Skip to content

sklearn Transformers

molfp.sklearn

scikit-learn compatible transformer classes for all fingerprint types.

Each transformer implements fit/transform/fit_transform and get_params/set_params, making fingerprints usable in sklearn Pipelines, GridSearchCV, cross_val_score, etc.

For feature selection, use the standalone functions after transforming: X = molfp.ECFPTransformer().fit_transform(smiles) X_clean = molfp.remove_low_variance(X, threshold=0.01)["X"] result = molfp.select_features(X, y, k=500)

MACCSTransformer

Bases: _BaseFingerprintTransformer

MACCS Keys (166 bits) sklearn transformer.

PubChemTransformer

Bases: _BaseFingerprintTransformer

PubChem/CACTVS (881 bits) sklearn transformer.

SubstructureTransformer

Bases: _BaseFingerprintTransformer

SubstructureFP (307 bits) sklearn transformer.

SubstructureCountTransformer

Bases: _BaseFingerprintTransformer

Substructure Count (307 features) sklearn transformer.

KlekotaRothTransformer

Bases: _BaseFingerprintTransformer

Klekota-Roth (4860 bits) sklearn transformer.

KlekotaRothCountTransformer

Bases: _BaseFingerprintTransformer

Klekota-Roth Count (4860 features) sklearn transformer.

EStateTransformer

Bases: _BaseFingerprintTransformer

E-State (79 bits) sklearn transformer.

ECFPTransformer

Bases: BaseEstimator, TransformerMixin

ECFP4 / Morgan (configurable bits) sklearn transformer.

FCFPTransformer

Bases: BaseEstimator, TransformerMixin

FCFP4 / Feature Morgan (configurable bits) sklearn transformer.

LingoTransformer

Bases: BaseEstimator, TransformerMixin

Lingo (hashed SMILES 4-grams) sklearn transformer.

ShortestPathTransformer

Bases: BaseEstimator, TransformerMixin

Shortest Path sklearn transformer.

HybridizationTransformer

Bases: BaseEstimator, TransformerMixin

Hybridization fingerprint sklearn transformer.

SignatureTransformer

Bases: BaseEstimator, TransformerMixin

Signature fingerprint sklearn transformer.

AtomPairs2DTransformer

Bases: BaseEstimator, TransformerMixin

AtomPairs2D (configurable bits) sklearn transformer.

AtomPairs2DCountTransformer

Bases: BaseEstimator, TransformerMixin

AtomPairs2D Count (configurable bits) sklearn transformer.

RDKTransformer

Bases: BaseEstimator, TransformerMixin

RDKit path-based (configurable bits) sklearn transformer.

RDKExtTransformer

Bases: BaseEstimator, TransformerMixin

RDKit Extended (configurable bits) sklearn transformer.

RDKGraphOnlyTransformer

Bases: BaseEstimator, TransformerMixin

RDKit GraphOnly (configurable bits) sklearn transformer.

AtomPairs3DTransformer

Bases: BaseEstimator, TransformerMixin

3D Atom Pairs (configurable bits) sklearn transformer.

AtomPairs3DCountTransformer

Bases: BaseEstimator, TransformerMixin

3D Atom Pairs Count (configurable bits) sklearn transformer.

get_transformer(name, **kwargs)

Get a sklearn transformer by fingerprint name.

Example: molfp.get_transformer("ecfp4", n_bits=2048, n_jobs=-1)