This document describes the ML.PRINCIPAL_COMPONENT_INFO function, which lets you see the statistics of the principal components in a principal component analysis (PCA) model, such as eigenvalue and explained variance ratio.


ML.PRINCIPAL_COMPONENT_INFO(MODEL `project_id.dataset.model`)


ML.PRINCIPAL_COMPONENT_INFO takes the following arguments:

  • project_id: a STRING value that specifies your project ID.
  • dataset: a STRING value that specifies the BigQuery dataset that contains the model.
  • model: a STRING value that specifies the name of the model.


ML.PRINCIPAL_COMPONENT_INFO returns the following columns:

  • principal_component_id: an INT64 that contains the principal component. The table is ordered in descending order of the eigenvalue value.
  • eigenvalue: a FLOAT64 value that contains the factor by which the eigenvector is scaled. Eigenvalue and explained variance are the same concepts in PCA.
  • explained_variance_ratio: a FLOAT64 value that contains the explained variance ratio, which is the ratio between the variance, also known as eigenvalue, of that principal component and the total variance. The total variance is the sum of the variances of all of the individual principal components.
  • cumulative_explained_variance_ratio: a FLOAT64 value that contains the cumulative explained variance ratio of the k-th principal component, which is the sum of the explained variance ratios of all the previous principal components, including the k-th principal component.


The following example retrieves the eigenvalue-related information of each principal component in the model mydataset.mymodel in your default project.


What's next