Data

This module contains wrappers for pandas.DataFrame data for estimators.

Storage

This is wrapper for pandas.DataFrame, which allows you to define dataset for estimator by a simple way.

class rep.data.storage.LabeledDataStorage(data, target=None, sample_weight=None, random_state=None, shuffle=False)

Bases: object

This class implements interface of data for estimators training. It contains data, labels and weights - all information to train model.

Parameters:

• ds (pandas.DataFrame) – data
• target (None or numbers.Number or array-like) – labels for classification and values for regression (set None for predict methods)
• sample_weight (None or numbers.Number or array-like) – weight (set None for predict methods)
• random_state (None or int or RandomState) – for pseudo random generator
• shuffle (bool) – shuffle or not data

col(index)

Get necessary columns

Parameters:index (None or str or list(str)) – names Return type:pandas.Series or pandas.DataFrame

eval_column(expression)

Evaluate some expression to get necessary data

Return type:numpy.array

get_data(features=None)

Get data for estimator

Parameters:features (None or list[str]) – set of feature names (if None then use all features in data storage) Return type:pandas.DataFrame

get_indices()

Get data indices

Return type:numpy.array

get_targets()

Get sample targets for estimator

Return type:numpy.array

get_weights(allow_nones=False)

Get sample weights for estimator

Return type:numpy.array