353 - accuracy: 0.9893 - mae: 0.0160 - mse: 0.0058 - val_loss: 1.0429 - val_accuracy: 0.7750 - val_mae: 0.1560 - val_mse: 0.1229 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1645/2000\n",
            "\n",
            "Epoch 1645: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0371 - accuracy: 0.9883 - mae: 0.0167 - mse: 0.0062 - val_loss: 1.0480 - val_accuracy: 0.7743 - val_mae: 0.1561 - val_mse: 0.1230 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1646/2000\n",
            "\n",
            "Epoch 1646: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0347 - accuracy: 0.9898 - mae: 0.0157 - mse: 0.0058 - val_loss: 1.0441 - val_accuracy: 0.7743 - val_mae: 0.1559 - val_mse: 0.1229 - lr: 1.0000e-04 - 283ms/epoch - 24ms/step\n",
            "Epoch 1647/2000\n",
            "\n",
            "Epoch 1647: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0371 - accuracy: 0.9872 - mae: 0.0167 - mse: 0.0065 - val_loss: 1.0467 - val_accuracy: 0.7743 - val_mae: 0.1557 - val_mse: 0.1229 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1648/2000\n",
            "\n",
            "Epoch 1648: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0360 - accuracy: 0.9887 - mae: 0.0161 - mse: 0.0061 - val_loss: 1.0458 - val_accuracy: 0.7770 - val_mae: 0.1559 - val_mse: 0.1231 - lr: 1.0000e-04 - 283ms/epoch - 24ms/step\n",
            "Epoch 1649/2000\n",
            "\n",
            "Epoch 1649: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0392 - accuracy: 0.9867 - mae: 0.0170 - mse: 0.0068 - val_loss: 1.0428 - val_accuracy: 0.7743 - val_mae: 0.1564 - val_mse: 0.1232 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1650/2000\n",
            "\n",
            "Epoch 1650: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0373 - accuracy: 0.9890 - mae: 0.0160 - mse: 0.0060 - val_loss: 1.0355 - val_accuracy: 0.7750 - val_mae: 0.1564 - val_mse: 0.1229 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1651/2000\n",
            "\n",
            "Epoch 1651: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0360 - accuracy: 0.9883 - mae: 0.0165 - mse: 0.0062 - val_loss: 1.0421 - val_accuracy: 0.7737 - val_mae: 0.1563 - val_mse: 0.1232 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1652/2000\n",
            "\n",
            "Epoch 1652: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0374 - accuracy: 0.9878 - mae: 0.0167 - mse: 0.0063 - val_loss: 1.0393 - val_accuracy: 0.7730 - val_mae: 0.1561 - val_mse: 0.1229 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1653/2000\n",
            "\n",
            "Epoch 1653: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0363 - accuracy: 0.9872 - mae: 0.0164 - mse: 0.0063 - val_loss: 1.0366 - val_accuracy: 0.7750 - val_mae: 0.1561 - val_mse: 0.1227 - lr: 1.0000e-04 - 282ms/epoch - 24ms/step\n",
            "Epoch 1654/2000\n",
            "\n",
            "Epoch 1654: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0350 - accuracy: 0.9885 - mae: 0.0163 - mse: 0.0060 - val_loss: 1.0407 - val_accuracy: 0.7737 - val_mae: 0.1564 - val_mse: 0.1233 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1655/2000\n",
            "\n",
            "Epoch 1655: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0334 - accuracy: 0.9899 - mae: 0.0156 - mse: 0.0056 - val_loss: 1.0442 - val_accuracy: 0.7733 - val_mae: 0.1563 - val_mse: 0.1233 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1656/2000\n",
            "\n",
            "Epoch 1656: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0380 - accuracy: 0.9877 - mae: 0.0171 - mse: 0.0066 - val_loss: 1.0470 - val_accuracy: 0.7760 - val_mae: 0.1560 - val_mse: 0.1230 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1657/2000\n",
            "\n",
            "Epoch 1657: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0338 - accuracy: 0.9887 - mae: 0.0154 - mse: 0.0057 - val_loss: 1.0577 - val_accuracy: 0.7730 - val_mae: 0.1561 - val_mse: 0.1234 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1658/2000\n",
            "\n",
            "Epoch 1658: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0385 - accuracy: 0.9887 - mae: 0.0164 - mse: 0.0062 - val_loss: 1.0545 - val_accuracy: 0.7733 - val_mae: 0.1560 - val_mse: 0.1233 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1659/2000\n",
            "\n",
            "Epoch 1659: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0339 - accuracy: 0.9893 - mae: 0.0155 - mse: 0.0058 - val_loss: 1.0592 - val_accuracy: 0.7737 - val_mae: 0.1555 - val_mse: 0.1232 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1660/2000\n",
            "\n",
            "Epoch 1660: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0331 - accuracy: 0.9891 - mae: 0.0153 - mse: 0.0057 - val_loss: 1.0620 - val_accuracy: 0.7737 - val_mae: 0.1552 - val_mse: 0.1231 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1661/2000\n",
            "\n",
            "Epoch 1661: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0360 - accuracy: 0.9887 - mae: 0.0160 - mse: 0.0061 - val_loss: 1.0604 - val_accuracy: 0.7747 - val_mae: 0.1556 - val_mse: 0.1232 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1662/2000\n",
            "\n",
            "Epoch 1662: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0368 - accuracy: 0.9879 - mae: 0.0162 - mse: 0.0063 - val_loss: 1.0574 - val_accuracy: 0.7757 - val_mae: 0.1559 - val_mse: 0.1234 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1663/2000\n",
            "\n",
            "Epoch 1663: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0336 - accuracy: 0.9887 - mae: 0.0154 - mse: 0.0057 - val_loss: 1.0589 - val_accuracy: 0.7763 - val_mae: 0.1561 - val_mse: 0.1237 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1664/2000\n",
            "\n",
            "Epoch 1664: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0351 - accuracy: 0.9895 - mae: 0.0156 - mse: 0.0058 - val_loss: 1.0564 - val_accuracy: 0.7753 - val_mae: 0.1557 - val_mse: 0.1233 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1665/2000\n",
            "\n",
            "Epoch 1665: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0370 - accuracy: 0.9889 - mae: 0.0163 - mse: 0.0063 - val_loss: 1.0551 - val_accuracy: 0.7743 - val_mae: 0.1556 - val_mse: 0.1232 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1666/2000\n",
            "\n",
            "Epoch 1666: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0348 - accuracy: 0.9882 - mae: 0.0158 - mse: 0.0059 - val_loss: 1.0576 - val_accuracy: 0.7727 - val_mae: 0.1557 - val_mse: 0.1232 - lr: 1.0000e-04 - 319ms/epoch - 27ms/step\n",
            "Epoch 1667/2000\n",
            "\n",
            "Epoch 1667: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0358 - accuracy: 0.9895 - mae: 0.0160 - mse: 0.0062 - val_loss: 1.0531 - val_accuracy: 0.7723 - val_mae: 0.1559 - val_mse: 0.1231 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1668/2000\n",
            "\n",
            "Epoch 1668: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0348 - accuracy: 0.9884 - mae: 0.0160 - mse: 0.0061 - val_loss: 1.0536 - val_accuracy: 0.7733 - val_mae: 0.1560 - val_mse: 0.1231 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1669/2000\n",
            "\n",
            "Epoch 1669: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0332 - accuracy: 0.9904 - mae: 0.0159 - mse: 0.0056 - val_loss: 1.0571 - val_accuracy: 0.7733 - val_mae: 0.1558 - val_mse: 0.1232 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1670/2000\n",
            "\n",
            "Epoch 1670: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0384 - accuracy: 0.9881 - mae: 0.0160 - mse: 0.0064 - val_loss: 1.0571 - val_accuracy: 0.7730 - val_mae: 0.1558 - val_mse: 0.1231 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1671/2000\n",
            "\n",
            "Epoch 1671: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0367 - accuracy: 0.9884 - mae: 0.0168 - mse: 0.0064 - val_loss: 1.0586 - val_accuracy: 0.7740 - val_mae: 0.1559 - val_mse: 0.1233 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1672/2000\n",
            "\n",
            "Epoch 1672: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0356 - accuracy: 0.9885 - mae: 0.0160 - mse: 0.0061 - val_loss: 1.0566 - val_accuracy: 0.7747 - val_mae: 0.1556 - val_mse: 0.1231 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1673/2000\n",
            "\n",
            "Epoch 1673: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0362 - accuracy: 0.9884 - mae: 0.0157 - mse: 0.0061 - val_loss: 1.0597 - val_accuracy: 0.7720 - val_mae: 0.1562 - val_mse: 0.1236 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1674/2000\n",
            "\n",
            "Epoch 1674: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0350 - accuracy: 0.9889 - mae: 0.0161 - mse: 0.0061 - val_loss: 1.0493 - val_accuracy: 0.7730 - val_mae: 0.1561 - val_mse: 0.1233 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1675/2000\n",
            "\n",
            "Epoch 1675: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0375 - accuracy: 0.9872 - mae: 0.0167 - mse: 0.0066 - val_loss: 1.0583 - val_accuracy: 0.7720 - val_mae: 0.1560 - val_mse: 0.1235 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1676/2000\n",
            "\n",
            "Epoch 1676: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0390 - accuracy: 0.9868 - mae: 0.0171 - mse: 0.0068 - val_loss: 1.0497 - val_accuracy: 0.7727 - val_mae: 0.1561 - val_mse: 0.1233 - lr: 1.0000e-04 - 306ms/epoch - 25ms/step\n",
            "Epoch 1677/2000\n",
            "\n",
            "Epoch 1677: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0361 - accuracy: 0.9883 - mae: 0.0162 - mse: 0.0062 - val_loss: 1.0477 - val_accuracy: 0.7720 - val_mae: 0.1561 - val_mse: 0.1232 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1678/2000\n",
            "\n",
            "Epoch 1678: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0361 - accuracy: 0.9885 - mae: 0.0159 - mse: 0.0060 - val_loss: 1.0580 - val_accuracy: 0.7750 - val_mae: 0.1559 - val_mse: 0.1234 - lr: 1.0000e-04 - 313ms/epoch - 26ms/step\n",
            "Epoch 1679/2000\n",
            "\n",
            "Epoch 1679: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0350 - accuracy: 0.9884 - mae: 0.0162 - mse: 0.0061 - val_loss: 1.0540 - val_accuracy: 0.7750 - val_mae: 0.1557 - val_mse: 0.1231 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1680/2000\n",
            "\n",
            "Epoch 1680: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0348 - accuracy: 0.9885 - mae: 0.0158 - mse: 0.0060 - val_loss: 1.0489 - val_accuracy: 0.7727 - val_mae: 0.1561 - val_mse: 0.1232 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1681/2000\n",
            "\n",
            "Epoch 1681: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0352 - accuracy: 0.9887 - mae: 0.0157 - mse: 0.0059 - val_loss: 1.0477 - val_accuracy: 0.7740 - val_mae: 0.1563 - val_mse: 0.1233 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1682/2000\n",
            "\n",
            "Epoch 1682: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0363 - accuracy: 0.9877 - mae: 0.0166 - mse: 0.0063 - val_loss: 1.0480 - val_accuracy: 0.7723 - val_mae: 0.1561 - val_mse: 0.1230 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1683/2000\n",
            "\n",
            "Epoch 1683: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0385 - accuracy: 0.9877 - mae: 0.0166 - mse: 0.0067 - val_loss: 1.0459 - val_accuracy: 0.7727 - val_mae: 0.1560 - val_mse: 0.1228 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1684/2000\n",
            "\n",
            "Epoch 1684: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0348 - accuracy: 0.9883 - mae: 0.0160 - mse: 0.0060 - val_loss: 1.0521 - val_accuracy: 0.7760 - val_mae: 0.1558 - val_mse: 0.1228 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1685/2000\n",
            "\n",
            "Epoch 1685: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0351 - accuracy: 0.9888 - mae: 0.0159 - mse: 0.0060 - val_loss: 1.0510 - val_accuracy: 0.7737 - val_mae: 0.1561 - val_mse: 0.1231 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1686/2000\n",
            "\n",
            "Epoch 1686: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0348 - accuracy: 0.9893 - mae: 0.0156 - mse: 0.0058 - val_loss: 1.0606 - val_accuracy: 0.7740 - val_mae: 0.1561 - val_mse: 0.1235 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1687/2000\n",
            "\n",
            "Epoch 1687: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0367 - accuracy: 0.9881 - mae: 0.0162 - mse: 0.0063 - val_loss: 1.0539 - val_accuracy: 0.7740 - val_mae: 0.1562 - val_mse: 0.1235 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1688/2000\n",
            "\n",
            "Epoch 1688: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0382 - accuracy: 0.9863 - mae: 0.0169 - mse: 0.0069 - val_loss: 1.0485 - val_accuracy: 0.7717 - val_mae: 0.1561 - val_mse: 0.1232 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1689/2000\n",
            "\n",
            "Epoch 1689: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0367 - accuracy: 0.9872 - mae: 0.0168 - mse: 0.0064 - val_loss: 1.0467 - val_accuracy: 0.7723 - val_mae: 0.1559 - val_mse: 0.1231 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1690/2000\n",
            "\n",
            "Epoch 1690: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0344 - accuracy: 0.9893 - mae: 0.0153 - mse: 0.0057 - val_loss: 1.0494 - val_accuracy: 0.7740 - val_mae: 0.1556 - val_mse: 0.1229 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1691/2000\n",
            "\n",
            "Epoch 1691: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0388 - accuracy: 0.9870 - mae: 0.0170 - mse: 0.0068 - val_loss: 1.0520 - val_accuracy: 0.7730 - val_mae: 0.1556 - val_mse: 0.1231 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1692/2000\n",
            "\n",
            "Epoch 1692: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0331 - accuracy: 0.9889 - mae: 0.0154 - mse: 0.0057 - val_loss: 1.0530 - val_accuracy: 0.7733 - val_mae: 0.1554 - val_mse: 0.1230 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1693/2000\n",
            "\n",
            "Epoch 1693: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0329 - accuracy: 0.9898 - mae: 0.0151 - mse: 0.0054 - val_loss: 1.0632 - val_accuracy: 0.7713 - val_mae: 0.1553 - val_mse: 0.1230 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1694/2000\n",
            "\n",
            "Epoch 1694: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0376 - accuracy: 0.9863 - mae: 0.0166 - mse: 0.0068 - val_loss: 1.0596 - val_accuracy: 0.7730 - val_mae: 0.1553 - val_mse: 0.1230 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1695/2000\n",
            "\n",
            "Epoch 1695: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0346 - accuracy: 0.9887 - mae: 0.0158 - mse: 0.0058 - val_loss: 1.0544 - val_accuracy: 0.7733 - val_mae: 0.1559 - val_mse: 0.1232 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1696/2000\n",
            "\n",
            "Epoch 1696: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0332 - accuracy: 0.9906 - mae: 0.0155 - mse: 0.0054 - val_loss: 1.0578 - val_accuracy: 0.7700 - val_mae: 0.1555 - val_mse: 0.1232 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1697/2000\n",
            "\n",
            "Epoch 1697: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0342 - accuracy: 0.9891 - mae: 0.0149 - mse: 0.0057 - val_loss: 1.0623 - val_accuracy: 0.7710 - val_mae: 0.1551 - val_mse: 0.1232 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1698/2000\n",
            "\n",
            "Epoch 1698: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0306 - accuracy: 0.9907 - mae: 0.0144 - mse: 0.0051 - val_loss: 1.0663 - val_accuracy: 0.7733 - val_mae: 0.1553 - val_mse: 0.1234 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1699/2000\n",
            "\n",
            "Epoch 1699: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0351 - accuracy: 0.9874 - mae: 0.0153 - mse: 0.0061 - val_loss: 1.0616 - val_accuracy: 0.7740 - val_mae: 0.1555 - val_mse: 0.1233 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1700/2000\n",
            "\n",
            "Epoch 1700: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0351 - accuracy: 0.9887 - mae: 0.0154 - mse: 0.0060 - val_loss: 1.0618 - val_accuracy: 0.7747 - val_mae: 0.1556 - val_mse: 0.1235 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1701/2000\n",
            "\n",
            "Epoch 1701: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0328 - accuracy: 0.9902 - mae: 0.0150 - mse: 0.0054 - val_loss: 1.0557 - val_accuracy: 0.7747 - val_mae: 0.1557 - val_mse: 0.1233 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1702/2000\n",
            "\n",
            "Epoch 1702: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0314 - accuracy: 0.9905 - mae: 0.0145 - mse: 0.0051 - val_loss: 1.0625 - val_accuracy: 0.7730 - val_mae: 0.1559 - val_mse: 0.1235 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1703/2000\n",
            "\n",
            "Epoch 1703: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0363 - accuracy: 0.9892 - mae: 0.0157 - mse: 0.0061 - val_loss: 1.0671 - val_accuracy: 0.7760 - val_mae: 0.1556 - val_mse: 0.1236 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1704/2000\n",
            "\n",
            "Epoch 1704: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0309 - accuracy: 0.9902 - mae: 0.0145 - mse: 0.0052 - val_loss: 1.0666 - val_accuracy: 0.7760 - val_mae: 0.1554 - val_mse: 0.1233 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1705/2000\n",
            "\n",
            "Epoch 1705: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0366 - accuracy: 0.9877 - mae: 0.0162 - mse: 0.0064 - val_loss: 1.0653 - val_accuracy: 0.7730 - val_mae: 0.1559 - val_mse: 0.1235 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1706/2000\n",
            "\n",
            "Epoch 1706: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0355 - accuracy: 0.9883 - mae: 0.0156 - mse: 0.0060 - val_loss: 1.0642 - val_accuracy: 0.7723 - val_mae: 0.1561 - val_mse: 0.1237 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1707/2000\n",
            "\n",
            "Epoch 1707: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0357 - accuracy: 0.9880 - mae: 0.0155 - mse: 0.0062 - val_loss: 1.0601 - val_accuracy: 0.7740 - val_mae: 0.1558 - val_mse: 0.1234 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1708/2000\n",
            "\n",
            "Epoch 1708: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0313 - accuracy: 0.9910 - mae: 0.0145 - mse: 0.0052 - val_loss: 1.0621 - val_accuracy: 0.7743 - val_mae: 0.1558 - val_mse: 0.1234 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1709/2000\n",
            "\n",
            "Epoch 1709: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0347 - accuracy: 0.9877 - mae: 0.0159 - mse: 0.0061 - val_loss: 1.0625 - val_accuracy: 0.7730 - val_mae: 0.1556 - val_mse: 0.1234 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1710/2000\n",
            "\n",
            "Epoch 1710: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0353 - accuracy: 0.9883 - mae: 0.0156 - mse: 0.0062 - val_loss: 1.0651 - val_accuracy: 0.7760 - val_mae: 0.1555 - val_mse: 0.1235 - lr: 1.0000e-04 - 320ms/epoch - 27ms/step\n",
            "Epoch 1711/2000\n",
            "\n",
            "Epoch 1711: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0344 - accuracy: 0.9890 - mae: 0.0152 - mse: 0.0059 - val_loss: 1.0630 - val_accuracy: 0.7760 - val_mae: 0.1552 - val_mse: 0.1231 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1712/2000\n",
            "\n",
            "Epoch 1712: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0317 - accuracy: 0.9907 - mae: 0.0140 - mse: 0.0051 - val_loss: 1.0611 - val_accuracy: 0.7730 - val_mae: 0.1552 - val_mse: 0.1229 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1713/2000\n",
            "\n",
            "Epoch 1713: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0328 - accuracy: 0.9887 - mae: 0.0155 - mse: 0.0056 - val_loss: 1.0707 - val_accuracy: 0.7737 - val_mae: 0.1553 - val_mse: 0.1231 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1714/2000\n",
            "\n",
            "Epoch 1714: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0365 - accuracy: 0.9875 - mae: 0.0162 - mse: 0.0064 - val_loss: 1.0653 - val_accuracy: 0.7737 - val_mae: 0.1552 - val_mse: 0.1229 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1715/2000\n",
            "\n",
            "Epoch 1715: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0337 - accuracy: 0.9893 - mae: 0.0150 - mse: 0.0057 - val_loss: 1.0638 - val_accuracy: 0.7713 - val_mae: 0.1558 - val_mse: 0.1233 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1716/2000\n",
            "\n",
            "Epoch 1716: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0346 - accuracy: 0.9889 - mae: 0.0153 - mse: 0.0059 - val_loss: 1.0737 - val_accuracy: 0.7720 - val_mae: 0.1563 - val_mse: 0.1240 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1717/2000\n",
            "\n",
            "Epoch 1717: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0358 - accuracy: 0.9882 - mae: 0.0158 - mse: 0.0061 - val_loss: 1.0586 - val_accuracy: 0.7727 - val_mae: 0.1559 - val_mse: 0.1236 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1718/2000\n",
            "\n",
            "Epoch 1718: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0361 - accuracy: 0.9885 - mae: 0.0160 - mse: 0.0060 - val_loss: 1.0648 - val_accuracy: 0.7730 - val_mae: 0.1561 - val_mse: 0.1236 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1719/2000\n",
            "\n",
            "Epoch 1719: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0344 - accuracy: 0.9888 - mae: 0.0157 - mse: 0.0059 - val_loss: 1.0568 - val_accuracy: 0.7740 - val_mae: 0.1558 - val_mse: 0.1232 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1720/2000\n",
            "\n",
            "Epoch 1720: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0345 - accuracy: 0.9885 - mae: 0.0155 - mse: 0.0060 - val_loss: 1.0642 - val_accuracy: 0.7720 - val_mae: 0.1556 - val_mse: 0.1232 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1721/2000\n",
            "\n",
            "Epoch 1721: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0335 - accuracy: 0.9894 - mae: 0.0155 - mse: 0.0057 - val_loss: 1.0655 - val_accuracy: 0.7727 - val_mae: 0.1554 - val_mse: 0.1231 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1722/2000\n",
            "\n",
            "Epoch 1722: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0362 - accuracy: 0.9880 - mae: 0.0157 - mse: 0.0061 - val_loss: 1.0708 - val_accuracy: 0.7733 - val_mae: 0.1558 - val_mse: 0.1234 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1723/2000\n",
            "\n",
            "Epoch 1723: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0318 - accuracy: 0.9892 - mae: 0.0150 - mse: 0.0054 - val_loss: 1.0614 - val_accuracy: 0.7737 - val_mae: 0.1560 - val_mse: 0.1234 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1724/2000\n",
            "\n",
            "Epoch 1724: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0366 - accuracy: 0.9887 - mae: 0.0157 - mse: 0.0062 - val_loss: 1.0586 - val_accuracy: 0.7717 - val_mae: 0.1561 - val_mse: 0.1234 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1725/2000\n",
            "\n",
            "Epoch 1725: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0354 - accuracy: 0.9883 - mae: 0.0157 - mse: 0.0061 - val_loss: 1.0544 - val_accuracy: 0.7750 - val_mae: 0.1559 - val_mse: 0.1233 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1726/2000\n",
            "\n",
            "Epoch 1726: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0348 - accuracy: 0.9896 - mae: 0.0157 - mse: 0.0059 - val_loss: 1.0652 - val_accuracy: 0.7710 - val_mae: 0.1555 - val_mse: 0.1233 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1727/2000\n",
            "\n",
            "Epoch 1727: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0366 - accuracy: 0.9872 - mae: 0.0162 - mse: 0.0064 - val_loss: 1.0591 - val_accuracy: 0.7727 - val_mae: 0.1556 - val_mse: 0.1231 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1728/2000\n",
            "\n",
            "Epoch 1728: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0368 - accuracy: 0.9883 - mae: 0.0159 - mse: 0.0063 - val_loss: 1.0596 - val_accuracy: 0.7720 - val_mae: 0.1561 - val_mse: 0.1233 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1729/2000\n",
            "\n",
            "Epoch 1729: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0324 - accuracy: 0.9896 - mae: 0.0150 - mse: 0.0056 - val_loss: 1.0636 - val_accuracy: 0.7737 - val_mae: 0.1560 - val_mse: 0.1234 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1730/2000\n",
            "\n",
            "Epoch 1730: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0341 - accuracy: 0.9887 - mae: 0.0156 - mse: 0.0058 - val_loss: 1.0646 - val_accuracy: 0.7737 - val_mae: 0.1560 - val_mse: 0.1237 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1731/2000\n",
            "\n",
            "Epoch 1731: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0347 - accuracy: 0.9889 - mae: 0.0156 - mse: 0.0059 - val_loss: 1.0635 - val_accuracy: 0.7713 - val_mae: 0.1556 - val_mse: 0.1235 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1732/2000\n",
            "\n",
            "Epoch 1732: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0335 - accuracy: 0.9889 - mae: 0.0153 - mse: 0.0057 - val_loss: 1.0653 - val_accuracy: 0.7737 - val_mae: 0.1555 - val_mse: 0.1235 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1733/2000\n",
            "\n",
            "Epoch 1733: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0326 - accuracy: 0.9900 - mae: 0.0145 - mse: 0.0054 - val_loss: 1.0714 - val_accuracy: 0.7737 - val_mae: 0.1553 - val_mse: 0.1234 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1734/2000\n",
            "\n",
            "Epoch 1734: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0332 - accuracy: 0.9905 - mae: 0.0149 - mse: 0.0056 - val_loss: 1.0762 - val_accuracy: 0.7730 - val_mae: 0.1548 - val_mse: 0.1231 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1735/2000\n",
            "\n",
            "Epoch 1735: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0320 - accuracy: 0.9891 - mae: 0.0147 - mse: 0.0055 - val_loss: 1.0731 - val_accuracy: 0.7740 - val_mae: 0.1549 - val_mse: 0.1231 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1736/2000\n",
            "\n",
            "Epoch 1736: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0332 - accuracy: 0.9899 - mae: 0.0152 - mse: 0.0056 - val_loss: 1.0760 - val_accuracy: 0.7763 - val_mae: 0.1553 - val_mse: 0.1233 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1737/2000\n",
            "\n",
            "Epoch 1737: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0350 - accuracy: 0.9888 - mae: 0.0152 - mse: 0.0060 - val_loss: 1.0743 - val_accuracy: 0.7737 - val_mae: 0.1553 - val_mse: 0.1233 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1738/2000\n",
            "\n",
            "Epoch 1738: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0345 - accuracy: 0.9881 - mae: 0.0154 - mse: 0.0061 - val_loss: 1.0748 - val_accuracy: 0.7730 - val_mae: 0.1558 - val_mse: 0.1236 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1739/2000\n",
            "\n",
            "Epoch 1739: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0352 - accuracy: 0.9884 - mae: 0.0155 - mse: 0.0061 - val_loss: 1.0684 - val_accuracy: 0.7763 - val_mae: 0.1555 - val_mse: 0.1234 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1740/2000\n",
            "\n",
            "Epoch 1740: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0331 - accuracy: 0.9896 - mae: 0.0150 - mse: 0.0057 - val_loss: 1.0815 - val_accuracy: 0.7747 - val_mae: 0.1557 - val_mse: 0.1238 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1741/2000\n",
            "\n",
            "Epoch 1741: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0322 - accuracy: 0.9906 - mae: 0.0145 - mse: 0.0054 - val_loss: 1.0714 - val_accuracy: 0.7757 - val_mae: 0.1556 - val_mse: 0.1235 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1742/2000\n",
            "\n",
            "Epoch 1742: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0352 - accuracy: 0.9883 - mae: 0.0156 - mse: 0.0060 - val_loss: 1.0658 - val_accuracy: 0.7737 - val_mae: 0.1562 - val_mse: 0.1237 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1743/2000\n",
            "\n",
            "Epoch 1743: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0313 - accuracy: 0.9895 - mae: 0.0146 - mse: 0.0054 - val_loss: 1.0613 - val_accuracy: 0.7750 - val_mae: 0.1561 - val_mse: 0.1236 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1744/2000\n",
            "\n",
            "Epoch 1744: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0334 - accuracy: 0.9896 - mae: 0.0153 - mse: 0.0056 - val_loss: 1.0678 - val_accuracy: 0.7753 - val_mae: 0.1559 - val_mse: 0.1237 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1745/2000\n",
            "\n",
            "Epoch 1745: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0346 - accuracy: 0.9889 - mae: 0.0151 - mse: 0.0059 - val_loss: 1.0755 - val_accuracy: 0.7770 - val_mae: 0.1555 - val_mse: 0.1234 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1746/2000\n",
            "\n",
            "Epoch 1746: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0288 - accuracy: 0.9912 - mae: 0.0135 - mse: 0.0047 - val_loss: 1.0730 - val_accuracy: 0.7767 - val_mae: 0.1552 - val_mse: 0.1232 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1747/2000\n",
            "\n",
            "Epoch 1747: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0340 - accuracy: 0.9893 - mae: 0.0146 - mse: 0.0058 - val_loss: 1.0690 - val_accuracy: 0.7740 - val_mae: 0.1555 - val_mse: 0.1232 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1748/2000\n",
            "\n",
            "Epoch 1748: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0348 - accuracy: 0.9883 - mae: 0.0158 - mse: 0.0060 - val_loss: 1.0662 - val_accuracy: 0.7750 - val_mae: 0.1559 - val_mse: 0.1233 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1749/2000\n",
            "\n",
            "Epoch 1749: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0360 - accuracy: 0.9877 - mae: 0.0162 - mse: 0.0062 - val_loss: 1.0635 - val_accuracy: 0.7753 - val_mae: 0.1557 - val_mse: 0.1232 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1750/2000\n",
            "\n",
            "Epoch 1750: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0352 - accuracy: 0.9887 - mae: 0.0155 - mse: 0.0060 - val_loss: 1.0720 - val_accuracy: 0.7753 - val_mae: 0.1558 - val_mse: 0.1235 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1751/2000\n",
            "\n",
            "Epoch 1751: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0335 - accuracy: 0.9893 - mae: 0.0151 - mse: 0.0057 - val_loss: 1.0681 - val_accuracy: 0.7730 - val_mae: 0.1555 - val_mse: 0.1232 - lr: 1.0000e-04 - 326ms/epoch - 27ms/step\n",
            "Epoch 1752/2000\n",
            "\n",
            "Epoch 1752: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0306 - accuracy: 0.9905 - mae: 0.0142 - mse: 0.0052 - val_loss: 1.0685 - val_accuracy: 0.7780 - val_mae: 0.1551 - val_mse: 0.1229 - lr: 1.0000e-04 - 319ms/epoch - 27ms/step\n",
            "Epoch 1753/2000\n",
            "\n",
            "Epoch 1753: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0317 - accuracy: 0.9902 - mae: 0.0143 - mse: 0.0054 - val_loss: 1.0788 - val_accuracy: 0.7737 - val_mae: 0.1555 - val_mse: 0.1234 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1754/2000\n",
            "\n",
            "Epoch 1754: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0330 - accuracy: 0.9890 - mae: 0.0153 - mse: 0.0057 - val_loss: 1.0669 - val_accuracy: 0.7757 - val_mae: 0.1555 - val_mse: 0.1231 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1755/2000\n",
            "\n",
            "Epoch 1755: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0354 - accuracy: 0.9877 - mae: 0.0152 - mse: 0.0061 - val_loss: 1.0682 - val_accuracy: 0.7717 - val_mae: 0.1555 - val_mse: 0.1230 - lr: 1.0000e-04 - 315ms/epoch - 26ms/step\n",
            "Epoch 1756/2000\n",
            "\n",
            "Epoch 1756: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0331 - accuracy: 0.9894 - mae: 0.0151 - mse: 0.0057 - val_loss: 1.0699 - val_accuracy: 0.7737 - val_mae: 0.1556 - val_mse: 0.1234 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1757/2000\n",
            "\n",
            "Epoch 1757: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0329 - accuracy: 0.9892 - mae: 0.0147 - mse: 0.0057 - val_loss: 1.0758 - val_accuracy: 0.7733 - val_mae: 0.1558 - val_mse: 0.1236 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1758/2000\n",
            "\n",
            "Epoch 1758: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0342 - accuracy: 0.9885 - mae: 0.0155 - mse: 0.0060 - val_loss: 1.0712 - val_accuracy: 0.7707 - val_mae: 0.1559 - val_mse: 0.1236 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1759/2000\n",
            "\n",
            "Epoch 1759: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0342 - accuracy: 0.9901 - mae: 0.0151 - mse: 0.0056 - val_loss: 1.0696 - val_accuracy: 0.7713 - val_mae: 0.1559 - val_mse: 0.1235 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1760/2000\n",
            "\n",
            "Epoch 1760: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0351 - accuracy: 0.9885 - mae: 0.0156 - mse: 0.0061 - val_loss: 1.0747 - val_accuracy: 0.7727 - val_mae: 0.1554 - val_mse: 0.1234 - lr: 1.0000e-04 - 311ms/epoch - 26ms/step\n",
            "Epoch 1761/2000\n",
            "\n",
            "Epoch 1761: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0312 - accuracy: 0.9904 - mae: 0.0139 - mse: 0.0052 - val_loss: 1.0775 - val_accuracy: 0.7753 - val_mae: 0.1551 - val_mse: 0.1234 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1762/2000\n",
            "\n",
            "Epoch 1762: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0326 - accuracy: 0.9898 - mae: 0.0146 - mse: 0.0055 - val_loss: 1.0789 - val_accuracy: 0.7737 - val_mae: 0.1551 - val_mse: 0.1234 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1763/2000\n",
            "\n",
            "Epoch 1763: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0327 - accuracy: 0.9883 - mae: 0.0150 - mse: 0.0058 - val_loss: 1.0789 - val_accuracy: 0.7720 - val_mae: 0.1554 - val_mse: 0.1234 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1764/2000\n",
            "\n",
            "Epoch 1764: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0356 - accuracy: 0.9884 - mae: 0.0157 - mse: 0.0062 - val_loss: 1.0752 - val_accuracy: 0.7740 - val_mae: 0.1554 - val_mse: 0.1232 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1765/2000\n",
            "\n",
            "Epoch 1765: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0333 - accuracy: 0.9894 - mae: 0.0146 - mse: 0.0056 - val_loss: 1.0788 - val_accuracy: 0.7737 - val_mae: 0.1554 - val_mse: 0.1232 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1766/2000\n",
            "\n",
            "Epoch 1766: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0353 - accuracy: 0.9888 - mae: 0.0154 - mse: 0.0059 - val_loss: 1.0765 - val_accuracy: 0.7727 - val_mae: 0.1557 - val_mse: 0.1234 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1767/2000\n",
            "\n",
            "Epoch 1767: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0312 - accuracy: 0.9901 - mae: 0.0144 - mse: 0.0054 - val_loss: 1.0715 - val_accuracy: 0.7730 - val_mae: 0.1557 - val_mse: 0.1233 - lr: 1.0000e-04 - 335ms/epoch - 28ms/step\n",
            "Epoch 1768/2000\n",
            "\n",
            "Epoch 1768: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0342 - accuracy: 0.9884 - mae: 0.0148 - mse: 0.0058 - val_loss: 1.0701 - val_accuracy: 0.7733 - val_mae: 0.1560 - val_mse: 0.1235 - lr: 1.0000e-04 - 315ms/epoch - 26ms/step\n",
            "Epoch 1769/2000\n",
            "\n",
            "Epoch 1769: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0318 - accuracy: 0.9897 - mae: 0.0147 - mse: 0.0054 - val_loss: 1.0657 - val_accuracy: 0.7757 - val_mae: 0.1558 - val_mse: 0.1231 - lr: 1.0000e-04 - 306ms/epoch - 25ms/step\n",
            "Epoch 1770/2000\n",
            "\n",
            "Epoch 1770: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0331 - accuracy: 0.9903 - mae: 0.0148 - mse: 0.0056 - val_loss: 1.0657 - val_accuracy: 0.7760 - val_mae: 0.1556 - val_mse: 0.1231 - lr: 1.0000e-04 - 306ms/epoch - 25ms/step\n",
            "Epoch 1771/2000\n",
            "\n",
            "Epoch 1771: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0319 - accuracy: 0.9887 - mae: 0.0146 - mse: 0.0055 - val_loss: 1.0744 - val_accuracy: 0.7723 - val_mae: 0.1556 - val_mse: 0.1235 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1772/2000\n",
            "\n",
            "Epoch 1772: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0302 - accuracy: 0.9911 - mae: 0.0138 - mse: 0.0050 - val_loss: 1.0698 - val_accuracy: 0.7723 - val_mae: 0.1553 - val_mse: 0.1234 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1773/2000\n",
            "\n",
            "Epoch 1773: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0322 - accuracy: 0.9891 - mae: 0.0143 - mse: 0.0056 - val_loss: 1.0726 - val_accuracy: 0.7740 - val_mae: 0.1554 - val_mse: 0.1237 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1774/2000\n",
            "\n",
            "Epoch 1774: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0319 - accuracy: 0.9896 - mae: 0.0144 - mse: 0.0054 - val_loss: 1.0746 - val_accuracy: 0.7713 - val_mae: 0.1555 - val_mse: 0.1238 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1775/2000\n",
            "\n",
            "Epoch 1775: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0323 - accuracy: 0.9902 - mae: 0.0146 - mse: 0.0054 - val_loss: 1.0736 - val_accuracy: 0.7737 - val_mae: 0.1555 - val_mse: 0.1237 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1776/2000\n",
            "\n",
            "Epoch 1776: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0316 - accuracy: 0.9906 - mae: 0.0145 - mse: 0.0055 - val_loss: 1.0752 - val_accuracy: 0.7740 - val_mae: 0.1554 - val_mse: 0.1237 - lr: 1.0000e-04 - 317ms/epoch - 26ms/step\n",
            "Epoch 1777/2000\n",
            "\n",
            "Epoch 1777: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0317 - accuracy: 0.9900 - mae: 0.0144 - mse: 0.0054 - val_loss: 1.0802 - val_accuracy: 0.7700 - val_mae: 0.1557 - val_mse: 0.1237 - lr: 1.0000e-04 - 315ms/epoch - 26ms/step\n",
            "Epoch 1778/2000\n",
            "\n",
            "Epoch 1778: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0346 - accuracy: 0.9887 - mae: 0.0152 - mse: 0.0060 - val_loss: 1.0757 - val_accuracy: 0.7733 - val_mae: 0.1556 - val_mse: 0.1235 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1779/2000\n",
            "\n",
            "Epoch 1779: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0334 - accuracy: 0.9890 - mae: 0.0151 - mse: 0.0058 - val_loss: 1.0817 - val_accuracy: 0.7700 - val_mae: 0.1555 - val_mse: 0.1235 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1780/2000\n",
            "\n",
            "Epoch 1780: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0310 - accuracy: 0.9904 - mae: 0.0140 - mse: 0.0052 - val_loss: 1.0697 - val_accuracy: 0.7770 - val_mae: 0.1554 - val_mse: 0.1234 - lr: 1.0000e-04 - 316ms/epoch - 26ms/step\n",
            "Epoch 1781/2000\n",
            "\n",
            "Epoch 1781: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0335 - accuracy: 0.9893 - mae: 0.0151 - mse: 0.0056 - val_loss: 1.0706 - val_accuracy: 0.7733 - val_mae: 0.1556 - val_mse: 0.1237 - lr: 1.0000e-04 - 314ms/epoch - 26ms/step\n",
            "Epoch 1782/2000\n",
            "\n",
            "Epoch 1782: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0332 - accuracy: 0.9887 - mae: 0.0151 - mse: 0.0057 - val_loss: 1.0656 - val_accuracy: 0.7720 - val_mae: 0.1561 - val_mse: 0.1241 - lr: 1.0000e-04 - 321ms/epoch - 27ms/step\n",
            "Epoch 1783/2000\n",
            "\n",
            "Epoch 1783: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0320 - accuracy: 0.9893 - mae: 0.0148 - mse: 0.0055 - val_loss: 1.0710 - val_accuracy: 0.7720 - val_mae: 0.1561 - val_mse: 0.1242 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1784/2000\n",
            "\n",
            "Epoch 1784: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0326 - accuracy: 0.9896 - mae: 0.0147 - mse: 0.0056 - val_loss: 1.0671 - val_accuracy: 0.7737 - val_mae: 0.1557 - val_mse: 0.1236 - lr: 1.0000e-04 - 319ms/epoch - 27ms/step\n",
            "Epoch 1785/2000\n",
            "\n",
            "Epoch 1785: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0319 - accuracy: 0.9901 - mae: 0.0143 - mse: 0.0055 - val_loss: 1.0768 - val_accuracy: 0.7730 - val_mae: 0.1559 - val_mse: 0.1240 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1786/2000\n",
            "\n",
            "Epoch 1786: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0290 - accuracy: 0.9918 - mae: 0.0139 - mse: 0.0047 - val_loss: 1.0739 - val_accuracy: 0.7723 - val_mae: 0.1555 - val_mse: 0.1234 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1787/2000\n",
            "\n",
            "Epoch 1787: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0324 - accuracy: 0.9900 - mae: 0.0143 - mse: 0.0054 - val_loss: 1.0753 - val_accuracy: 0.7747 - val_mae: 0.1554 - val_mse: 0.1234 - lr: 1.0000e-04 - 313ms/epoch - 26ms/step\n",
            "Epoch 1788/2000\n",
            "\n",
            "Epoch 1788: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0326 - accuracy: 0.9898 - mae: 0.0149 - mse: 0.0056 - val_loss: 1.0723 - val_accuracy: 0.7730 - val_mae: 0.1553 - val_mse: 0.1234 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1789/2000\n",
            "\n",
            "Epoch 1789: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0349 - accuracy: 0.9894 - mae: 0.0154 - mse: 0.0060 - val_loss: 1.0758 - val_accuracy: 0.7737 - val_mae: 0.1555 - val_mse: 0.1236 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1790/2000\n",
            "\n",
            "Epoch 1790: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0311 - accuracy: 0.9897 - mae: 0.0147 - mse: 0.0054 - val_loss: 1.0848 - val_accuracy: 0.7737 - val_mae: 0.1556 - val_mse: 0.1240 - lr: 1.0000e-04 - 311ms/epoch - 26ms/step\n",
            "Epoch 1791/2000\n",
            "\n",
            "Epoch 1791: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0347 - accuracy: 0.9883 - mae: 0.0152 - mse: 0.0061 - val_loss: 1.0784 - val_accuracy: 0.7757 - val_mae: 0.1553 - val_mse: 0.1236 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1792/2000\n",
            "\n",
            "Epoch 1792: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0352 - accuracy: 0.9893 - mae: 0.0153 - mse: 0.0060 - val_loss: 1.0784 - val_accuracy: 0.7747 - val_mae: 0.1560 - val_mse: 0.1240 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1793/2000\n",
            "\n",
            "Epoch 1793: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0341 - accuracy: 0.9893 - mae: 0.0152 - mse: 0.0058 - val_loss: 1.0771 - val_accuracy: 0.7727 - val_mae: 0.1557 - val_mse: 0.1237 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1794/2000\n",
            "\n",
            "Epoch 1794: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0318 - accuracy: 0.9897 - mae: 0.0147 - mse: 0.0055 - val_loss: 1.0804 - val_accuracy: 0.7780 - val_mae: 0.1551 - val_mse: 0.1232 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1795/2000\n",
            "\n",
            "Epoch 1795: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0342 - accuracy: 0.9895 - mae: 0.0149 - mse: 0.0057 - val_loss: 1.0781 - val_accuracy: 0.7750 - val_mae: 0.1554 - val_mse: 0.1234 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1796/2000\n",
            "\n",
            "Epoch 1796: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0346 - accuracy: 0.9895 - mae: 0.0149 - mse: 0.0059 - val_loss: 1.0737 - val_accuracy: 0.7737 - val_mae: 0.1558 - val_mse: 0.1238 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1797/2000\n",
            "\n",
            "Epoch 1797: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0339 - accuracy: 0.9888 - mae: 0.0147 - mse: 0.0058 - val_loss: 1.0710 - val_accuracy: 0.7737 - val_mae: 0.1556 - val_mse: 0.1234 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1798/2000\n",
            "\n",
            "Epoch 1798: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0332 - accuracy: 0.9890 - mae: 0.0151 - mse: 0.0057 - val_loss: 1.0613 - val_accuracy: 0.7750 - val_mae: 0.1557 - val_mse: 0.1233 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1799/2000\n",
            "\n",
            "Epoch 1799: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0329 - accuracy: 0.9894 - mae: 0.0151 - mse: 0.0056 - val_loss: 1.0712 - val_accuracy: 0.7747 - val_mae: 0.1557 - val_mse: 0.1234 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1800/2000\n",
            "\n",
            "Epoch 1800: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0327 - accuracy: 0.9896 - mae: 0.0147 - mse: 0.0056 - val_loss: 1.0757 - val_accuracy: 0.7777 - val_mae: 0.1553 - val_mse: 0.1233 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1801/2000\n",
            "\n",
            "Epoch 1801: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0327 - accuracy: 0.9895 - mae: 0.0146 - mse: 0.0056 - val_loss: 1.0783 - val_accuracy: 0.7753 - val_mae: 0.1554 - val_mse: 0.1236 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1802/2000\n",
            "\n",
            "Epoch 1802: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0337 - accuracy: 0.9892 - mae: 0.0151 - mse: 0.0059 - val_loss: 1.0735 - val_accuracy: 0.7757 - val_mae: 0.1558 - val_mse: 0.1241 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1803/2000\n",
            "\n",
            "Epoch 1803: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0302 - accuracy: 0.9911 - mae: 0.0137 - mse: 0.0050 - val_loss: 1.0822 - val_accuracy: 0.7757 - val_mae: 0.1561 - val_mse: 0.1243 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1804/2000\n",
            "\n",
            "Epoch 1804: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0307 - accuracy: 0.9898 - mae: 0.0140 - mse: 0.0053 - val_loss: 1.0799 - val_accuracy: 0.7740 - val_mae: 0.1558 - val_mse: 0.1240 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1805/2000\n",
            "\n",
            "Epoch 1805: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0340 - accuracy: 0.9889 - mae: 0.0149 - mse: 0.0059 - val_loss: 1.0724 - val_accuracy: 0.7747 - val_mae: 0.1559 - val_mse: 0.1237 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1806/2000\n",
            "\n",
            "Epoch 1806: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0319 - accuracy: 0.9893 - mae: 0.0145 - mse: 0.0056 - val_loss: 1.0701 - val_accuracy: 0.7710 - val_mae: 0.1562 - val_mse: 0.1238 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1807/2000\n",
            "\n",
            "Epoch 1807: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0318 - accuracy: 0.9907 - mae: 0.0141 - mse: 0.0053 - val_loss: 1.0750 - val_accuracy: 0.7713 - val_mae: 0.1561 - val_mse: 0.1240 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1808/2000\n",
            "\n",
            "Epoch 1808: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0309 - accuracy: 0.9901 - mae: 0.0144 - mse: 0.0053 - val_loss: 1.0816 - val_accuracy: 0.7703 - val_mae: 0.1559 - val_mse: 0.1243 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1809/2000\n",
            "\n",
            "Epoch 1809: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0308 - accuracy: 0.9901 - mae: 0.0139 - mse: 0.0053 - val_loss: 1.0785 - val_accuracy: 0.7730 - val_mae: 0.1558 - val_mse: 0.1240 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1810/2000\n",
            "\n",
            "Epoch 1810: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0324 - accuracy: 0.9899 - mae: 0.0141 - mse: 0.0053 - val_loss: 1.0790 - val_accuracy: 0.7707 - val_mae: 0.1560 - val_mse: 0.1242 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1811/2000\n",
            "\n",
            "Epoch 1811: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0293 - accuracy: 0.9908 - mae: 0.0137 - mse: 0.0050 - val_loss: 1.0785 - val_accuracy: 0.7727 - val_mae: 0.1560 - val_mse: 0.1243 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1812/2000\n",
            "\n",
            "Epoch 1812: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0333 - accuracy: 0.9896 - mae: 0.0147 - mse: 0.0057 - val_loss: 1.0785 - val_accuracy: 0.7703 - val_mae: 0.1568 - val_mse: 0.1248 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1813/2000\n",
            "\n",
            "Epoch 1813: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0333 - accuracy: 0.9887 - mae: 0.0148 - mse: 0.0058 - val_loss: 1.0654 - val_accuracy: 0.7717 - val_mae: 0.1569 - val_mse: 0.1244 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1814/2000\n",
            "\n",
            "Epoch 1814: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0352 - accuracy: 0.9890 - mae: 0.0153 - mse: 0.0060 - val_loss: 1.0823 - val_accuracy: 0.7703 - val_mae: 0.1569 - val_mse: 0.1250 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1815/2000\n",
            "\n",
            "Epoch 1815: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0336 - accuracy: 0.9900 - mae: 0.0144 - mse: 0.0056 - val_loss: 1.0771 - val_accuracy: 0.7743 - val_mae: 0.1564 - val_mse: 0.1244 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1816/2000\n",
            "\n",
            "Epoch 1816: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0325 - accuracy: 0.9898 - mae: 0.0144 - mse: 0.0055 - val_loss: 1.0858 - val_accuracy: 0.7737 - val_mae: 0.1563 - val_mse: 0.1246 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1817/2000\n",
            "\n",
            "Epoch 1817: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0357 - accuracy: 0.9877 - mae: 0.0154 - mse: 0.0062 - val_loss: 1.0796 - val_accuracy: 0.7720 - val_mae: 0.1558 - val_mse: 0.1242 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1818/2000\n",
            "\n",
            "Epoch 1818: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0323 - accuracy: 0.9889 - mae: 0.0144 - mse: 0.0056 - val_loss: 1.0864 - val_accuracy: 0.7723 - val_mae: 0.1557 - val_mse: 0.1243 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1819/2000\n",
            "\n",
            "Epoch 1819: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0351 - accuracy: 0.9889 - mae: 0.0152 - mse: 0.0059 - val_loss: 1.0766 - val_accuracy: 0.7703 - val_mae: 0.1556 - val_mse: 0.1237 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1820/2000\n",
            "\n",
            "Epoch 1820: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0337 - accuracy: 0.9887 - mae: 0.0143 - mse: 0.0058 - val_loss: 1.0733 - val_accuracy: 0.7743 - val_mae: 0.1550 - val_mse: 0.1231 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1821/2000\n",
            "\n",
            "Epoch 1821: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0317 - accuracy: 0.9895 - mae: 0.0146 - mse: 0.0055 - val_loss: 1.0856 - val_accuracy: 0.7733 - val_mae: 0.1554 - val_mse: 0.1239 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1822/2000\n",
            "\n",
            "Epoch 1822: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0335 - accuracy: 0.9888 - mae: 0.0147 - mse: 0.0058 - val_loss: 1.0779 - val_accuracy: 0.7747 - val_mae: 0.1551 - val_mse: 0.1234 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1823/2000\n",
            "\n",
            "Epoch 1823: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0302 - accuracy: 0.9908 - mae: 0.0141 - mse: 0.0051 - val_loss: 1.0874 - val_accuracy: 0.7757 - val_mae: 0.1552 - val_mse: 0.1236 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1824/2000\n",
            "\n",
            "Epoch 1824: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0324 - accuracy: 0.9892 - mae: 0.0144 - mse: 0.0056 - val_loss: 1.0817 - val_accuracy: 0.7777 - val_mae: 0.1548 - val_mse: 0.1233 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1825/2000\n",
            "\n",
            "Epoch 1825: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0307 - accuracy: 0.9902 - mae: 0.0137 - mse: 0.0052 - val_loss: 1.0877 - val_accuracy: 0.7773 - val_mae: 0.1549 - val_mse: 0.1232 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1826/2000\n",
            "\n",
            "Epoch 1826: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0320 - accuracy: 0.9898 - mae: 0.0146 - mse: 0.0055 - val_loss: 1.0830 - val_accuracy: 0.7770 - val_mae: 0.1550 - val_mse: 0.1233 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1827/2000\n",
            "\n",
            "Epoch 1827: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0316 - accuracy: 0.9901 - mae: 0.0140 - mse: 0.0052 - val_loss: 1.0839 - val_accuracy: 0.7753 - val_mae: 0.1552 - val_mse: 0.1236 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1828/2000\n",
            "\n",
            "Epoch 1828: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0317 - accuracy: 0.9891 - mae: 0.0141 - mse: 0.0055 - val_loss: 1.0888 - val_accuracy: 0.7750 - val_mae: 0.1555 - val_mse: 0.1240 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1829/2000\n",
            "\n",
            "Epoch 1829: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0288 - accuracy: 0.9918 - mae: 0.0131 - mse: 0.0048 - val_loss: 1.0920 - val_accuracy: 0.7767 - val_mae: 0.1552 - val_mse: 0.1240 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1830/2000\n",
            "\n",
            "Epoch 1830: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0293 - accuracy: 0.9902 - mae: 0.0139 - mse: 0.0051 - val_loss: 1.0911 - val_accuracy: 0.7787 - val_mae: 0.1547 - val_mse: 0.1235 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1831/2000\n",
            "\n",
            "Epoch 1831: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0321 - accuracy: 0.9890 - mae: 0.0143 - mse: 0.0056 - val_loss: 1.0952 - val_accuracy: 0.7757 - val_mae: 0.1549 - val_mse: 0.1238 - lr: 1.0000e-04 - 316ms/epoch - 26ms/step\n",
            "Epoch 1832/2000\n",
            "\n",
            "Epoch 1832: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0331 - accuracy: 0.9888 - mae: 0.0146 - mse: 0.0058 - val_loss: 1.0886 - val_accuracy: 0.7767 - val_mae: 0.1550 - val_mse: 0.1236 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1833/2000\n",
            "\n",
            "Epoch 1833: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0306 - accuracy: 0.9901 - mae: 0.0138 - mse: 0.0052 - val_loss: 1.0889 - val_accuracy: 0.7763 - val_mae: 0.1551 - val_mse: 0.1237 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1834/2000\n",
            "\n",
            "Epoch 1834: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0312 - accuracy: 0.9903 - mae: 0.0138 - mse: 0.0053 - val_loss: 1.0850 - val_accuracy: 0.7780 - val_mae: 0.1551 - val_mse: 0.1237 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1835/2000\n",
            "\n",
            "Epoch 1835: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0306 - accuracy: 0.9903 - mae: 0.0137 - mse: 0.0051 - val_loss: 1.0889 - val_accuracy: 0.7763 - val_mae: 0.1552 - val_mse: 0.1239 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1836/2000\n",
            "\n",
            "Epoch 1836: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0303 - accuracy: 0.9914 - mae: 0.0135 - mse: 0.0051 - val_loss: 1.0917 - val_accuracy: 0.7740 - val_mae: 0.1553 - val_mse: 0.1240 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1837/2000\n",
            "\n",
            "Epoch 1837: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0297 - accuracy: 0.9907 - mae: 0.0136 - mse: 0.0050 - val_loss: 1.0936 - val_accuracy: 0.7753 - val_mae: 0.1549 - val_mse: 0.1237 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1838/2000\n",
            "\n",
            "Epoch 1838: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0301 - accuracy: 0.9907 - mae: 0.0138 - mse: 0.0050 - val_loss: 1.0914 - val_accuracy: 0.7737 - val_mae: 0.1549 - val_mse: 0.1237 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1839/2000\n",
            "\n",
            "Epoch 1839: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0293 - accuracy: 0.9917 - mae: 0.0135 - mse: 0.0048 - val_loss: 1.0848 - val_accuracy: 0.7730 - val_mae: 0.1552 - val_mse: 0.1236 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1840/2000\n",
            "\n",
            "Epoch 1840: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0339 - accuracy: 0.9882 - mae: 0.0149 - mse: 0.0059 - val_loss: 1.0845 - val_accuracy: 0.7713 - val_mae: 0.1555 - val_mse: 0.1238 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1841/2000\n",
            "\n",
            "Epoch 1841: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0303 - accuracy: 0.9903 - mae: 0.0141 - mse: 0.0052 - val_loss: 1.0818 - val_accuracy: 0.7750 - val_mae: 0.1551 - val_mse: 0.1234 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1842/2000\n",
            "\n",
            "Epoch 1842: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0347 - accuracy: 0.9891 - mae: 0.0148 - mse: 0.0059 - val_loss: 1.0848 - val_accuracy: 0.7727 - val_mae: 0.1550 - val_mse: 0.1236 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1843/2000\n",
            "\n",
            "Epoch 1843: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0323 - accuracy: 0.9896 - mae: 0.0147 - mse: 0.0057 - val_loss: 1.0838 - val_accuracy: 0.7730 - val_mae: 0.1554 - val_mse: 0.1238 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1844/2000\n",
            "\n",
            "Epoch 1844: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0316 - accuracy: 0.9891 - mae: 0.0146 - mse: 0.0055 - val_loss: 1.0840 - val_accuracy: 0.7740 - val_mae: 0.1557 - val_mse: 0.1239 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1845/2000\n",
            "\n",
            "Epoch 1845: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0325 - accuracy: 0.9896 - mae: 0.0144 - mse: 0.0056 - val_loss: 1.0914 - val_accuracy: 0.7733 - val_mae: 0.1556 - val_mse: 0.1242 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1846/2000\n",
            "\n",
            "Epoch 1846: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0307 - accuracy: 0.9906 - mae: 0.0139 - mse: 0.0052 - val_loss: 1.0879 - val_accuracy: 0.7743 - val_mae: 0.1558 - val_mse: 0.1244 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1847/2000\n",
            "\n",
            "Epoch 1847: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0308 - accuracy: 0.9899 - mae: 0.0139 - mse: 0.0053 - val_loss: 1.0856 - val_accuracy: 0.7737 - val_mae: 0.1556 - val_mse: 0.1239 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1848/2000\n",
            "\n",
            "Epoch 1848: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0305 - accuracy: 0.9902 - mae: 0.0141 - mse: 0.0052 - val_loss: 1.0841 - val_accuracy: 0.7760 - val_mae: 0.1553 - val_mse: 0.1236 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1849/2000\n",
            "\n",
            "Epoch 1849: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0283 - accuracy: 0.9921 - mae: 0.0131 - mse: 0.0046 - val_loss: 1.0902 - val_accuracy: 0.7757 - val_mae: 0.1552 - val_mse: 0.1237 - lr: 1.0000e-04 - 319ms/epoch - 27ms/step\n",
            "Epoch 1850/2000\n",
            "\n",
            "Epoch 1850: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0312 - accuracy: 0.9900 - mae: 0.0139 - mse: 0.0053 - val_loss: 1.0915 - val_accuracy: 0.7750 - val_mae: 0.1552 - val_mse: 0.1236 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1851/2000\n",
            "\n",
            "Epoch 1851: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0308 - accuracy: 0.9891 - mae: 0.0138 - mse: 0.0053 - val_loss: 1.0920 - val_accuracy: 0.7773 - val_mae: 0.1546 - val_mse: 0.1232 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1852/2000\n",
            "\n",
            "Epoch 1852: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0351 - accuracy: 0.9877 - mae: 0.0151 - mse: 0.0063 - val_loss: 1.0932 - val_accuracy: 0.7750 - val_mae: 0.1554 - val_mse: 0.1239 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1853/2000\n",
            "\n",
            "Epoch 1853: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0308 - accuracy: 0.9901 - mae: 0.0138 - mse: 0.0053 - val_loss: 1.0814 - val_accuracy: 0.7773 - val_mae: 0.1554 - val_mse: 0.1233 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1854/2000\n",
            "\n",
            "Epoch 1854: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0346 - accuracy: 0.9891 - mae: 0.0145 - mse: 0.0059 - val_loss: 1.0863 - val_accuracy: 0.7727 - val_mae: 0.1557 - val_mse: 0.1239 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1855/2000\n",
            "\n",
            "Epoch 1855: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0294 - accuracy: 0.9894 - mae: 0.0140 - mse: 0.0052 - val_loss: 1.0878 - val_accuracy: 0.7723 - val_mae: 0.1557 - val_mse: 0.1240 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1856/2000\n",
            "\n",
            "Epoch 1856: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0298 - accuracy: 0.9901 - mae: 0.0137 - mse: 0.0050 - val_loss: 1.0893 - val_accuracy: 0.7717 - val_mae: 0.1557 - val_mse: 0.1240 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 1857/2000\n",
            "\n",
            "Epoch 1857: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0319 - accuracy: 0.9894 - mae: 0.0142 - mse: 0.0056 - val_loss: 1.0888 - val_accuracy: 0.7743 - val_mae: 0.1552 - val_mse: 0.1236 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1858/2000\n",
            "\n",
            "Epoch 1858: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0339 - accuracy: 0.9881 - mae: 0.0148 - mse: 0.0060 - val_loss: 1.0861 - val_accuracy: 0.7737 - val_mae: 0.1549 - val_mse: 0.1233 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1859/2000\n",
            "\n",
            "Epoch 1859: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0321 - accuracy: 0.9893 - mae: 0.0147 - mse: 0.0056 - val_loss: 1.0844 - val_accuracy: 0.7707 - val_mae: 0.1549 - val_mse: 0.1233 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1860/2000\n",
            "\n",
            "Epoch 1860: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0319 - accuracy: 0.9889 - mae: 0.0143 - mse: 0.0056 - val_loss: 1.0821 - val_accuracy: 0.7733 - val_mae: 0.1547 - val_mse: 0.1232 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1861/2000\n",
            "\n",
            "Epoch 1861: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0321 - accuracy: 0.9897 - mae: 0.0139 - mse: 0.0055 - val_loss: 1.0891 - val_accuracy: 0.7737 - val_mae: 0.1553 - val_mse: 0.1238 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1862/2000\n",
            "\n",
            "Epoch 1862: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0307 - accuracy: 0.9899 - mae: 0.0139 - mse: 0.0052 - val_loss: 1.0865 - val_accuracy: 0.7740 - val_mae: 0.1550 - val_mse: 0.1232 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1863/2000\n",
            "\n",
            "Epoch 1863: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0333 - accuracy: 0.9896 - mae: 0.0144 - mse: 0.0056 - val_loss: 1.0888 - val_accuracy: 0.7760 - val_mae: 0.1550 - val_mse: 0.1232 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1864/2000\n",
            "\n",
            "Epoch 1864: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0330 - accuracy: 0.9898 - mae: 0.0147 - mse: 0.0056 - val_loss: 1.0831 - val_accuracy: 0.7770 - val_mae: 0.1549 - val_mse: 0.1232 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1865/2000\n",
            "\n",
            "Epoch 1865: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0300 - accuracy: 0.9901 - mae: 0.0138 - mse: 0.0052 - val_loss: 1.0947 - val_accuracy: 0.7723 - val_mae: 0.1551 - val_mse: 0.1236 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1866/2000\n",
            "\n",
            "Epoch 1866: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0315 - accuracy: 0.9895 - mae: 0.0141 - mse: 0.0054 - val_loss: 1.0900 - val_accuracy: 0.7730 - val_mae: 0.1550 - val_mse: 0.1236 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1867/2000\n",
            "\n",
            "Epoch 1867: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0302 - accuracy: 0.9911 - mae: 0.0135 - mse: 0.0050 - val_loss: 1.0967 - val_accuracy: 0.7740 - val_mae: 0.1550 - val_mse: 0.1239 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1868/2000\n",
            "\n",
            "Epoch 1868: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0313 - accuracy: 0.9898 - mae: 0.0142 - mse: 0.0055 - val_loss: 1.0998 - val_accuracy: 0.7747 - val_mae: 0.1549 - val_mse: 0.1238 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1869/2000\n",
            "\n",
            "Epoch 1869: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0308 - accuracy: 0.9901 - mae: 0.0139 - mse: 0.0053 - val_loss: 1.0922 - val_accuracy: 0.7750 - val_mae: 0.1546 - val_mse: 0.1232 - lr: 1.0000e-04 - 310ms/epoch - 26ms/step\n",
            "Epoch 1870/2000\n",
            "\n",
            "Epoch 1870: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0304 - accuracy: 0.9902 - mae: 0.0136 - mse: 0.0051 - val_loss: 1.0995 - val_accuracy: 0.7763 - val_mae: 0.1548 - val_mse: 0.1236 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 1871/2000\n",
            "\n",
            "Epoch 1871: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0308 - accuracy: 0.9893 - mae: 0.0141 - mse: 0.0053 - val_loss: 1.0991 - val_accuracy: 0.7767 - val_mae: 0.1544 - val_mse: 0.1234 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1872/2000\n",
            "\n",
            "Epoch 1872: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0285 - accuracy: 0.9908 - mae: 0.0129 - mse: 0.0048 - val_loss: 1.1028 - val_accuracy: 0.7747 - val_mae: 0.1545 - val_mse: 0.1236 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1873/2000\n",
            "\n",
            "Epoch 1873: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0332 - accuracy: 0.9883 - mae: 0.0141 - mse: 0.0057 - val_loss: 1.1059 - val_accuracy: 0.7740 - val_mae: 0.1550 - val_mse: 0.1239 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1874/2000\n",
            "\n",
            "Epoch 1874: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0306 - accuracy: 0.9898 - mae: 0.0136 - mse: 0.0052 - val_loss: 1.0970 - val_accuracy: 0.7727 - val_mae: 0.1556 - val_mse: 0.1240 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1875/2000\n",
            "\n",
            "Epoch 1875: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0323 - accuracy: 0.9898 - mae: 0.0144 - mse: 0.0055 - val_loss: 1.0998 - val_accuracy: 0.7727 - val_mae: 0.1559 - val_mse: 0.1243 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1876/2000\n",
            "\n",
            "Epoch 1876: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0324 - accuracy: 0.9904 - mae: 0.0144 - mse: 0.0055 - val_loss: 1.0930 - val_accuracy: 0.7707 - val_mae: 0.1560 - val_mse: 0.1243 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1877/2000\n",
            "\n",
            "Epoch 1877: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0316 - accuracy: 0.9895 - mae: 0.0143 - mse: 0.0055 - val_loss: 1.0940 - val_accuracy: 0.7720 - val_mae: 0.1555 - val_mse: 0.1240 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1878/2000\n",
            "\n",
            "Epoch 1878: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0304 - accuracy: 0.9908 - mae: 0.0138 - mse: 0.0051 - val_loss: 1.0899 - val_accuracy: 0.7747 - val_mae: 0.1548 - val_mse: 0.1235 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1879/2000\n",
            "\n",
            "Epoch 1879: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0299 - accuracy: 0.9905 - mae: 0.0140 - mse: 0.0051 - val_loss: 1.0993 - val_accuracy: 0.7737 - val_mae: 0.1548 - val_mse: 0.1237 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1880/2000\n",
            "\n",
            "Epoch 1880: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0346 - accuracy: 0.9893 - mae: 0.0149 - mse: 0.0058 - val_loss: 1.0986 - val_accuracy: 0.7750 - val_mae: 0.1548 - val_mse: 0.1235 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1881/2000\n",
            "\n",
            "Epoch 1881: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0350 - accuracy: 0.9882 - mae: 0.0152 - mse: 0.0061 - val_loss: 1.0944 - val_accuracy: 0.7740 - val_mae: 0.1552 - val_mse: 0.1237 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1882/2000\n",
            "\n",
            "Epoch 1882: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0314 - accuracy: 0.9892 - mae: 0.0141 - mse: 0.0053 - val_loss: 1.0967 - val_accuracy: 0.7723 - val_mae: 0.1552 - val_mse: 0.1240 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1883/2000\n",
            "\n",
            "Epoch 1883: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0291 - accuracy: 0.9910 - mae: 0.0135 - mse: 0.0049 - val_loss: 1.1015 - val_accuracy: 0.7733 - val_mae: 0.1550 - val_mse: 0.1240 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1884/2000\n",
            "\n",
            "Epoch 1884: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0351 - accuracy: 0.9893 - mae: 0.0145 - mse: 0.0058 - val_loss: 1.0981 - val_accuracy: 0.7733 - val_mae: 0.1548 - val_mse: 0.1237 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1885/2000\n",
            "\n",
            "Epoch 1885: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0300 - accuracy: 0.9901 - mae: 0.0137 - mse: 0.0052 - val_loss: 1.0959 - val_accuracy: 0.7730 - val_mae: 0.1549 - val_mse: 0.1238 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1886/2000\n",
            "\n",
            "Epoch 1886: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0286 - accuracy: 0.9908 - mae: 0.0134 - mse: 0.0049 - val_loss: 1.0992 - val_accuracy: 0.7713 - val_mae: 0.1551 - val_mse: 0.1242 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1887/2000\n",
            "\n",
            "Epoch 1887: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0320 - accuracy: 0.9900 - mae: 0.0140 - mse: 0.0054 - val_loss: 1.0926 - val_accuracy: 0.7713 - val_mae: 0.1553 - val_mse: 0.1242 - lr: 1.0000e-04 - 315ms/epoch - 26ms/step\n",
            "Epoch 1888/2000\n",
            "\n",
            "Epoch 1888: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0336 - accuracy: 0.9898 - mae: 0.0142 - mse: 0.0058 - val_loss: 1.0884 - val_accuracy: 0.7720 - val_mae: 0.1556 - val_mse: 0.1241 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1889/2000\n",
            "\n",
            "Epoch 1889: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0312 - accuracy: 0.9899 - mae: 0.0141 - mse: 0.0053 - val_loss: 1.0868 - val_accuracy: 0.7693 - val_mae: 0.1560 - val_mse: 0.1243 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1890/2000\n",
            "\n",
            "Epoch 1890: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0330 - accuracy: 0.9896 - mae: 0.0144 - mse: 0.0057 - val_loss: 1.0872 - val_accuracy: 0.7720 - val_mae: 0.1556 - val_mse: 0.1240 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1891/2000\n",
            "\n",
            "Epoch 1891: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0306 - accuracy: 0.9909 - mae: 0.0136 - mse: 0.0050 - val_loss: 1.0935 - val_accuracy: 0.7747 - val_mae: 0.1553 - val_mse: 0.1240 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1892/2000\n",
            "\n",
            "Epoch 1892: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0308 - accuracy: 0.9894 - mae: 0.0136 - mse: 0.0054 - val_loss: 1.0930 - val_accuracy: 0.7757 - val_mae: 0.1552 - val_mse: 0.1238 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1893/2000\n",
            "\n",
            "Epoch 1893: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0324 - accuracy: 0.9896 - mae: 0.0141 - mse: 0.0056 - val_loss: 1.0882 - val_accuracy: 0.7760 - val_mae: 0.1554 - val_mse: 0.1239 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1894/2000\n",
            "\n",
            "Epoch 1894: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0307 - accuracy: 0.9903 - mae: 0.0134 - mse: 0.0052 - val_loss: 1.0846 - val_accuracy: 0.7733 - val_mae: 0.1557 - val_mse: 0.1240 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1895/2000\n",
            "\n",
            "Epoch 1895: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0317 - accuracy: 0.9894 - mae: 0.0146 - mse: 0.0053 - val_loss: 1.0862 - val_accuracy: 0.7730 - val_mae: 0.1558 - val_mse: 0.1242 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1896/2000\n",
            "\n",
            "Epoch 1896: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0286 - accuracy: 0.9913 - mae: 0.0134 - mse: 0.0047 - val_loss: 1.0859 - val_accuracy: 0.7757 - val_mae: 0.1549 - val_mse: 0.1235 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1897/2000\n",
            "\n",
            "Epoch 1897: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0301 - accuracy: 0.9909 - mae: 0.0134 - mse: 0.0051 - val_loss: 1.0981 - val_accuracy: 0.7753 - val_mae: 0.1544 - val_mse: 0.1233 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1898/2000\n",
            "\n",
            "Epoch 1898: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0287 - accuracy: 0.9915 - mae: 0.0126 - mse: 0.0048 - val_loss: 1.1066 - val_accuracy: 0.7757 - val_mae: 0.1548 - val_mse: 0.1239 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1899/2000\n",
            "\n",
            "Epoch 1899: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0297 - accuracy: 0.9898 - mae: 0.0137 - mse: 0.0052 - val_loss: 1.1051 - val_accuracy: 0.7767 - val_mae: 0.1548 - val_mse: 0.1239 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1900/2000\n",
            "\n",
            "Epoch 1900: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0319 - accuracy: 0.9890 - mae: 0.0137 - mse: 0.0056 - val_loss: 1.0995 - val_accuracy: 0.7740 - val_mae: 0.1551 - val_mse: 0.1239 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1901/2000\n",
            "\n",
            "Epoch 1901: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0290 - accuracy: 0.9909 - mae: 0.0131 - mse: 0.0049 - val_loss: 1.1006 - val_accuracy: 0.7737 - val_mae: 0.1554 - val_mse: 0.1243 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1902/2000\n",
            "\n",
            "Epoch 1902: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0306 - accuracy: 0.9902 - mae: 0.0138 - mse: 0.0052 - val_loss: 1.0954 - val_accuracy: 0.7747 - val_mae: 0.1553 - val_mse: 0.1242 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1903/2000\n",
            "\n",
            "Epoch 1903: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0280 - accuracy: 0.9911 - mae: 0.0131 - mse: 0.0049 - val_loss: 1.0935 - val_accuracy: 0.7737 - val_mae: 0.1558 - val_mse: 0.1244 - lr: 1.0000e-04 - 306ms/epoch - 25ms/step\n",
            "Epoch 1904/2000\n",
            "\n",
            "Epoch 1904: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0296 - accuracy: 0.9908 - mae: 0.0134 - mse: 0.0050 - val_loss: 1.0963 - val_accuracy: 0.7720 - val_mae: 0.1558 - val_mse: 0.1244 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1905/2000\n",
            "\n",
            "Epoch 1905: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0294 - accuracy: 0.9902 - mae: 0.0135 - mse: 0.0050 - val_loss: 1.1031 - val_accuracy: 0.7723 - val_mae: 0.1559 - val_mse: 0.1245 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1906/2000\n",
            "\n",
            "Epoch 1906: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0342 - accuracy: 0.9885 - mae: 0.0150 - mse: 0.0060 - val_loss: 1.1038 - val_accuracy: 0.7713 - val_mae: 0.1566 - val_mse: 0.1251 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1907/2000\n",
            "\n",
            "Epoch 1907: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0324 - accuracy: 0.9893 - mae: 0.0144 - mse: 0.0057 - val_loss: 1.0936 - val_accuracy: 0.7710 - val_mae: 0.1561 - val_mse: 0.1245 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1908/2000\n",
            "\n",
            "Epoch 1908: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0300 - accuracy: 0.9907 - mae: 0.0141 - mse: 0.0051 - val_loss: 1.1014 - val_accuracy: 0.7700 - val_mae: 0.1558 - val_mse: 0.1244 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1909/2000\n",
            "\n",
            "Epoch 1909: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0301 - accuracy: 0.9898 - mae: 0.0137 - mse: 0.0052 - val_loss: 1.1058 - val_accuracy: 0.7740 - val_mae: 0.1549 - val_mse: 0.1238 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1910/2000\n",
            "\n",
            "Epoch 1910: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0316 - accuracy: 0.9901 - mae: 0.0140 - mse: 0.0054 - val_loss: 1.1066 - val_accuracy: 0.7737 - val_mae: 0.1549 - val_mse: 0.1241 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1911/2000\n",
            "\n",
            "Epoch 1911: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0317 - accuracy: 0.9898 - mae: 0.0136 - mse: 0.0054 - val_loss: 1.1028 - val_accuracy: 0.7737 - val_mae: 0.1552 - val_mse: 0.1242 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1912/2000\n",
            "\n",
            "Epoch 1912: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0318 - accuracy: 0.9893 - mae: 0.0145 - mse: 0.0056 - val_loss: 1.0983 - val_accuracy: 0.7723 - val_mae: 0.1555 - val_mse: 0.1240 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1913/2000\n",
            "\n",
            "Epoch 1913: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0285 - accuracy: 0.9903 - mae: 0.0134 - mse: 0.0049 - val_loss: 1.1006 - val_accuracy: 0.7683 - val_mae: 0.1556 - val_mse: 0.1241 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1914/2000\n",
            "\n",
            "Epoch 1914: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0295 - accuracy: 0.9901 - mae: 0.0135 - mse: 0.0050 - val_loss: 1.1048 - val_accuracy: 0.7713 - val_mae: 0.1560 - val_mse: 0.1246 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1915/2000\n",
            "\n",
            "Epoch 1915: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0299 - accuracy: 0.9908 - mae: 0.0137 - mse: 0.0052 - val_loss: 1.1016 - val_accuracy: 0.7733 - val_mae: 0.1553 - val_mse: 0.1239 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1916/2000\n",
            "\n",
            "Epoch 1916: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0287 - accuracy: 0.9908 - mae: 0.0133 - mse: 0.0050 - val_loss: 1.1013 - val_accuracy: 0.7730 - val_mae: 0.1551 - val_mse: 0.1239 - lr: 1.0000e-04 - 283ms/epoch - 24ms/step\n",
            "Epoch 1917/2000\n",
            "\n",
            "Epoch 1917: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0314 - accuracy: 0.9898 - mae: 0.0136 - mse: 0.0055 - val_loss: 1.1025 - val_accuracy: 0.7737 - val_mae: 0.1546 - val_mse: 0.1237 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1918/2000\n",
            "\n",
            "Epoch 1918: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0332 - accuracy: 0.9893 - mae: 0.0141 - mse: 0.0057 - val_loss: 1.1043 - val_accuracy: 0.7727 - val_mae: 0.1547 - val_mse: 0.1237 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1919/2000\n",
            "\n",
            "Epoch 1919: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0298 - accuracy: 0.9904 - mae: 0.0133 - mse: 0.0051 - val_loss: 1.0972 - val_accuracy: 0.7777 - val_mae: 0.1544 - val_mse: 0.1231 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1920/2000\n",
            "\n",
            "Epoch 1920: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0309 - accuracy: 0.9907 - mae: 0.0141 - mse: 0.0053 - val_loss: 1.1016 - val_accuracy: 0.7760 - val_mae: 0.1545 - val_mse: 0.1233 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1921/2000\n",
            "\n",
            "Epoch 1921: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0284 - accuracy: 0.9914 - mae: 0.0131 - mse: 0.0048 - val_loss: 1.0986 - val_accuracy: 0.7757 - val_mae: 0.1541 - val_mse: 0.1231 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1922/2000\n",
            "\n",
            "Epoch 1922: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0319 - accuracy: 0.9902 - mae: 0.0134 - mse: 0.0054 - val_loss: 1.1069 - val_accuracy: 0.7737 - val_mae: 0.1549 - val_mse: 0.1238 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1923/2000\n",
            "\n",
            "Epoch 1923: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0309 - accuracy: 0.9903 - mae: 0.0136 - mse: 0.0053 - val_loss: 1.0993 - val_accuracy: 0.7713 - val_mae: 0.1553 - val_mse: 0.1240 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1924/2000\n",
            "\n",
            "Epoch 1924: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0334 - accuracy: 0.9891 - mae: 0.0142 - mse: 0.0056 - val_loss: 1.0922 - val_accuracy: 0.7727 - val_mae: 0.1553 - val_mse: 0.1238 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1925/2000\n",
            "\n",
            "Epoch 1925: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0284 - accuracy: 0.9910 - mae: 0.0133 - mse: 0.0048 - val_loss: 1.0924 - val_accuracy: 0.7740 - val_mae: 0.1552 - val_mse: 0.1238 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1926/2000\n",
            "\n",
            "Epoch 1926: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0298 - accuracy: 0.9898 - mae: 0.0134 - mse: 0.0051 - val_loss: 1.1003 - val_accuracy: 0.7733 - val_mae: 0.1554 - val_mse: 0.1242 - lr: 1.0000e-04 - 297ms/epoch - 25ms/step\n",
            "Epoch 1927/2000\n",
            "\n",
            "Epoch 1927: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0308 - accuracy: 0.9902 - mae: 0.0138 - mse: 0.0052 - val_loss: 1.1010 - val_accuracy: 0.7743 - val_mae: 0.1550 - val_mse: 0.1239 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1928/2000\n",
            "\n",
            "Epoch 1928: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0298 - accuracy: 0.9902 - mae: 0.0132 - mse: 0.0051 - val_loss: 1.1080 - val_accuracy: 0.7720 - val_mae: 0.1555 - val_mse: 0.1246 - lr: 1.0000e-04 - 311ms/epoch - 26ms/step\n",
            "Epoch 1929/2000\n",
            "\n",
            "Epoch 1929: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0302 - accuracy: 0.9900 - mae: 0.0138 - mse: 0.0053 - val_loss: 1.1119 - val_accuracy: 0.7737 - val_mae: 0.1549 - val_mse: 0.1241 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1930/2000\n",
            "\n",
            "Epoch 1930: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0300 - accuracy: 0.9898 - mae: 0.0139 - mse: 0.0052 - val_loss: 1.1022 - val_accuracy: 0.7757 - val_mae: 0.1545 - val_mse: 0.1235 - lr: 1.0000e-04 - 300ms/epoch - 25ms/step\n",
            "Epoch 1931/2000\n",
            "\n",
            "Epoch 1931: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0284 - accuracy: 0.9911 - mae: 0.0129 - mse: 0.0048 - val_loss: 1.1063 - val_accuracy: 0.7747 - val_mae: 0.1549 - val_mse: 0.1241 - lr: 1.0000e-04 - 302ms/epoch - 25ms/step\n",
            "Epoch 1932/2000\n",
            "\n",
            "Epoch 1932: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0277 - accuracy: 0.9917 - mae: 0.0127 - mse: 0.0046 - val_loss: 1.1030 - val_accuracy: 0.7753 - val_mae: 0.1548 - val_mse: 0.1238 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1933/2000\n",
            "\n",
            "Epoch 1933: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0316 - accuracy: 0.9893 - mae: 0.0143 - mse: 0.0056 - val_loss: 1.1037 - val_accuracy: 0.7760 - val_mae: 0.1548 - val_mse: 0.1239 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1934/2000\n",
            "\n",
            "Epoch 1934: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0306 - accuracy: 0.9897 - mae: 0.0131 - mse: 0.0052 - val_loss: 1.0990 - val_accuracy: 0.7743 - val_mae: 0.1547 - val_mse: 0.1239 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1935/2000\n",
            "\n",
            "Epoch 1935: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0315 - accuracy: 0.9893 - mae: 0.0144 - mse: 0.0055 - val_loss: 1.0972 - val_accuracy: 0.7723 - val_mae: 0.1551 - val_mse: 0.1241 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1936/2000\n",
            "\n",
            "Epoch 1936: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0301 - accuracy: 0.9893 - mae: 0.0135 - mse: 0.0053 - val_loss: 1.0989 - val_accuracy: 0.7743 - val_mae: 0.1551 - val_mse: 0.1240 - lr: 1.0000e-04 - 307ms/epoch - 26ms/step\n",
            "Epoch 1937/2000\n",
            "\n",
            "Epoch 1937: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0298 - accuracy: 0.9897 - mae: 0.0130 - mse: 0.0051 - val_loss: 1.0993 - val_accuracy: 0.7740 - val_mae: 0.1551 - val_mse: 0.1240 - lr: 1.0000e-04 - 319ms/epoch - 27ms/step\n",
            "Epoch 1938/2000\n",
            "\n",
            "Epoch 1938: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0316 - accuracy: 0.9902 - mae: 0.0137 - mse: 0.0053 - val_loss: 1.0969 - val_accuracy: 0.7750 - val_mae: 0.1551 - val_mse: 0.1237 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1939/2000\n",
            "\n",
            "Epoch 1939: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0283 - accuracy: 0.9905 - mae: 0.0129 - mse: 0.0048 - val_loss: 1.0966 - val_accuracy: 0.7743 - val_mae: 0.1550 - val_mse: 0.1237 - lr: 1.0000e-04 - 306ms/epoch - 26ms/step\n",
            "Epoch 1940/2000\n",
            "\n",
            "Epoch 1940: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0286 - accuracy: 0.9907 - mae: 0.0131 - mse: 0.0050 - val_loss: 1.1037 - val_accuracy: 0.7750 - val_mae: 0.1549 - val_mse: 0.1238 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1941/2000\n",
            "\n",
            "Epoch 1941: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0321 - accuracy: 0.9890 - mae: 0.0140 - mse: 0.0057 - val_loss: 1.1049 - val_accuracy: 0.7733 - val_mae: 0.1552 - val_mse: 0.1241 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1942/2000\n",
            "\n",
            "Epoch 1942: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0326 - accuracy: 0.9896 - mae: 0.0144 - mse: 0.0057 - val_loss: 1.1041 - val_accuracy: 0.7750 - val_mae: 0.1552 - val_mse: 0.1240 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1943/2000\n",
            "\n",
            "Epoch 1943: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0314 - accuracy: 0.9902 - mae: 0.0138 - mse: 0.0054 - val_loss: 1.1023 - val_accuracy: 0.7730 - val_mae: 0.1552 - val_mse: 0.1238 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1944/2000\n",
            "\n",
            "Epoch 1944: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0334 - accuracy: 0.9892 - mae: 0.0142 - mse: 0.0058 - val_loss: 1.0956 - val_accuracy: 0.7747 - val_mae: 0.1556 - val_mse: 0.1241 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1945/2000\n",
            "\n",
            "Epoch 1945: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0274 - accuracy: 0.9923 - mae: 0.0122 - mse: 0.0044 - val_loss: 1.0936 - val_accuracy: 0.7770 - val_mae: 0.1547 - val_mse: 0.1233 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1946/2000\n",
            "\n",
            "Epoch 1946: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0292 - accuracy: 0.9906 - mae: 0.0134 - mse: 0.0049 - val_loss: 1.1037 - val_accuracy: 0.7733 - val_mae: 0.1546 - val_mse: 0.1237 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1947/2000\n",
            "\n",
            "Epoch 1947: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0298 - accuracy: 0.9910 - mae: 0.0130 - mse: 0.0050 - val_loss: 1.0962 - val_accuracy: 0.7747 - val_mae: 0.1549 - val_mse: 0.1234 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1948/2000\n",
            "\n",
            "Epoch 1948: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0293 - accuracy: 0.9904 - mae: 0.0131 - mse: 0.0050 - val_loss: 1.1012 - val_accuracy: 0.7750 - val_mae: 0.1552 - val_mse: 0.1240 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1949/2000\n",
            "\n",
            "Epoch 1949: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0283 - accuracy: 0.9912 - mae: 0.0127 - mse: 0.0048 - val_loss: 1.1025 - val_accuracy: 0.7767 - val_mae: 0.1547 - val_mse: 0.1234 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1950/2000\n",
            "\n",
            "Epoch 1950: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0291 - accuracy: 0.9898 - mae: 0.0134 - mse: 0.0051 - val_loss: 1.1004 - val_accuracy: 0.7760 - val_mae: 0.1547 - val_mse: 0.1235 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1951/2000\n",
            "\n",
            "Epoch 1951: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0295 - accuracy: 0.9903 - mae: 0.0134 - mse: 0.0051 - val_loss: 1.1031 - val_accuracy: 0.7757 - val_mae: 0.1545 - val_mse: 0.1235 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1952/2000\n",
            "\n",
            "Epoch 1952: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0306 - accuracy: 0.9908 - mae: 0.0135 - mse: 0.0052 - val_loss: 1.1067 - val_accuracy: 0.7737 - val_mae: 0.1552 - val_mse: 0.1244 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1953/2000\n",
            "\n",
            "Epoch 1953: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0320 - accuracy: 0.9903 - mae: 0.0138 - mse: 0.0055 - val_loss: 1.1011 - val_accuracy: 0.7760 - val_mae: 0.1544 - val_mse: 0.1236 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1954/2000\n",
            "\n",
            "Epoch 1954: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0304 - accuracy: 0.9904 - mae: 0.0132 - mse: 0.0051 - val_loss: 1.1058 - val_accuracy: 0.7753 - val_mae: 0.1547 - val_mse: 0.1238 - lr: 1.0000e-04 - 281ms/epoch - 23ms/step\n",
            "Epoch 1955/2000\n",
            "\n",
            "Epoch 1955: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0305 - accuracy: 0.9915 - mae: 0.0134 - mse: 0.0050 - val_loss: 1.1007 - val_accuracy: 0.7743 - val_mae: 0.1551 - val_mse: 0.1241 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1956/2000\n",
            "\n",
            "Epoch 1956: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0280 - accuracy: 0.9914 - mae: 0.0127 - mse: 0.0046 - val_loss: 1.1000 - val_accuracy: 0.7743 - val_mae: 0.1556 - val_mse: 0.1244 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1957/2000\n",
            "\n",
            "Epoch 1957: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0289 - accuracy: 0.9906 - mae: 0.0135 - mse: 0.0050 - val_loss: 1.1001 - val_accuracy: 0.7733 - val_mae: 0.1556 - val_mse: 0.1243 - lr: 1.0000e-04 - 303ms/epoch - 25ms/step\n",
            "Epoch 1958/2000\n",
            "\n",
            "Epoch 1958: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0284 - accuracy: 0.9912 - mae: 0.0130 - mse: 0.0048 - val_loss: 1.1084 - val_accuracy: 0.7740 - val_mae: 0.1550 - val_mse: 0.1239 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1959/2000\n",
            "\n",
            "Epoch 1959: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0280 - accuracy: 0.9908 - mae: 0.0130 - mse: 0.0048 - val_loss: 1.1000 - val_accuracy: 0.7750 - val_mae: 0.1551 - val_mse: 0.1240 - lr: 1.0000e-04 - 296ms/epoch - 25ms/step\n",
            "Epoch 1960/2000\n",
            "\n",
            "Epoch 1960: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0276 - accuracy: 0.9910 - mae: 0.0129 - mse: 0.0046 - val_loss: 1.1049 - val_accuracy: 0.7737 - val_mae: 0.1554 - val_mse: 0.1246 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1961/2000\n",
            "\n",
            "Epoch 1961: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0297 - accuracy: 0.9906 - mae: 0.0135 - mse: 0.0050 - val_loss: 1.0994 - val_accuracy: 0.7737 - val_mae: 0.1550 - val_mse: 0.1241 - lr: 1.0000e-04 - 288ms/epoch - 24ms/step\n",
            "Epoch 1962/2000\n",
            "\n",
            "Epoch 1962: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0303 - accuracy: 0.9908 - mae: 0.0136 - mse: 0.0052 - val_loss: 1.1025 - val_accuracy: 0.7743 - val_mae: 0.1549 - val_mse: 0.1239 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1963/2000\n",
            "\n",
            "Epoch 1963: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0280 - accuracy: 0.9909 - mae: 0.0129 - mse: 0.0048 - val_loss: 1.0986 - val_accuracy: 0.7773 - val_mae: 0.1541 - val_mse: 0.1234 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1964/2000\n",
            "\n",
            "Epoch 1964: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0299 - accuracy: 0.9910 - mae: 0.0126 - mse: 0.0049 - val_loss: 1.1032 - val_accuracy: 0.7750 - val_mae: 0.1543 - val_mse: 0.1238 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1965/2000\n",
            "\n",
            "Epoch 1965: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0311 - accuracy: 0.9898 - mae: 0.0135 - mse: 0.0053 - val_loss: 1.0940 - val_accuracy: 0.7767 - val_mae: 0.1542 - val_mse: 0.1235 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1966/2000\n",
            "\n",
            "Epoch 1966: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0290 - accuracy: 0.9906 - mae: 0.0134 - mse: 0.0050 - val_loss: 1.1042 - val_accuracy: 0.7750 - val_mae: 0.1548 - val_mse: 0.1239 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1967/2000\n",
            "\n",
            "Epoch 1967: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0279 - accuracy: 0.9914 - mae: 0.0130 - mse: 0.0047 - val_loss: 1.1015 - val_accuracy: 0.7760 - val_mae: 0.1545 - val_mse: 0.1237 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1968/2000\n",
            "\n",
            "Epoch 1968: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0280 - accuracy: 0.9902 - mae: 0.0127 - mse: 0.0049 - val_loss: 1.1113 - val_accuracy: 0.7760 - val_mae: 0.1545 - val_mse: 0.1236 - lr: 1.0000e-04 - 283ms/epoch - 24ms/step\n",
            "Epoch 1969/2000\n",
            "\n",
            "Epoch 1969: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0286 - accuracy: 0.9908 - mae: 0.0128 - mse: 0.0049 - val_loss: 1.1069 - val_accuracy: 0.7760 - val_mae: 0.1540 - val_mse: 0.1234 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1970/2000\n",
            "\n",
            "Epoch 1970: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0298 - accuracy: 0.9899 - mae: 0.0130 - mse: 0.0053 - val_loss: 1.1064 - val_accuracy: 0.7747 - val_mae: 0.1543 - val_mse: 0.1239 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "Epoch 1971/2000\n",
            "\n",
            "Epoch 1971: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0288 - accuracy: 0.9893 - mae: 0.0130 - mse: 0.0052 - val_loss: 1.0997 - val_accuracy: 0.7747 - val_mae: 0.1547 - val_mse: 0.1239 - lr: 1.0000e-04 - 305ms/epoch - 25ms/step\n",
            "Epoch 1972/2000\n",
            "\n",
            "Epoch 1972: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0289 - accuracy: 0.9904 - mae: 0.0125 - mse: 0.0048 - val_loss: 1.0997 - val_accuracy: 0.7757 - val_mae: 0.1548 - val_mse: 0.1237 - lr: 1.0000e-04 - 285ms/epoch - 24ms/step\n",
            "Epoch 1973/2000\n",
            "\n",
            "Epoch 1973: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0298 - accuracy: 0.9898 - mae: 0.0132 - mse: 0.0052 - val_loss: 1.1017 - val_accuracy: 0.7730 - val_mae: 0.1549 - val_mse: 0.1239 - lr: 1.0000e-04 - 313ms/epoch - 26ms/step\n",
            "Epoch 1974/2000\n",
            "\n",
            "Epoch 1974: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0280 - accuracy: 0.9913 - mae: 0.0126 - mse: 0.0047 - val_loss: 1.0985 - val_accuracy: 0.7750 - val_mae: 0.1548 - val_mse: 0.1237 - lr: 1.0000e-04 - 291ms/epoch - 24ms/step\n",
            "Epoch 1975/2000\n",
            "\n",
            "Epoch 1975: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0295 - accuracy: 0.9909 - mae: 0.0129 - mse: 0.0049 - val_loss: 1.1012 - val_accuracy: 0.7743 - val_mae: 0.1547 - val_mse: 0.1239 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1976/2000\n",
            "\n",
            "Epoch 1976: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0285 - accuracy: 0.9917 - mae: 0.0128 - mse: 0.0047 - val_loss: 1.1009 - val_accuracy: 0.7757 - val_mae: 0.1541 - val_mse: 0.1234 - lr: 1.0000e-04 - 287ms/epoch - 24ms/step\n",
            "Epoch 1977/2000\n",
            "\n",
            "Epoch 1977: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0301 - accuracy: 0.9905 - mae: 0.0131 - mse: 0.0051 - val_loss: 1.1020 - val_accuracy: 0.7780 - val_mae: 0.1538 - val_mse: 0.1231 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1978/2000\n",
            "\n",
            "Epoch 1978: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0274 - accuracy: 0.9913 - mae: 0.0128 - mse: 0.0046 - val_loss: 1.1030 - val_accuracy: 0.7743 - val_mae: 0.1542 - val_mse: 0.1235 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1979/2000\n",
            "\n",
            "Epoch 1979: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0269 - accuracy: 0.9909 - mae: 0.0127 - mse: 0.0046 - val_loss: 1.1039 - val_accuracy: 0.7767 - val_mae: 0.1544 - val_mse: 0.1236 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1980/2000\n",
            "\n",
            "Epoch 1980: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0287 - accuracy: 0.9908 - mae: 0.0128 - mse: 0.0046 - val_loss: 1.1074 - val_accuracy: 0.7740 - val_mae: 0.1550 - val_mse: 0.1242 - lr: 1.0000e-04 - 294ms/epoch - 25ms/step\n",
            "Epoch 1981/2000\n",
            "\n",
            "Epoch 1981: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0314 - accuracy: 0.9896 - mae: 0.0136 - mse: 0.0054 - val_loss: 1.1031 - val_accuracy: 0.7750 - val_mae: 0.1550 - val_mse: 0.1240 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1982/2000\n",
            "\n",
            "Epoch 1982: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0276 - accuracy: 0.9912 - mae: 0.0125 - mse: 0.0046 - val_loss: 1.1120 - val_accuracy: 0.7733 - val_mae: 0.1551 - val_mse: 0.1242 - lr: 1.0000e-04 - 292ms/epoch - 24ms/step\n",
            "Epoch 1983/2000\n",
            "\n",
            "Epoch 1983: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0302 - accuracy: 0.9902 - mae: 0.0135 - mse: 0.0052 - val_loss: 1.1074 - val_accuracy: 0.7733 - val_mae: 0.1547 - val_mse: 0.1239 - lr: 1.0000e-04 - 298ms/epoch - 25ms/step\n",
            "Epoch 1984/2000\n",
            "\n",
            "Epoch 1984: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0288 - accuracy: 0.9912 - mae: 0.0129 - mse: 0.0048 - val_loss: 1.1087 - val_accuracy: 0.7737 - val_mae: 0.1550 - val_mse: 0.1242 - lr: 1.0000e-04 - 308ms/epoch - 26ms/step\n",
            "Epoch 1985/2000\n",
            "\n",
            "Epoch 1985: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0309 - accuracy: 0.9902 - mae: 0.0135 - mse: 0.0053 - val_loss: 1.1016 - val_accuracy: 0.7727 - val_mae: 0.1551 - val_mse: 0.1241 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1986/2000\n",
            "\n",
            "Epoch 1986: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0251 - accuracy: 0.9915 - mae: 0.0120 - mse: 0.0043 - val_loss: 1.1132 - val_accuracy: 0.7730 - val_mae: 0.1550 - val_mse: 0.1242 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 1987/2000\n",
            "\n",
            "Epoch 1987: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0293 - accuracy: 0.9904 - mae: 0.0129 - mse: 0.0050 - val_loss: 1.1071 - val_accuracy: 0.7753 - val_mae: 0.1545 - val_mse: 0.1235 - lr: 1.0000e-04 - 290ms/epoch - 24ms/step\n",
            "Epoch 1988/2000\n",
            "\n",
            "Epoch 1988: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0291 - accuracy: 0.9906 - mae: 0.0132 - mse: 0.0050 - val_loss: 1.1145 - val_accuracy: 0.7757 - val_mae: 0.1544 - val_mse: 0.1237 - lr: 1.0000e-04 - 295ms/epoch - 25ms/step\n",
            "Epoch 1989/2000\n",
            "\n",
            "Epoch 1989: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0301 - accuracy: 0.9889 - mae: 0.0135 - mse: 0.0054 - val_loss: 1.1098 - val_accuracy: 0.7770 - val_mae: 0.1543 - val_mse: 0.1235 - lr: 1.0000e-04 - 286ms/epoch - 24ms/step\n",
            "Epoch 1990/2000\n",
            "\n",
            "Epoch 1990: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0278 - accuracy: 0.9912 - mae: 0.0127 - mse: 0.0048 - val_loss: 1.1136 - val_accuracy: 0.7730 - val_mae: 0.1547 - val_mse: 0.1239 - lr: 1.0000e-04 - 293ms/epoch - 24ms/step\n",
            "Epoch 1991/2000\n",
            "\n",
            "Epoch 1991: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0304 - accuracy: 0.9894 - mae: 0.0133 - mse: 0.0053 - val_loss: 1.1078 - val_accuracy: 0.7747 - val_mae: 0.1542 - val_mse: 0.1235 - lr: 1.0000e-04 - 311ms/epoch - 26ms/step\n",
            "Epoch 1992/2000\n",
            "\n",
            "Epoch 1992: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0261 - accuracy: 0.9922 - mae: 0.0123 - mse: 0.0043 - val_loss: 1.1202 - val_accuracy: 0.7753 - val_mae: 0.1546 - val_mse: 0.1240 - lr: 1.0000e-04 - 289ms/epoch - 24ms/step\n",
            "Epoch 1993/2000\n",
            "\n",
            "Epoch 1993: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0271 - accuracy: 0.9916 - mae: 0.0123 - mse: 0.0045 - val_loss: 1.1127 - val_accuracy: 0.7747 - val_mae: 0.1544 - val_mse: 0.1238 - lr: 1.0000e-04 - 294ms/epoch - 24ms/step\n",
            "Epoch 1994/2000\n",
            "\n",
            "Epoch 1994: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0296 - accuracy: 0.9908 - mae: 0.0129 - mse: 0.0050 - val_loss: 1.1117 - val_accuracy: 0.7740 - val_mae: 0.1546 - val_mse: 0.1240 - lr: 1.0000e-04 - 320ms/epoch - 27ms/step\n",
            "Epoch 1995/2000\n",
            "\n",
            "Epoch 1995: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0312 - accuracy: 0.9903 - mae: 0.0126 - mse: 0.0051 - val_loss: 1.1022 - val_accuracy: 0.7740 - val_mae: 0.1552 - val_mse: 0.1241 - lr: 1.0000e-04 - 299ms/epoch - 25ms/step\n",
            "Epoch 1996/2000\n",
            "\n",
            "Epoch 1996: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0287 - accuracy: 0.9904 - mae: 0.0134 - mse: 0.0049 - val_loss: 1.0978 - val_accuracy: 0.7760 - val_mae: 0.1549 - val_mse: 0.1236 - lr: 1.0000e-04 - 301ms/epoch - 25ms/step\n",
            "Epoch 1997/2000\n",
            "\n",
            "Epoch 1997: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0270 - accuracy: 0.9913 - mae: 0.0124 - mse: 0.0045 - val_loss: 1.1041 - val_accuracy: 0.7767 - val_mae: 0.1545 - val_mse: 0.1234 - lr: 1.0000e-04 - 309ms/epoch - 26ms/step\n",
            "Epoch 1998/2000\n",
            "\n",
            "Epoch 1998: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0272 - accuracy: 0.9913 - mae: 0.0124 - mse: 0.0046 - val_loss: 1.1097 - val_accuracy: 0.7770 - val_mae: 0.1536 - val_mse: 0.1229 - lr: 1.0000e-04 - 323ms/epoch - 27ms/step\n",
            "Epoch 1999/2000\n",
            "\n",
            "Epoch 1999: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0289 - accuracy: 0.9904 - mae: 0.0129 - mse: 0.0049 - val_loss: 1.1166 - val_accuracy: 0.7770 - val_mae: 0.1536 - val_mse: 0.1231 - lr: 1.0000e-04 - 304ms/epoch - 25ms/step\n",
            "Epoch 2000/2000\n",
            "\n",
            "Epoch 2000: val_loss did not improve from 0.57547\n",
            "12/12 - 0s - loss: 0.0270 - accuracy: 0.9924 - mae: 0.0120 - mse: 0.0045 - val_loss: 1.1075 - val_accuracy: 0.7750 - val_mae: 0.1531 - val_mse: 0.1224 - lr: 1.0000e-04 - 284ms/epoch - 24ms/step\n",
            "CPU times: user 4h 15min 45s, sys: 1h 56min 45s, total: 6h 12min 31s\n",
            "Wall time: 10min 4s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Model Evaluation"
      ],
      "metadata": {
        "id": "CQZRdkOKfvnL"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from matplotlib import pyplot as plt\n",
        "%matplotlib inline\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(hist.history['loss'])\n",
        "plt.plot(hist.history['val_loss'])\n",
        "\n",
        "plt.title('Model loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['train_loss', 'val_loss'], loc='upper left')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "kQGQMtu6fvHH",
        "outputId": "4aaef96c-c164-4218-c7d6-c65497d61284"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(hist.history['accuracy'])\n",
        "plt.legend(['accuracy'], loc='upper left')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 447
        },
        "id": "oOO9pVR-c_ef",
        "outputId": "51a59b33-6978-47cf-ee21-5812be47c013"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7e18f54e42e0>"
            ]
          },
          "metadata": {},
          "execution_count": 74
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(hist.history['accuracy'])\n",
        "plt.plot(hist.history['val_loss'])\n",
        "plt.legend(['accuracy', 'val_loss'], loc='upper left')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 447
        },
        "id": "0FKANVkijBpO",
        "outputId": "7adafff4-f069-4f1b-e76f-637e5706c372"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7e19065aa9e0>"
            ]
          },
          "metadata": {},
          "execution_count": 75
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "c0aYnVRiM9K7"
      },
      "source": [
        "# Test Animal Classification"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import classification_report, confusion_matrix\n",
        "import seaborn as sns"
      ],
      "metadata": {
        "id": "5Ibp4aWgz9xJ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# bird, dog, horse\n",
        "actual = np.array(\n",
        "    ['Bird', 'Dog', 'Horse', 'Bird', 'Dog', 'Bird', 'Dog',  'Horse', 'Horse', 'Bird'])\n",
        "predicted = np.array(\n",
        "    ['Bird', 'Dog', 'Dog', 'Bird', 'Dog', 'Bird', 'Dog',  'Horse', 'Horse', 'Dog'])"
      ],
      "metadata": {
        "id": "vh464CDvNm0j"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from keras.models import load_model\n",
        "animal_model = load_model('./models/animal_cnn.h5')"
      ],
      "metadata": {
        "id": "sjzMgT9aOOQF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Alml5e7VND8y",
        "outputId": "e40fb68e-de36-4387-ee99-d803cd913545"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "94/94 [==============================] - 0s 2ms/step\n",
            "Confusion Matrix and Performance of Model Architecture\n",
            "Model trained with images of 3 classes with input size of 15x15 and 1 channel.\n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "        Bird       0.76      0.78      0.77      1000\n",
            "         Dog       0.74      0.75      0.74      1000\n",
            "       Horse       0.82      0.78      0.80      1000\n",
            "\n",
            "    accuracy                           0.77      3000\n",
            "   macro avg       0.77      0.77      0.77      3000\n",
            "weighted avg       0.77      0.77      0.77      3000\n",
            "\n"
          ]
        }
      ],
      "source": [
        "# convert one-hot encoded labels to integer labels\n",
        "y_true = np.argmax(y_test_animal, axis=1)\n",
        "\n",
        "# get model predictions\n",
        "y_predict = animal_model.predict(X_test_final)\n",
        "\n",
        "# convert predictions to integers\n",
        "y_predict = np.argmax(y_predict, axis=1)\n",
        "\n",
        "#rounded_labels = np.argmax(y_test, axis=1) # convert label to single-digit\n",
        "\n",
        "cm = confusion_matrix(y_true, y_predict)\n",
        "print(\"Confusion Matrix and Performance of Model Architecture\")\n",
        "print(\"Model trained with images of 3 classes with input size of 15x15 and 1 channel.\\n\")\n",
        "target_names = ['Bird', 'Dog', 'Horse']\n",
        "print(classification_report(y_true, y_predict, target_names=target_names))\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "sns.heatmap(cm,\n",
        "            annot=True,\n",
        "            fmt='g',\n",
        "            cmap=\"Blues\",\n",
        "            xticklabels=['Bird', 'Dog', 'Horse'],\n",
        "            yticklabels=['Bird', 'Dog', 'Horse'])\n",
        "plt.ylabel('Prediction', fontsize=13)\n",
        "plt.xlabel('Actual', fontsize=13)\n",
        "plt.title('Confusion Matrix', fontsize=17)\n",
        "plt.show()\n",
        "\n",
        "print(cm)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 533
        },
        "id": "f0g06UDbOV5l",
        "outputId": "45cc80c4-b0f8-482a-9a90-42fac88854a0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[783 134  83]\n",
            " [159 749  92]\n",
            " [ 93 129 778]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Test Animal Classification with Financial Model"
      ],
      "metadata": {
        "id": "Gr7CCX2D3gt7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!ls"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ky_DT-5536si",
        "outputId": "55eab72e-bd31-4dad-9dbf-44971a1ab763"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "'1) Process Stock Data.ipynb'\t    model.png\n",
            "'2) Label Stock Data.ipynb'\t    models\n",
            "'3) Calculate Initial Indicators'  'Stock CNN 2k'\n",
            "'4) Original Stock CNN.ipynb'\t   'Stock CNN 2k Intersected Features'\n",
            "'5) Feature Selection'\t\t   'Stock CNN 2k K Best Features'\n",
            " Animals\t\t\t   \"Stock CNN 2k K Best Features Paper's Architecture\"\n",
            "'archived implementations'\t   \"Stock CNN 2k Paper's Architecture\"\n",
            " Backtester\t\t\t   'Stock CNN 3k'\n",
            " data\t\t\t\t    ta-lib\n",
            "'Expand Indicators'\t\t    ta-lib-0.4.0-src.tar.gz\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import classification_report, confusion_matrix\n",
        "import seaborn as sns"
      ],
      "metadata": {
        "id": "KHPFmiVxPO8h"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# bird, dog, horse\n",
        "actual = np.array(\n",
        "    ['Bird', 'Dog', 'Horse', 'Bird', 'Dog', 'Bird', 'Dog',  'Horse', 'Horse', 'Bird'])\n",
        "predicted = np.array(\n",
        "    ['Bird', 'Dog', 'Dog', 'Bird', 'Dog', 'Bird', 'Dog',  'Horse', 'Horse', 'Dog'])"
      ],
      "metadata": {
        "id": "DJLTLgJePO8m"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from keras.models import load_model\n",
        "financial_model = load_model('./models/stock_cnn_best_model_old_intersected_features_new_arch.h5')"
      ],
      "metadata": {
        "id": "DoN8mhztPO8m"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# convert one-hot encoded labels to integer labels\n",
        "y_true = np.argmax(y_test_animal, axis=1)\n",
        "\n",
        "# get model predictions\n",
        "y_predict = financial_model.predict(X_test_final)\n",
        "\n",
        "# convert predictions to integers\n",
        "y_predict = np.argmax(y_predict, axis=1)\n",
        "\n",
        "#rounded_labels = np.argmax(y_test, axis=1) # convert label to single-digit\n",
        "\n",
        "cm = confusion_matrix(y_true, y_predict)\n",
        "print(\"Confusion Matrix and Performance of Finance Model\")\n",
        "print(\"Model trained with financial images created using stock indicators.\\n\")\n",
        "target_names = ['Bird', 'Dog', 'Horse']\n",
        "print(classification_report(y_true, y_predict, target_names=target_names))\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SOLjvT-d35Om",
        "outputId": "8d0f4ad6-c847-4995-f429-c93600d187fb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "94/94 [==============================] - 1s 5ms/step\n",
            "Confusion Matrix and Performance of Finance Model\n",
            "Model trained with financial images created using stock indicators.\n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "        Bird       0.25      0.20      0.22      1000\n",
            "         Dog       0.00      0.00      0.00      1000\n",
            "       Horse       0.35      0.76      0.48      1000\n",
            "\n",
            "    accuracy                           0.32      3000\n",
            "   macro avg       0.20      0.32      0.23      3000\n",
            "weighted avg       0.20      0.32      0.23      3000\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_classification.py:1344: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
            "  _warn_prf(average, modifier, msg_start, len(result))\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_classification.py:1344: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
            "  _warn_prf(average, modifier, msg_start, len(result))\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/metrics/_classification.py:1344: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
            "  _warn_prf(average, modifier, msg_start, len(result))\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "sns.heatmap(cm,\n",
        "            annot=True,\n",
        "            fmt='g',\n",
        "            cmap=\"Blues\",\n",
        "            xticklabels=['Bird', 'Dog', 'Horse'],\n",
        "            yticklabels=['Bird', 'Dog', 'Horse'])\n",
        "plt.ylabel('Prediction', fontsize=13)\n",
        "plt.xlabel('Actual', fontsize=13)\n",
        "plt.title('Confusion Matrix', fontsize=17)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 481
        },
        "id": "kXNgfn9dQ841",
        "outputId": "542526c5-ee42-4ede-d700-60e571cc9bd1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}