chemicalchecker.core.data.DataFactory

class DataFactory[source]

Bases: object

DataFactory class.

Methods

make_data

Initialize any type of Signature.

signaturize

From matrix to signature.
static make_data(cctype, *args, **kwargs)[source]

Initialize any type of Signature.

Parameters:

  • cctype (str) – the signature type: ‘sign0-3’, ‘clus0-3’, ‘neig0-3’ ‘proj0-3’.

  • args – passed to signature constructor

  • kwargs – passed to signature constructor

static signaturize(cctype, signature_path, matrix, keys=None, dataset_code=None)[source]

From matrix to signature.

Produce a signature-like structure for the given matrix input.

Parameters:

  • signature_path (str) – Destination for the signature.

  • matrix (np.array) – Matrix where row are Molecules and columns are features.

  • keys (np.array) – List of Molecule names. If None incremental keys are used to maintain the original order.

  • dataset_code (str) – The code for the newly generated signature.