load_mnist_dataset must include train_test_split

[SaashaJoshi]

The train_test_split feature is not currently supported within the load_mnist_dataset. The user will have to call the split before calling the wrapper function.

It will be a better design to include the split within the function such that both the train and test datasets or Dataloaders are retrieved, split, and normalized with one function only. Another option could be to let load_mnist_dataset take an argument that mentions if the user is loading the train or test dataloader. To achieve this, logic inside the wrapper function must be split into two sub-functions or conditionals.

Ideas for better design are welcome.

[Rajesh-1234567]

Hey @SaashaJoshi !

You can solve the issue by either:

Integrating train_test_split within the load_mnist_dataset function, returning both train and test datasets after splitting and normalizing.
Adding an argument to specify whether to load the train or test dataset, with the split and normalization handled accordingly.

[SaashaJoshi]

Hi @Rajesh-1234567,
This is amazing! Feel free to open a PR whenever you are ready or ask any questions by tagging me.

[SaashaJoshi]

Hi @Rajesh-1234567, any questions or updates?

[sohamyedgaonkar]

hey @SaashaJoshi i just wanna know is this issue still open and which all files are included(how can i recreate the bug)

[SaashaJoshi]

Hi @sohamyedgaonkar, Sure go ahead. You can work on this since I haven't received any update on this yet.
This is not a bug per se but the wrapper load_mnist_dataset is a function that handles most of the data cleaning, including normalization and bathing. I was hoping we could also have a train_test_split in the wrapper as well so that a user doesn't have to split the dataset separately. Let me know if you have any questions.

[sohamyedgaonkar]

hey @SaashaJoshi which files do you want to integrate it in ? (so that i can directly change some lines and pull a request

[SaashaJoshi]

Hi, @sohamyedgaonkar Sorry for not looking at this message before. I would want you to integrate test_train_split as an argument in the load_mnist_dataset fucntion. I also mentioned this in the PR. Let me know if you would need any more help!