Project Name: MNIST Classification

Description

This project focuses on training and evaluating neural networks for handwritten digit classification using the MNIST dataset. The code is organized into three main files: S5.ipynb, model.py, and utils.py.

---

S5.ipynb

This Jupyter Notebook contains the main code for the MNIST classification task. The notebook is structured as follows:

1. Imports and Setup

   • Import necessary libraries and modules.
   • Check for CUDA availability.
   • Define data transformations and create data loaders.

2. Data Visualization

   • Display sample images from the training dataset.

3. Model Summary

   • Utilize torchsummary to display the summary of the neural network model (Net and Net2).

4. Model Training

   • Train the Net model using SGD optimizer, learning rate scheduler, and negative log-likelihood loss.
   • Print training and testing accuracies over epochs.

5. Visualize Training Progress

   • Plot training and testing losses, as well as training and testing accuracies.

---

model.py

This file contains the definition of the neural network models used for the MNIST classification task:

1. Net Class

   • Convolutional Neural Network with four convolutional layers and two fully connected layers.
   • Implements the forward pass with ReLU activation and log-softmax output.

2. Net2 Class

   • Similar to Net but with bias terms in convolutional and fully connected layers set to False.

---

utils.py

This file includes utility functions and data transformations:

1. Data Transformations

   • Defines training and testing data transformations using torchvision.transforms.

2. Neural Network Training Functions

   • GetCorrectPredCount: Returns the count of correct predictions.
   • train: Performs training of the neural network.
   • test: Evaluates the neural network on the test dataset.

3. Graph Plotting Function

   • allgraphs: Plots training and testing losses, as well as training and testing accuracies.

---

How to Use

1. Open and run the cells in S5.ipynb sequentially to train and evaluate the model.
2. Do not forget to add 'model.py' and 'utils.py' as auxillary files onto the Colab environment

---

Notes

• Ensure that required libraries (torch, torchvision, matplotlib, tqdm) are installed.
• Make sure to have GPU support for faster training (if available).