Class TrainingRun (1.16.0)

Information about a single training query run for the model.

.. attribute:: training_options

Options that were used for this training run, includes user specified and default options that were used.

Output of each iteration run, results.size() <= max_iterations.

Inheritance

builtins.object > google.protobuf.pyext._message.CMessage > builtins.object > google.protobuf.message.Message > TrainingRun

Classes

IterationResult

Information about a single iteration of the training run.

.. attribute:: index

Index of the iteration, 0 based.

Loss computed on the training data at the end of iteration.

Learn rate used for this iteration.

TrainingOptions

Protocol buffer.

Type of loss function used during training run.

L1 regularization coefficient.

When early_stop is true, stops training when accuracy improvement is less than 'min_relative_progress'. Used only for iterative training algorithms.

Whether to stop early when the loss doesn't improve significantly any more (compared to min_relative_progress). Used only for iterative training algorithms.

The data split type for training and evaluation, e.g. RANDOM.

The column to split data with. This column won't be used as a feature. 1. When data_split_method is CUSTOM, the corresponding column should be boolean. The rows with true value tag are eval data, and the false are training data. 2. When data_split_method is SEQ, the first DATA_SPLIT_EVAL_FRACTION rows (from smallest to largest) in the corresponding column are used as training data, and the rest are eval data. It respects the order in Orderable data types: https://cloud.google.com/bigquery/docs/reference/standard- sql/data-types#data-type-properties

Specifies the initial learning rate for the line search learn rate strategy.

[Beta] Distance type for clustering models.

[Beta] Google Cloud Storage URI from which the model was imported. Only applicable for imported models.