DenseFeatures

class rectools.dataset.features.DenseFeatures(values: ndarray, names: Tuple[str, ...])[source]

Bases: object

Storage for dense features.

Dense features are represented as a dense matrix, where rows correspond to objects, columns - to features.

Usually you do not need to create this object directly, use from_dataframe class method instead. If you want to use custom logic, use from_iterables class method instead of direct creation.

Parameters

values (np.ndarray) – Matrix of feature values in the classic format: rows - objects, columns - features.
names (tuple(str)) – Names of features (number of names must be equal to the number of columns in values).

Inherited-members

Methods

`from_dataframe`(df, id_map[, id_col])	Create DenseFeatures object from dataframe.
`from_iterables`(values, names)	Create class instance from any iterables of feature values and names.
`get_dense`()	Return values in dense format.
`get_sparse`()	Return values in sparse format.
`take`(ids)	Take a subset of features for given subject (user or item) ids.

Attributes

`values`
`names`

classmethod from_dataframe(df: DataFrame, id_map: IdMap, id_col: str = 'id') → DenseFeatures[source]

Create DenseFeatures object from dataframe.

Assume that feature values are values in dataframe, and feature names are column names.

Parameters

df (pd.Dataframe) – Table in classic format: rows corresponds to objects, columns - to features. One special column id_col must contain object external ids.
id_map (IdMap) – Mapping between external and internal ids. Sets of ids in id_map and in df must be equal.
id_col (str, default id) – Name of column containing object ids.

Return type

DenseFeatures

classmethod from_iterables(values: Iterable[Iterable[float]], names: Iterable[str]) → DenseFeatures[source]

Create class instance from any iterables of feature values and names.

Parameters

values (iterable(iterable(float))) – Feature values matrix. E.g. list of lists: [[1, 2, 3], [4, 5, 6]].
names (iterable(str)) – Feature names.

Return type

DenseFeatures

get_dense() → ndarray[source]

Return values in dense format.

Return type: ndarray

get_sparse() → csr_matrix[source]

Return values in sparse format.

Return type: csr_matrix

take(ids: Union[Sequence[int], ndarray]) → DenseFeatures[source]

Take a subset of features for given subject (user or item) ids.

Parameters: ids (array-like) – Array of internal ids to select features for.
Return type: DenseFeatures