Variable naming convention in DeepChem

[arunppsg]

To represent number of, DeepChem library uses various conventions:

• nb_epochs to represent number of epochs in model.fit method
• num_tasks / n_tasks to represent number of tasks for a model
• num_classes / n_classes to represent number of classes

It will be helpful if we could agree on a standard convention to follow and deprecate others. Often, its confusing whether to use num_tasks or n_tasks - both as a developer and user of DeepChem.

Similarly, there are a couple of other instances which I am listing down here:

• hidden_dim / hidden_size
• emb_dim / embedding_dim

[harshit4311]

I would love working on this issue, could you please assign this to me?

[harshit4311]

I would love working on this issue, could you please assign this to me?

I believe that adopting whole variable names like num_tasks instead of abbreviations such as n_tasks would be highly beneficial for clarity and consistency throughout the library. Here are a few reasons why: 1.) Readability. 2.) Consistency 3.) Documentation

By embracing whole variable names like num_tasks, we can streamline communication and promote a more user-friendly experience for everyone involved with DeepChem. Let's work together to implement these standardised naming conventions for better code readability and usability.