Number Recognition using PyTorch

This project demonstrates the implementation of the LeNet-5 convolutional neural network for handwritten digit recognition using the MNIST dataset.

Project Structure

The project is organized as follows:

• main.py: Contains the main script to train and evaluate the LeNet-5 model.
• model.py: Defines the LeNet-5 model architecture.
• utils.py: Provides utility functions for data loading, preprocessing, and evaluation.
• requirements.txt: Lists the required dependencies for running the project.

Usage

1. Prepare the dataset:

• Download the MNIST dataset and place it in the appropriate directory.
• Ensure the dataset is properly split into training and validation sets.

2. Train the model:

• Adjust the data_path argument according to the location of your dataset.
• Optionally, specify other training parameters such as learning rate, batch size, etc.

3. Evaluate the model:

• Adjust the data_path argument according to the location of your dataset.

4. Predict using the trained model:

• Modify the inference() function in utils.py to accept and process the input image.
• Use the modified function to make predictions on new images.

Results

• For a given input image, the model predicted the digit as 8 with a probability of 99%. predicted: 8, prob: 99.9998927116394 %

Acknowledgements

• The MNIST dataset is used for training and evaluation.

🔗 Links