Grapes Image Classification

This repository contains code for a grape image classification project using TensorFlow and VGG16 model.

Overview

The code performs the following tasks:

• Downloads and preprocesses the grape images dataset
• Trains a VGG16 model on the training set
• Evaluates the model on the validation and test sets
• Generates classification reports and confusion matrix
• Makes predictions on new images

Dependencies

The following dependencies are required to run the code:

• Python 3.7
• TensorFlow 2.x
• NumPy
• Pandas
• Seaborn
• Matplotlib
• PIL

Usage

To run the code:

1. Clone the repository: git clone https://github.com/majid0110/grapes-image-classification.git
2. Install the required dependencies: pip install -r requirements.txt
3. Run the main script: python main.py

File Structure

The repository has the following file structure:

grapes-image-classification/ ├── main.py ├── model.py ├── Grape_leaves_disease_detection.ipynb ├── dataset/ │ ├── grapes images/ │ │ ├── train/ │ │ ├── validation/ │ │ └── test/ │ └── grape_leaf.jpeg └── README.md

• main.py: The main script that trains the model and makes predictions.
• model.py: Defines the model architecture and training configuration.
• dataset/: Contains the dataset folders and example test image.
• README.md: The README file you're currently reading.

Results

Here are some results and visualizations obtained from running the code:

• Training and validation accuracy/loss curves
• Classification report on the validation set
• Confusion matrix
• Predicted class for a sample test image