Add tests for MNIST data loader and update data loading functionality

[ketayon]

This PR addresses issue #101 by adding train_test_split functionality within the load_mnist_dataset function. Now, users no longer need to manually split the MNIST dataset into training and testing sets before loading.

Changes:

1. Added train-test splitting based on the split_ratio argument, allowing for flexible dataset sizes.
2. Updated the function to allow loading only the train or test dataloader, or both, enhancing usability.
3. Added comprehensive tests in test_mnist_data_loader.py to ensure the functionality of the load_mnist_dataset function, including:

• Verifying that the dataset splits correctly according to the specified split_ratio.
• Testing the functionality of loading train and test dataloaders independently or together.
• Checking that the data normalization functionality works as intended.

Let me know if any further changes are required.

[SaashaJoshi]

Hi @ketayon, thank you for the contribution. I will get back to reviewing as soon as I can. However, are you in a rush to get this done before hacktoberfest gets over?

[ketayon]

Hello, no, I don’t take part in hacktoberfest.On 22 Oct 2024, at 20:49, Saasha Joshi ***@***.***> wrote:Hi @ketayon, thank you for the contribution. I will get back to reviewing as soon as I can. However, are you in a rush to get this done before hacktoberfest gets over?