vsm.model.Lsa

class vsm.model.Lsa(td_matrix=array([], dtype=float64), context_type=None)

Methods

__init__([td_matrix, dtype, context_type])	Initialize Lsa.
load(f)	Loads LSA model data from a numpy archive file with extension
save(f)	Saves model data as a numpy archive file with extension npz.
train([k_factors])	Trains the model.

__init__(td_matrix=array([], dtype=float64), context_type=None)

Initialize Lsa.

Parameters:

• td_matrix (np.array, optional) – Term-Document matrix. Default is an empty array.
• context_type (string, optional) – Name of tokenization whose tokens will be treated as documents. Default is None.

static load(f)

Loads LSA model data from a numpy archive file with extension npz. The expected keys for the component matrices are word_matrix, eigenvalues and doc_matrix.

Parameters:f (str-like or file-like object) – Designates the file from which to load data. See numpy.load for further details. Returns:a saved Lsa model. See Also:numpy.load()

save(f)

Saves model data as a numpy archive file with extension npz. The keys for the component matrices are word_matrix, eigenvalues and doc_matrix.

Parameters:f (str-like or file-like object) – Designates the file to which to save data. See numpy.savez for further details. See Also:numpy.savez()

train(k_factors=300)

Trains the model.

Parameters:k_factors (int, optional) – Default is 300.