Cat and Dog Image Classifier

Cat:

Dog:

Overview

This Python script implements an image classification model using a Support Vector Machine (SVM) with a Convolutional Neural Network (CNN). The model is trained to distinguish between images of cats and dogs. The project is part of Prodigy Infotech's third task.

Instructions for Use

1. Importing Important Modules:

   • Pandas: Data manipulation and analysis.
   • NumPy: Numerical operations.
   • TensorFlow: Neural network construction and training.

2. Preprocessing the Training Data:

   • Data augmentation is applied using TensorFlow's ImageDataGenerator to prevent overfitting.
   • Training and test sets are generated from the provided dataset.

3. Creating the Model:

   • The CNN model is built using TensorFlow's Keras API.
   • Layers include convolutional layers, pooling layers, flattening layers, fully connected layers, and an output layer.

4. Training the CNN:

   • The model is compiled with the Adam optimizer and hinge loss function.
   • Training involves fitting the model to the training set and validating it on the test set for multiple epochs.

5. Plotting Training and Validation Metrics:

   • Code is provided to plot training and validation loss, as well as training and validation accuracy over epochs.

6. Saving the Trained Model:

   • The trained CNN model is saved as a .h5 file for future use or deployment.

7. Testing the Model on a Sample Image:

   • A sample image from the test set is loaded, preprocessed, and used to make predictions.
   • The result is displayed alongside the image.

Conclusion

This project serves as a tutorial on image classification using SVM with CNN. Feel free to modify the code or address any bugs. For questions or improvements, please contact the author.

Author

Advait Dongre