rdmc.conformer_generation.solvation#

Modules for including computing solvation corrections.

class rdmc.conformer_generation.solvation.ConfSolv(trained_model_dir: str, track_stats: bool | None = False)#

Bases: Estimator

Class for estimating conformer energies in solution with neural networks.

Parameters:

  • trained_model_dir (str) – The path to the directory storing the trained ConfSolv model.

  • track_stats (bool, optional) – Whether to track timing stats. Defaults to False.

predict_energies(mol_data: List[dict], **kwargs) → List[dict]#

Predict conformer free energies in a given solvent.

Parameters:

mol_data (List[dict]) – A list of molecule dictionaries.

Returns:

mol_data (List[dict]) – A list of molecule dictionaries with energy values updated.

class rdmc.conformer_generation.solvation.Estimator(track_stats: bool | None = False)#

Bases: object

The abstract class for energy estimator.

Parameters:

track_stats (bool, optional) – Whether to track timing stats. Defaults to False.

predict_energies(mol_data: List[dict], **kwargs)#

The abstract method for predicting energies. It will be implemented in actual classes. The method needs to take mol_data which is a dictionary containing info about the conformers of the molecules. It will return the molecule as the same mol_data object with the energy values altered.

Parameters:

mol_data (List[dict]) – A list of molecule dictionaries.

Returns:

mol_data (List[dict]) – A list of molecule dictionaries with energy values updated.