bigframes.ml.preprocessing.OneHotEncoder#

class bigframes.ml.preprocessing.OneHotEncoder(drop: Literal['most_frequent'] | None = None, min_frequency: int | None = None, max_categories: int | None = None)[source]#

Encode categorical features as a one-hot format.

The input to this transformer should be an array-like of integers or strings, denoting the values taken on by categorical (discrete) features. The features are encoded using a one-hot (aka ‘one-of-K’ or ‘dummy’) encoding scheme.

Note that this method deviates from Scikit-Learn; instead of producing sparse binary columns, the encoding is a single column of STRUCT<index INT64, value DOUBLE>.

Examples:

Given a dataset with two features, we let the encoder find the unique values per feature and transform the data to a binary one-hot encoding.

>>> from bigframes.ml.preprocessing import OneHotEncoder
>>> import bigframes.pandas as bpd

>>> enc = OneHotEncoder()
>>> X = bpd.DataFrame({"a": ["Male", "Female", "Female"], "b": ["1", "3", "2"]})
>>> enc.fit(X)
OneHotEncoder()

>>> print(enc.transform(bpd.DataFrame({"a": ["Female", "Male"], "b": ["1", "4"]})))
                onehotencoded_a               onehotencoded_b
0  [{'index': 1, 'value': 1.0}]  [{'index': 1, 'value': 1.0}]
1  [{'index': 2, 'value': 1.0}]  [{'index': 0, 'value': 1.0}]

[2 rows x 2 columns]
Parameters:

  • drop (Optional[Literal["most_frequent"]], default None) – Specifies a methodology to use to drop one of the categories per feature. This is useful in situations where perfectly collinear features cause problems, such as when feeding the resulting data into an unregularized linear regression model. However, dropping one category breaks the symmetry of the original representation and can therefore induce a bias in downstream models, for instance for penalized linear classification or regression models. Default None: retain all the categories. “most_frequent”: Drop the most frequent category found in the string expression. Selecting this value causes the function to use dummy encoding.

  • min_frequency (Optional[int], default None) – Specifies the minimum frequency below which a category will be considered infrequent. Default None. int: categories with a smaller cardinality will be considered infrequent as index 0.

  • max_categories (Optional[int], default None) – Specifies an upper limit to the number of output features for each input feature when considering infrequent categories. If there are infrequent categories, max_categories includes the category representing the infrequent categories along with the frequent categories. Default None. Set limit to 1,000,000.

Attributes

FREQUENCY_THRESHOLD_DEFAULT

TOP_K_DEFAULT

Methods

__init__([drop, min_frequency, max_categories])

fit(X[, y])

Fit OneHotEncoder to X.

fit_transform(X[, y])

get_params([deep])

Get parameters for this estimator.

to_gbq(model_name[, replace])

Save the transformer as a BigQuery model.

transform(X)

Transform X using one-hot encoding.
On this page

This Page