astartes package

Subpackages

• astartes.samplers package
  • Subpackages
    • astartes.samplers.extrapolation package
      • Submodules
      • astartes.samplers.extrapolation.dbscan module
      • astartes.samplers.extrapolation.kmeans module
      • astartes.samplers.extrapolation.molecular_weight module
      • astartes.samplers.extrapolation.optisim module
      • astartes.samplers.extrapolation.scaffold module
      • astartes.samplers.extrapolation.sphere_exclusion module
      • astartes.samplers.extrapolation.target_property module
      • astartes.samplers.extrapolation.time_based module
      • Module contents
    • astartes.samplers.interpolation package
      • Submodules
      • astartes.samplers.interpolation.kennardstone module
      • astartes.samplers.interpolation.random_split module
      • astartes.samplers.interpolation.spxy module
      • Module contents
  • Submodules
  • astartes.samplers.abstract_sampler module
    • AbstractSampler
      • AbstractSampler.__init__()
      • AbstractSampler.get_clusters()
      • AbstractSampler.get_config()
      • AbstractSampler.get_sample_idxs()
      • AbstractSampler.get_sorted_cluster_counter()
  • Module contents
• astartes.utils package
  • Submodules
  • astartes.utils.aimsim_featurizer module
    • Molecule
      • Molecule.mol_graph
      • Molecule.mol_text
      • Molecule.mol_property_val
      • Molecule.descriptor
      • Molecule.get_descriptor_val()
      • Molecule.match_fingerprint_from()
      • Molecule.get_similarity_to()
      • Molecule.get_name()
      • Molecule.get_mol_property_val()
      • Molecule.draw()
      • Molecule.is_same()
      • Molecule.__init__()
      • Molecule.draw()
      • Molecule.get_descriptor_val()
      • Molecule.get_mol_property_val()
      • Molecule.get_name()
      • Molecule.get_similarity_to()
      • Molecule.is_same()
      • Molecule.match_fingerprint_from()
      • Molecule.set_descriptor()
    • featurize_molecules()
  • astartes.utils.array_type_helpers module
    • convert_to_array()
    • panda_handla()
    • return_helper()
  • astartes.utils.exceptions module
    • InvalidConfigurationError
      • InvalidConfigurationError.__init__()
    • InvalidModelTypeError
      • InvalidModelTypeError.__init__()
    • MoleculesNotInstalledError
      • MoleculesNotInstalledError.__init__()
    • SamplerNotImplementedError
      • SamplerNotImplementedError.__init__()
    • UncastableInputError
      • UncastableInputError.__init__()
  • astartes.utils.fast_kennard_stone module
    • fast_kennard_stone()
  • astartes.utils.sampler_factory module
    • SamplerFactory
      • SamplerFactory.__init__()
      • SamplerFactory.get_sampler()
  • astartes.utils.user_utils module
    • display_results_as_table()
    • generate_regression_results_dict()
  • astartes.utils.warnings module
    • ConversionWarning
      • ConversionWarning.__init__()
    • ImperfectSplittingWarning
      • ImperfectSplittingWarning.__init__()
    • NoMatchingScaffold
      • NoMatchingScaffold.__init__()
    • NormalizationWarning
      • NormalizationWarning.__init__()
  • Module contents
    • generate_regression_results_dict()

Submodules

astartes.main module

astartes.main.train_test_split(X: array, y: array | None = None, labels: array | None = None, train_size: float = 0.75, test_size: float | None = None, sampler: str = 'random', random_state: int | None = None, hopts: dict = {}, return_indices: bool = False)

Deterministic train_test_splitting of arbitrary arrays.

Parameters:

• X (np.array) – Numpy array of feature vectors.

• y (np.array, optional) – Targets corresponding to X, must be of same size. Defaults to None.

• labels (np.array, optional) – Labels corresponding to X, must be of same size. Defaults to None.

• train_size (float, optional) – Fraction of dataset to use in training set. Defaults to 0.75.

• test_size (float, optional) – Fraction of dataset to use in test set. Defaults to None.

• sampler (str, optional) – Sampler to use, see IMPLEMENTED_INTER/EXTRAPOLATION_SAMPLERS. Defaults to “random”.

• random_state (int, optional) – The random seed used throughout astartes.

• hopts (dict, optional) – Hyperparameters for the sampler used above. Defaults to {}.

• return_indices (bool, optional) – True to return indices of train/test instead of values. Defaults to False.

Returns:

X, y, and labels train/val/test data, or indices.

Return type:

np.array

astartes.main.train_val_test_split(X: array | DataFrame, y: array | Series | None = None, labels: array | Series | None = None, train_size: float = 0.8, val_size: float = 0.1, test_size: float = 0.1, sampler: str = 'random', random_state: int | None = None, hopts: dict = {}, return_indices: bool = False)

Deterministic train_test_splitting of arbitrary arrays.

Parameters:

• X (np.array, pd.DataFrame) – Numpy array or pandas DataFrame of feature vectors.

• y (np.array, pd.Series, optional) – Targets corresponding to X, must be of same size. Defaults to None.

• labels (np.array, pd.Series, optional) – Labels corresponding to X, must be of same size. Defaults to None.

• train_size (float, optional) – Fraction of dataset to use in training set. Defaults to 0.8.

• val_size (float, optional) – Fraction of dataset to use in validation set. Defaults to 0.1.

• test_size (float, optional) – Fraction of dataset to use in test set. Defaults to 0.1.

• sampler (str, optional) – Sampler to use, see IMPLEMENTED_INTER/EXTRAPOLATION_SAMPLERS. Defaults to “random”.

• random_state (int, optional) – The random seed used throughout astartes.

• hopts (dict, optional) – Hyperparameters for the sampler used above. Defaults to {}.

• return_indices (bool, optional) – True to return indices of train/test after values. Defaults to False.

Returns:

X, y, and labels train/val/test data, or indices.

Return type:

np.array(s)

astartes.molecules module

astartes.molecules.train_test_split_molecules(molecules: array, y: array | None = None, labels: array | None = None, train_size: float = 0.75, test_size: float | None = None, sampler: str = 'random', random_state: int | None = None, hopts: dict = {}, fingerprint: str = 'morgan_fingerprint', fprints_hopts: dict = {}, return_indices: bool = False)

Deterministic train/test splitting of molecules (SMILES strings or RDKit objects).

Parameters:

• molecules (np.array) – List of SMILES strings or RDKit molecule objects representing molecules or reactions.

• y (np.array, optional) – Targets corresponding to SMILES, must be of same size. Defaults to None.

• labels (np.array, optional) – Labels corresponding to SMILES, must be of same size. Defaults to None.

• train_size (float, optional) – Fraction of dataset to use in training (test+train~1). Defaults to 0.75.

• test_size (float, optional) – Fraction of dataset to use in test set. Defaults to None.

• sampler (str, optional) – Sampler to use, see IMPLEMENTED_INTER/EXTRAPOLATION_SAMPLERS. Defaults to “random”.

• random_state (int, optional) – The random seed used throughout astartes. Defaults to None.

• hopts (dict, optional) – Hyperparameters for the sampler used above. Defaults to {}.

• fingerprint (str, optional) – Molecular fingerprint to be used from AIMSim. Defaults to “morgan_fingerprint”.

• fprints_hopts (dict, optional) – Hyperparameters for AIMSim featurization. Defaults to {}.

• return_indices (bool, optional) – True to return indices of train/test after the values. Defaults to False.

Returns:

X, y, and labels train/test data, or indices.

Return type:

np.array

astartes.molecules.train_val_test_split_molecules(molecules: array, y: array | None = None, labels: array | None = None, train_size: float = 0.8, val_size: float = 0.1, test_size: float = 0.1, sampler: str = 'random', random_state: int | None = None, hopts: dict = {}, fingerprint: str = 'morgan_fingerprint', fprints_hopts: dict = {}, return_indices: bool = False)

Deterministic train_test_splitting of molecules (SMILES strings or RDKit objects).

Parameters:

• molecules (np.array) – List of SMILES strings or RDKit molecule objects representing molecules or reactions.

• y (np.array, optional) – Targets corresponding to SMILES, must be of same size. Defaults to None.

• labels (np.array, optional) – Labels corresponding to SMILES, must be of same size. Defaults to None.

• train_size (float, optional) – Fraction of dataset to use in training set. Defaults to 0.8.

• val_size (float, optional) – Fraction of dataset to use in validation set. Defaults to 0.1.

• test_size (float, optional) – Fraction of dataset to use in test set. Defaults to 0.1.

• sampler (str, optional) – Sampler to use, see IMPLEMENTED_INTER/EXTRAPOLATION_SAMPLERS. Defaults to “random”.

• random_state (int, optional) – The random seed used throughout astartes. Defaults to 42.

• hopts (dict, optional) – Hyperparameters for the sampler used above. Defaults to {}.

• fingerprint (str, optional) – Molecular fingerprint to be used from AIMSim. Defaults to “morgan_fingerprint”.

• fprints_hopts (dict, optional) – Hyperparameters for AIMSim featurization. Defaults to {}.

• return_indices (bool, optional) – True to return indices of train/test after the values. Defaults to False.

Returns:

X, y, and labels train/val/test data, or indices.

Return type:

np.array

Module contents