This project aims to classify images of dogs and cats using a Convolutional Neural Network (CNN) implemented in Python with TensorFlow. The model is trained on a dataset of dog and cat images, and the performance is evaluated based on accuracy and loss metrics.
train.py: Script to train the CNN model using TensorFlow's Keras API. It includes data preprocessing, model creation, training, and saving the model.predict.py: Script to load the saved model and make predictions on new images.model.py: Contains the function to create the CNN model.data_preprocessing.py: Contains the function to create data generators for training and validation.__init__.py: Marks the directory as a Python package.
The CNN model is built using Keras with the following layers:
- Convolutional layers
- MaxPooling layers
- Dense layers
- Dropout layer
To train the model, run the following command:
python train.pyThis will preprocess the data, train the model, and save the trained model to the saved_models directory.
To make predictions using the trained model, run the following command:
python predict.pyThis will load the saved model and make predictions on the images in the Datasets/test directory.
- Python 3.12.4
- TensorFlow
- Keras
- Matplotlib
- Pandas
- Numpy
Install the required packages using:
pip install -r requirements.txtThis project demonstrates how to build, train, and evaluate a CNN model for image classification. The trained model can distinguish between images of dogs and cats with high accuracy. The visualizations of the training process and the prediction results provide insights into the model's performance.