Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
This PR adds the
GNINAModelEnsemble
class, which allow to combine different pre-trained GNINA models into a single class. Related to #33 and #35.This PR also adds functions to easily load an ensemble of pre-trained models using their names, and to perform inference from the CLI. For the time being, the CLI requires a list of protein-ligand pairs in a file, instead of the protein and ligand directly as in GNINA.
The class assumes that the models perform both pose prediction and binding affinity prediction (which is the case for the pre-trained GNINA models, hence the GNINA name). A less strict class to be used with custom models (pose prediction only, or pose and flexible residues) will follow.
The
dense
model is not yet available since it appears to be problematic and it is under investigation. This means that the GNINAdefault
model is not yet available.PR Checklist