The goal of this project was to understand the inner workings of neural networks without relying on high-level libraries like TensorFlow or PyTorch. To do this, I decided to build the Neural Network from scratch using only Python and numpy to classify handwritten digits from the MNIST dataset.
Last Updated: 10/12/2024
Data preparation: Ensure that the MNIST dataset is downloaded and placed in the appropriate directory (e.g., /content/data/mnist_train.csv in your Colab notebook).
Run the code: Execute the Python script or notebook containing the code. The training process will be displayed, including the accuracy at each iteration.
Evaluate performance: After training, use the trained model to make predictions on the test set and assess its accuracy.
Experiment with different architectures: Explore deeper networks, convolutional layers, or recurrent neural networks.
Optimize hyperparameters: Fine-tune the learning rate, number of epochs, and other parameters to improve performance.
Implement regularization techniques: Prevent overfitting by using techniques like dropout or L1/L2 regularization.
This project is licensed under the MIT License.
Thank you to the authors of the resources and tutorials that inspired this project.