MNIST classification using LeNet5 model

This repository showcases a deep learning project implementing the LeNet-5 architecture [1]. The project utilizes the MNIST dataset to demonstrate LeNet-5's effectiveness in image classification.

Table of Contents

• Running the Project
• Directory Structure
• Dataset
• LeNet-5 Model
• Utilities
• References

Running the Project

1. Clone the repository:

   git clone https//github.com/Alessio1599/LeNet-5-MNIST-Classification.git
   cd LeNet-5-MNIST-Classification

2. Install required packages:

   pip install -r requirements.txt

3. Run the main script:

   cd code
   python main.py

   Example of the use of flags

   python main_wandb.py --epochs=10 --batch_size=250

Directory structure

LeNet-5-MNIST-Classification/
├── LICENSE
├── LeNet5-MNIST.ipynb
├── README.md
├── code
│   ├── main.py
│   ├── main_wandb.py
│   └── util.py
└── requirements.txt

Dataset

The project uses the MNIST dataset, which is automatically downloaded when running the scripts or notebooks.

LeNet-5 Model

The model is built using the LeNet-5 architecture, which consists of:

• Convolutional layers
• Average pooling layers
• Fully connected layers

Utilities

The utility functions help in visualizing the training process and evaluating the model performance:

• plot_history(history, metric): Plots the training and validation loss and metrics.
• show_confusion_matrix(conf_matrix, class_names): Displays the confusion matrix with class labels.
• Image_inspection(data): Displays 10 random images from the dataset.

References

[1] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner. "Gradient-based learning applied to document recognition." Proceedings of the IEEE, 86(11):2278-2324, November 1998