Le Net

Introduction

This repository hosts the LeNet model for image classification using the MNIST dataset, available in the MindSpore library on Hugging Face's platform. LeNet, introduced by Yann LeCun in 1998, is known for its simplicity and effectiveness in recognizing handwritten characters.

Architecture

LeNet-5 is a multi-layer convolutional neural network architecture designed for image classification tasks. It consists of several layers, including convolutional layers, pooling layers, and fully connected layers, which work together to classify input images. This architecture was pivotal in advancing the field of document recognition.

Training

The LeNet model is trained on the MNIST dataset, which consists of 60,000 training images and 10,000 test images of handwritten digits. The model utilizes gradient-based learning to optimize its parameters for accurate classification.

Guide: Running Locally

To run the LeNet model locally, follow these steps:

1. Clone the Repository: Download the repository from Hugging Face using Git.
2. Install Dependencies: Ensure you have Python and MindSpore installed. Use pip to install any additional required packages.
3. Download MNIST Dataset: Ensure the MNIST dataset is available in your working directory or modify the code to fetch it directly.
4. Run the Model: Use the provided script to train and evaluate the model on your local machine.

For faster training and experimentation, consider using cloud GPUs such as AWS EC2, Google Cloud Platform, or Azure.

License

This project is licensed under the Apache-2.0 License, allowing for permissive use, modification, and distribution.