Implement MatchMaker model #67
Conversation
| """ | ||
|
|
||
| # The left drug | ||
| hidden_left = torch.cat([context_features, drug_features_left], dim=1) |
There was a problem hiding this comment.
since the code is exactly duplicated between the left and right drug (besides the input), consider splitting this into a helper function
|
Please run |
Codecov Report
@@ Coverage Diff @@
## main #67 +/- ##
==========================================
+ Coverage 93.75% 93.97% +0.22%
==========================================
Files 29 29
Lines 800 846 +46
==========================================
+ Hits 750 795 +45
- Misses 50 51 +1
Continue to review full report at Codecov.
|
chemicalx/models/matchmaker.py
Outdated
| :param out_channels: The number of output channels. | ||
| :param dropout_rate: The rate of dropout before the scoring head is used. | ||
| """ | ||
|
|
There was a problem hiding this comment.
remove blank line (please read CI results carefully or run tox -e flake8 locally)
Co-authored-by: Charles Tapley Hoyt <cthoyt@gmail.com>
|
I'd like to push a few simplifications to this model before merge - like using some nn.Sequentials |
This reverts commit e025146.
Summary
Closes #23
Changes
Implement the MatchMaker model.