Fixed issue moleculenet classification data sets not running properly.

[dickeygh]

Features.py:

1. Added lines 54-74 to fix the issue of some SMILES not featurizing properly (this was causing the issues with the classification data sets).
2. Changed lines 110-116 to work for both multi-label (clintox and sider) and single label classification.

This PR fixes the issues that the last PR was having. The issue had something to do with self.task_type and the formatting of if statements. I will be working on making this part of the code neater in a later PR.

I have tested this PR to make sure that it runs properly for both regression and classification. It seems to be running properly.

This PR also adds the new data sets to the development branch.

[pep8speaks]

Hello @dickeygh! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:

• In the file core/features.py:

Line 50:5: E303 too many blank lines (3) Line 50:80: E501 line too long (113 > 79 characters) Line 57:14: E711 comparison to None should be 'if cond is None:'

• In the file main.py:

Line 31:9: E265 block comment should start with '# ' Line 31:9: E303 too many blank lines (2) Line 32:35: E261 at least two spaces before inline comment Line 32:80: E501 line too long (106 > 79 characters) Line 38:9: E265 block comment should start with '# '

Comment last updated at 2020-07-15 17:55:05 UTC