This project is a Convolutional Neural Network (CNN) classifier for images of cats, dogs, and pandas. The dataset is organized into the dataset directory, with three subdirectories: cat, dog, and panda, each containing 1000 images.
The dataset is split into training, validation, and test sets with the following proportions:
- Training Set: 70%
- Validation Set: 15%
- Test Set: 15%
I train the model for 45 epochs. At the end of each epoch, the following metrics are recorded and plotted:
- Training Loss
- Validation Loss
- Training Accuracy
- Validation Accuracy
These metrics, plotted against the number of epochs, help us observe and judge the training performance of the model.
Finally, I evaluate the model's performance using the Test Loss and Test Accuracy.
- Clone or download this project.
- Install the required dependencies:
pip install -r requirements.txt - Run the main script:
python main.py
Training process metrics:
Final test results:
Test Loss: 0.6178
Test Accuracy: 71.78%