Attributes

• dataSplitColumn (String.t): 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 Defaults to: null.
• dataSplitEvalFraction (float()): The fraction of evaluation data over the whole input data. The rest of data will be used as training data. The format should be double. Accurate to two decimal places. Default value is 0.2. Defaults to: null.

• dataSplitMethod (String.t): The data split type for training and evaluation, e.g. RANDOM. Defaults to: null.

  • Enum - one of [DATA_SPLIT_METHOD_UNSPECIFIED, RANDOM, CUSTOM, SEQUENTIAL, NO_SPLIT, AUTO_SPLIT]

• distanceType (String.t): [Beta] Distance type for clustering models. Defaults to: null.

  • Enum - one of [DISTANCE_TYPE_UNSPECIFIED, EUCLIDEAN, COSINE]
• earlyStop (boolean()): Whether to stop early when the loss doesn't improve significantly any more (compared to min_relative_progress). Used only for iterative training algorithms. Defaults to: null.
• initialLearnRate (float()): Specifies the initial learning rate for the line search learn rate strategy. Defaults to: null.
• inputLabelColumns ([String.t]): Name of input label columns in training data. Defaults to: null.
• l1Regularization (float()): L1 regularization coefficient. Defaults to: null.
• l2Regularization (float()): L2 regularization coefficient. Defaults to: null.
• labelClassWeights (%{optional(String.t) => float()}): Weights associated with each label class, for rebalancing the training data. Only applicable for classification models. Defaults to: null.
• learnRate (float()): Learning rate in training. Used only for iterative training algorithms. Defaults to: null.

• learnRateStrategy (String.t): The strategy to determine learn rate for the current iteration. Defaults to: null.

  • Enum - one of [LEARN_RATE_STRATEGY_UNSPECIFIED, LINE_SEARCH, CONSTANT]

• lossType (String.t): Type of loss function used during training run. Defaults to: null.

  • Enum - one of [LOSS_TYPE_UNSPECIFIED, MEAN_SQUARED_LOSS, MEAN_LOG_LOSS]
• maxIterations (String.t): The maximum number of iterations in training. Used only for iterative training algorithms. Defaults to: null.
• minRelativeProgress (float()): When early_stop is true, stops training when accuracy improvement is less than 'min_relative_progress'. Used only for iterative training algorithms. Defaults to: null.
• modelUri (String.t): [Beta] Google Cloud Storage URI from which the model was imported. Only applicable for imported models. Defaults to: null.
• numClusters (String.t): [Beta] Number of clusters for clustering models. Defaults to: null.

• optimizationStrategy (String.t): Optimization strategy for training linear regression models. Defaults to: null.

  • Enum - one of [OPTIMIZATION_STRATEGY_UNSPECIFIED, BATCH_GRADIENT_DESCENT, NORMAL_EQUATION]
• warmStart (boolean()): Whether to train a model from the last checkpoint. Defaults to: null.