bigframes.ml.decomposition.PCA#

class bigframes.ml.decomposition.PCA(n_components: int | float | None = None, *, svd_solver: Literal['full', 'randomized', 'auto'] = 'auto')[source]#

Principal component analysis (PCA).

Examples:

>>> import bigframes.pandas as bpd
>>> from bigframes.ml.decomposition import PCA
>>> X = bpd.DataFrame({"feat0": [-1, -2, -3, 1, 2, 3], "feat1": [-1, -1, -2, 1, 1, 2]})
>>> pca = PCA(n_components=2).fit(X)
>>> pca.predict(X) 
    principal_component_1  principal_component_2
0              -0.755243               0.157628
1               -1.05405              -0.141179
2              -1.809292               0.016449
3               0.755243              -0.157628
4                1.05405               0.141179
5               1.809292              -0.016449

[6 rows x 2 columns]
>>> pca.explained_variance_ratio_ 
    principal_component_id  explained_variance_ratio
0                       1                   0.00901
1                       0                   0.99099

[2 rows x 2 columns]

Parameters:

• n_components (int, float or None, default None) Number of components to keep. If n_components is not set, all components are kept, n_components = min(n_samples, n_features). If 0 < n_components < 1, select the number of components such that the amount of variance that needs to be explained is greater than the percentage specified by n_components.

• svd_solver ("full", "randomized" or "auto", default "auto") The solver to use to calculate the principal components. Details: https://cloud.google.com/bigquery/docs/reference/standard-sql/bigqueryml-syntax-create-pca#pca_solver.

property components_: DataFrame#

Principal axes in feature space, representing the directions of maximum variance in the data.

Returns:

DataFrame of principal components, containing following columns:

principal_component_id: An integer that identifies the principal component.

feature: The column name that contains the feature.

numerical_value: If feature is numeric, the value of feature for the principal component that principal_component_id identifies. If feature isnt numeric, the value is NULL.

categorical_value: A list of mappings containing information about categorical features. Each mapping contains the following fields:

categorical_value.category: The name of each category.

categorical_value.value: The value of categorical_value.category for the centroid that centroid_id identifies.

The output contains one row per feature per component.

Return type:

bigframes.dataframe.DataFrame

detect_anomalies(X: DataFrame | Series | DataFrame | Series, *, contamination: float = 0.1) → DataFrame[source]#

Detect the anomaly data points of the input.

Parameters:

• X (bigframes.dataframe.DataFrame or bigframes.series.Series) Series or a DataFrame to detect anomalies.

• contamination (float, default 0.1) Identifies the proportion of anomalies in the training dataset that are used to create the model. The value must be in the range [0, 0.5].

Returns:

detected DataFrame.

Return type:

bigframes.dataframe.DataFrame

property explained_variance_: DataFrame#

The amount of variance explained by each of the selected components.

Returns:

DataFrame containing following columns:

principal_component_id: An integer that identifies the principal component.

explained_variance: The factor by which the eigenvector is scaled. Eigenvalue and explained variance are the same concepts in PCA.

Return type:

bigframes.dataframe.DataFrame

property explained_variance_ratio_: DataFrame#

Percentage of variance explained by each of the selected components.

Returns:

DataFrame containing following columns:

principal_component_id: An integer that identifies the principal component.

explained_variance_ratio: the total variance is the sum of variances, also known as eigenvalues, of all of the individual principal components. The explained variance ratio by a principal component is the ratio between the variance, also known as eigenvalue, of that principal component and the total variance.

Return type:

bigframes.dataframe.DataFrame

predict(X: DataFrame | Series | DataFrame | Series) → DataFrame[source]#

Predict the closest cluster for each sample in X.

Parameters:

X (bigframes.dataframe.DataFrame or bigframes.series.Series or pandas.core.frame.DataFrame or pandas.core.series.Series) Series or a DataFrame to predict.

Returns:

Predicted DataFrames.

Return type:

bigframes.dataframe.DataFrame

score(X=None, y=None) → DataFrame[source]#

Calculate evaluation metrics of the model.

Note

Output matches that of the BigQuery ML.EVALUATE function. See: https://cloud.google.com/bigquery/docs/reference/standard-sql/bigqueryml-syntax-evaluate#pca_models for the outputs relevant to this model type.

Parameters:

• X (default None) Ignored.

• y (default None) Ignored.

Returns:

DataFrame that represents model metrics.

Return type:

bigframes.dataframe.DataFrame

to_gbq(model_name: str, replace: bool = False) → PCA[source]#

Save the model to BigQuery.

Parameters:

• model_name (str) The name of the model.

• replace (bool, default False) Determine whether to replace if the model already exists. Default to False.

Returns:

Saved model.

Return type:

PCA