Dog Breed Classifier

Overview

This project implements a dog breed classifier using a convolutional neural network (CNN) with TensorFlow and Keras. The model is trained to classify dog breeds based on a dataset of images.

Project Structure

• image_classifier.py: The main script that contains the code for model training and prediction.
• saved_models: Directory to save the trained model (model.keras).
• dog-breeds
  • train: Training set directory.
  • valid: Validation set directory.
  • test: Testing set directory.

Setup and Dependencies

1. Install the required dependencies:

   pip install tensorflow pillow

2. Run the script:

   python image_classifier.py

Usage

1. Train the model by running image_classifier.py. Adjust the script to point to your dataset.
2. Use the trained model to make predictions on dog breed types. You can input a new image path when prompted.

Important Files

• image_classifier.py: Main script for model training and prediction.
• model/model.keras: Trained model saved in keras format.
• dog-breeds: Directory containing the dataset.

Note

• The dataset should be organized with subdirectories for each class in both the training and testing sets.
• Modify the script paths to match the location of your dataset.