Add CNN class for MNIST classification and import it in main.py

[sweep-nightly]

Description

This PR adds a new CNN class for handling MNIST classification in the cnn.py file. The CNN class is designed to replace the existing feed-forward neural network (Net class) in main.py. The CNN architecture includes convolutional layers, pooling layers, and fully connected layers, making it suitable for image classification tasks like MNIST.

Summary of Changes

• Created a new file src/cnn.py to define the CNN class.
• Added the CNN class in cnn.py, which inherits from torch.nn.Module and includes the necessary layers for the CNN architecture.
• Updated the main.py file to import the CNN class from cnn.py and use it for training.
• Adjusted the training loop in main.py to work with the new CNN model.
• Added docstrings and comments to the CNN class and its methods for clarity and maintainability.

Please review and merge this PR to incorporate the CNN class for improved MNIST classification.

Fixes #9.

---

🎉 Latest improvements to Sweep:

• Sweep can now passively improve your repository! Check out Rules to learn more.

---

💡 To get Sweep to edit this pull request, you can:

• Comment below, and Sweep can edit the entire PR
• Comment on a file, Sweep will only modify the commented file
• Edit the original issue to get Sweep to recreate the PR from scratch

[sweep-nightly]

Rollback Files For Sweep

• Rollback changes to src/cnn.py
• Rollback changes to src/cnn.py
• Rollback changes to src/main.py

[sweep-nightly]

Apply Sweep Rules to your PR?

• Apply: All docstrings and comments should be up to date.
• Apply: Code should be properly formatted and indented.
• Apply: Variable and function names should be descriptive and follow a consistent naming convention.
• Apply: Imports should be organized and grouped together.
• Apply: There should be no unused imports or variables.
• Apply: Code should be properly commented and include docstrings for functions and classes.