Adding Usage Examples to Splitters

[GreatRSingh]

Pull Request Template

Description

Fix #3230

This PR aims to add example documentation for Splitters.

• RandomSplitter - Already Done
• RandomGroupSplitter
• RandomStratifiedSplitter
• SingletaskStratifiedSplitter - Already Done
• IndexSplitter
• SpecifiedSplitter
• MolecularWeightSplitter
• MaxMinSplitter
• ButinaSplitter
• FingerprintSplitter
• ScaffoldSplitter - Already Done

Type of change

Please check the option that is related to your PR.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
  • In this case, we recommend to discuss your modification on GitHub issues before creating the PR
• Documentations (modification for documents)

Checklist

• My code follows the style guidelines of this project
  • Run yapf -i <modified file> and check no errors (yapf version must be 0.32.0)
  • Run mypy -p deepchem and check no errors
  • Run flake8 <modified file> --count and check no errors
  • Run python -m doctest <modified file> and check no errors
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• I have added tests that prove my fix is effective or that my feature works
• New unit tests pass locally with my changes
• I have checked my code and corrected any misspellings

[GreatRSingh]

@gusty1g sir, check the example for MolecularWeightSplitter.

[arunppsg]

A splitter.split method will return indices for split and not datasets. Instead, if we use train_dataset, test_dataset = splitter.train_test_split(dataset), it will be more correct.

Also, the example should demonstrate how the splitter works. I would rewrite it something like this:

n_samples = 5
n_features = 2
X = np.random.rand(n_samples, n_features)
y = np.random.rand(n_samples)
splitter = dc.splits.IndexSplitter()
dataset = dc.data.NumpyDataset(X, y)

>>> train_dataset, test_dataset = splitter.train_test_split(dataset)

>>> print(train_dataset.ids)
[0, 1, 2]
>>> print (test_dataset.ids)
[3, 4]

The above example demonstrates how the splitter splits by index.

[GreatRSingh]

I mistook split function for something else.
I fixed that in the latest commit.

[arunppsg]

Same as comment for IndexSplitter.

[GreatRSingh]

@gusty1g can you suggest example for RandomGroupSplitter. I am having difficulty in making example for it.
In the group parameter.

[arunppsg]

@gusty1g can you suggest example for RandomGroupSplitter. I am having difficulty in making example for it. In the group parameter.

import deepchem as dc
import numpy as np

groups = [0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3]
splitter = dc.splits.RandomGroupSplitter(groups=groups)

dataset = dc.data.NumpyDataset(X=np.arange(12))   # 12 elements
train, test = splitter.train_test_split(dataset, frac_train=0.75, seed=0)

print (train.ids)  # array([6, 7, 8, 9, 10, 11, 3, 4, 5], dtype=object) 

In above example, note that the values 6, 7, 8 belong to a group (2), 9, 10, 11 belong to another group (3) and 3 4 5 belong to group 1. The splitter preserves the group (i.e a group goes either into train or test) while splitting the dataset.