Rock Paper Scissors.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import csv\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from tensorflow.keras import optimizers "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(27455, 28, 28)\n",
      "(27455,)\n",
      "(7172, 28, 28)\n",
      "(7172,)\n"
     ]
    }
   ],
   "source": [
    "def get_data(filename):\n",
    "    with open(filename) as training_file:\n",
    "        csv_reader = csv.reader(training_file, delimiter=',')\n",
    "        first_line = True\n",
    "        temp_images = []\n",
    "        temp_labels = []\n",
    "        for row in csv_reader:\n",
    "            if first_line:\n",
    "                # print(\"Ignoring first line\")\n",
    "                first_line = False\n",
    "            else:\n",
    "                temp_labels.append(row[0])\n",
    "                image_data = row[1:785]\n",
    "                image_data_as_array = np.array_split(image_data, 28)\n",
    "                temp_images.append(image_data_as_array)\n",
    "        images = np.array(temp_images).astype('float')\n",
    "        labels = np.array(temp_labels).astype('float')\n",
    "    return images, labels\n",
    "\n",
    "\n",
    "training_images, training_labels = get_data('sign_mnist_train.csv')\n",
    "testing_images, testing_labels = get_data('sign_mnist_test.csv')\n",
    "\n",
    "print(training_images.shape)\n",
    "print(training_labels.shape)\n",
    "print(testing_images.shape)\n",
    "print(testing_labels.shape)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(27455, 28, 28, 1)\n",
      "(7172, 28, 28, 1)\n"
     ]
    }
   ],
   "source": [
    "training_images = np.expand_dims(training_images, axis=3)\n",
    "testing_images = np.expand_dims(testing_images, axis=3)\n",
    "\n",
    "train_datagen = ImageDataGenerator(\n",
    "    rescale=1. / 255,\n",
    "    rotation_range=40,\n",
    "    width_shift_range=0.2,\n",
    "    height_shift_range=0.2,\n",
    "    shear_range=0.2,\n",
    "    zoom_range=0.2,\n",
    "    horizontal_flip=True,\n",
    "    fill_mode='nearest')\n",
    "\n",
    "validation_datagen = ImageDataGenerator(\n",
    "    rescale=1. / 255)\n",
    "\n",
    "print(training_images.shape)\n",
    "print(testing_images.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/15\n",
      "858/857 [==============================] - 43s 50ms/step - loss: 2.8148 - acc: 0.1451 - val_loss: 2.1766 - val_acc: 0.3118\n",
      "Epoch 2/15\n",
      "858/857 [==============================] - 43s 50ms/step - loss: 2.0473 - acc: 0.3549 - val_loss: 1.4318 - val_acc: 0.5238\n",
      "Epoch 3/15\n",
      "858/857 [==============================] - 41s 48ms/step - loss: 1.6730 - acc: 0.4617 - val_loss: 1.0278 - val_acc: 0.6627\n",
      "Epoch 4/15\n",
      "858/857 [==============================] - 43s 50ms/step - loss: 1.4387 - acc: 0.5347 - val_loss: 0.9661 - val_acc: 0.6832\n",
      "Epoch 5/15\n",
      "858/857 [==============================] - 40s 47ms/step - loss: 1.2789 - acc: 0.5846 - val_loss: 0.7557 - val_acc: 0.7464\n",
      "Epoch 6/15\n",
      "858/857 [==============================] - 42s 49ms/step - loss: 1.1483 - acc: 0.6280 - val_loss: 0.6893 - val_acc: 0.7652\n",
      "Epoch 7/15\n",
      "858/857 [==============================] - 43s 50ms/step - loss: 1.0674 - acc: 0.6463 - val_loss: 0.5681 - val_acc: 0.8123\n",
      "Epoch 8/15\n",
      "858/857 [==============================] - 43s 50ms/step - loss: 0.9864 - acc: 0.6749 - val_loss: 0.5519 - val_acc: 0.8161\n",
      "Epoch 9/15\n",
      "858/857 [==============================] - 41s 48ms/step - loss: 0.9256 - acc: 0.6929 - val_loss: 0.4372 - val_acc: 0.8450\n",
      "Epoch 10/15\n",
      "858/857 [==============================] - 42s 49ms/step - loss: 0.8702 - acc: 0.7111 - val_loss: 0.5596 - val_acc: 0.8161\n",
      "Epoch 11/15\n",
      "858/857 [==============================] - 41s 47ms/step - loss: 0.8045 - acc: 0.7330 - val_loss: 0.4838 - val_acc: 0.8390\n",
      "Epoch 12/15\n",
      "858/857 [==============================] - 44s 51ms/step - loss: 0.7850 - acc: 0.7373 - val_loss: 0.3898 - val_acc: 0.8692\n",
      "Epoch 13/15\n",
      "858/857 [==============================] - 42s 49ms/step - loss: 0.7470 - acc: 0.7505 - val_loss: 0.4548 - val_acc: 0.8419\n",
      "Epoch 14/15\n",
      "858/857 [==============================] - 43s 50ms/step - loss: 0.7143 - acc: 0.7608 - val_loss: 0.4215 - val_acc: 0.8558\n",
      "Epoch 15/15\n",
      "858/857 [==============================] - 41s 48ms/step - loss: 0.6872 - acc: 0.7713 - val_loss: 0.3378 - val_acc: 0.8875\n",
      "7172/7172 [==============================] - 4s 561us/sample - loss: 154.4213 - acc: 0.6518\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[154.42131806696452, 0.6518405]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = tf.keras.models.Sequential([\n",
    "    tf.keras.layers.Conv2D(64, (3, 3), activation='relu', input_shape=(28, 28, 1)),\n",
    "    tf.keras.layers.MaxPooling2D(2, 2),\n",
    "    tf.keras.layers.Conv2D(64, (3, 3), activation='relu'),\n",
    "    tf.keras.layers.MaxPooling2D(2, 2),\n",
    "    tf.keras.layers.Flatten(),\n",
    "    tf.keras.layers.Dense(128, activation=tf.nn.relu),\n",
    "    tf.keras.layers.Dense(26, activation=tf.nn.softmax)])\n",
    "\n",
    "model.compile(optimizer = 'adam',\n",
    "              loss = 'sparse_categorical_crossentropy',\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "history = model.fit(train_datagen.flow(training_images, training_labels, batch_size=32),\n",
    "                              steps_per_epoch=len(training_images) / 32,\n",
    "                              epochs=15,\n",
    "                              validation_data=validation_datagen.flow(testing_images, testing_labels, batch_size=32),\n",
    "                              validation_steps=len(testing_images) / 32)\n",
    "\n",
    "model.evaluate(testing_images, testing_labels)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "acc = history.history['acc']\n",
    "val_acc = history.history['val_acc']\n",
    "loss = history.history['loss']\n",
    "val_loss = history.history['val_loss']\n",
    "\n",
    "epochs = range(len(acc))\n",
    "\n",
    "plt.plot(epochs, acc, 'r', label='Training accuracy')\n",
    "plt.plot(epochs, val_acc, 'b', label='Validation accuracy')\n",
    "plt.title('Training and validation accuracy')\n",
    "plt.legend()\n",
    "plt.figure()\n",
    "\n",
    "plt.plot(epochs, loss, 'r', label='Training Loss')\n",
    "plt.plot(epochs, val_loss, 'b', label='Validation Loss')\n",
    "plt.title('Training and validation loss')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}