Refactor training loop from script to class

[sweep-nightly]

Description

This PR refactors the training loop from a script to a class in order to improve code organization and reusability. The existing code in src/main.py is encapsulated into a new class named MNISTTrainer, which includes methods for loading and preprocessing the data, defining the model, training the model, and saving the trained model.

Summary of Changes

• Created a new class named MNISTTrainer in src/main.py.
• Moved the code for loading and preprocessing the data into a method named load_data.
• Moved the code for defining the model into a method named define_model.
• Created a new method named train_model to implement the training loop.
• Created a new method named save_model to save the trained model.
• Updated the script to create an instance of MNISTTrainer and call the methods in the correct order.
• Added proper documentation with docstrings and comments for all methods.

Please review and merge this PR. Thank you!

Fixes #6.

---

🎉 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/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.