com.google.api.services.bigquery.model.TrainingOptions Maven / Gradle / Ivy
/*
* Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except
* in compliance with the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software distributed under the License
* is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express
* or implied. See the License for the specific language governing permissions and limitations under
* the License.
*/
/*
* This code was generated by https://github.com/googleapis/google-api-java-client-services/
* Modify at your own risk.
*/
package com.google.api.services.bigquery.model;
/**
* Options used in model training.
*
* This is the Java data model class that specifies how to parse/serialize into the JSON that is
* transmitted over HTTP when working with the BigQuery API. For a detailed explanation see:
* https://developers.google.com/api-client-library/java/google-http-java-client/json
*
*
* @author Google, Inc.
*/
@SuppressWarnings("javadoc")
public final class TrainingOptions extends com.google.api.client.json.GenericJson {
/**
* Whether to enable auto ARIMA or not.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Boolean autoArima;
/**
* The max value of non-seasonal p and q.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key @com.google.api.client.json.JsonString
private java.lang.Long autoArimaMaxOrder;
/**
* Batch size for dnn models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key @com.google.api.client.json.JsonString
private java.lang.Long batchSize;
/**
* The data frequency of a time series.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String dataFrequency;
/**
* 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
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String dataSplitColumn;
/**
* 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.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Double dataSplitEvalFraction;
/**
* The data split type for training and evaluation, e.g. RANDOM.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String dataSplitMethod;
/**
* Distance type for clustering models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String distanceType;
/**
* Dropout probability for dnn models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Double dropout;
/**
* 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 value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Boolean earlyStop;
/**
* Feedback type that specifies which algorithm to run for matrix factorization.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String feedbackType;
/**
* Hidden units for dnn models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key @com.google.api.client.json.JsonString
private java.util.List hiddenUnits;
/**
* The geographical region based on which the holidays are considered in time series modeling. If
* a valid value is specified, then holiday effects modeling is enabled.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String holidayRegion;
/**
* The number of periods ahead that need to be forecasted.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key @com.google.api.client.json.JsonString
private java.lang.Long horizon;
/**
* Include drift when fitting an ARIMA model.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Boolean includeDrift;
/**
* Specifies the initial learning rate for the line search learn rate strategy.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Double initialLearnRate;
/**
* Name of input label columns in training data.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.util.List inputLabelColumns;
/**
* Item column specified for matrix factorization models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String itemColumn;
/**
* The column used to provide the initial centroids for kmeans algorithm when
* kmeans_initialization_method is CUSTOM.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String kmeansInitializationColumn;
/**
* The method used to initialize the centroids for kmeans algorithm.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String kmeansInitializationMethod;
/**
* L1 regularization coefficient.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Double l1Regularization;
/**
* L2 regularization coefficient.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Double l2Regularization;
/**
* Weights associated with each label class, for rebalancing the training data. Only applicable
* for classification models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.util.Map labelClassWeights;
/**
* Learning rate in training. Used only for iterative training algorithms.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Double learnRate;
/**
* The strategy to determine learn rate for the current iteration.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String learnRateStrategy;
/**
* Type of loss function used during training run.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String lossType;
/**
* The maximum number of iterations in training. Used only for iterative training algorithms.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key @com.google.api.client.json.JsonString
private java.lang.Long maxIterations;
/**
* Maximum depth of a tree for boosted tree models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key @com.google.api.client.json.JsonString
private java.lang.Long maxTreeDepth;
/**
* When early_stop is true, stops training when accuracy improvement is less than
* 'min_relative_progress'. Used only for iterative training algorithms.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Double minRelativeProgress;
/**
* Minimum split loss for boosted tree models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Double minSplitLoss;
/**
* Google Cloud Storage URI from which the model was imported. Only applicable for imported
* models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String modelUri;
/**
* A specification of the non-seasonal part of the ARIMA model: the three components (p, d, q) are
* the AR order, the degree of differencing, and the MA order.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private ArimaOrder nonSeasonalOrder;
/**
* Number of clusters for clustering models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key @com.google.api.client.json.JsonString
private java.lang.Long numClusters;
/**
* Num factors specified for matrix factorization models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key @com.google.api.client.json.JsonString
private java.lang.Long numFactors;
/**
* Optimization strategy for training linear regression models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String optimizationStrategy;
/**
* Whether to preserve the input structs in output feature names. Suppose there is a struct A with
* field b. When false (default), the output feature name is A_b. When true, the output feature
* name is A.b.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Boolean preserveInputStructs;
/**
* Subsample fraction of the training data to grow tree to prevent overfitting for boosted tree
* models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Double subsample;
/**
* Column to be designated as time series data for ARIMA model.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String timeSeriesDataColumn;
/**
* The time series id column that was used during ARIMA model training.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String timeSeriesIdColumn;
/**
* Column to be designated as time series timestamp for ARIMA model.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String timeSeriesTimestampColumn;
/**
* User column specified for matrix factorization models.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.String userColumn;
/**
* Hyperparameter for matrix factoration when implicit feedback type is specified.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Double walsAlpha;
/**
* Whether to train a model from the last checkpoint.
* The value may be {@code null}.
*/
@com.google.api.client.util.Key
private java.lang.Boolean warmStart;
/**
* Whether to enable auto ARIMA or not.
* @return value or {@code null} for none
*/
public java.lang.Boolean getAutoArima() {
return autoArima;
}
/**
* Whether to enable auto ARIMA or not.
* @param autoArima autoArima or {@code null} for none
*/
public TrainingOptions setAutoArima(java.lang.Boolean autoArima) {
this.autoArima = autoArima;
return this;
}
/**
* The max value of non-seasonal p and q.
* @return value or {@code null} for none
*/
public java.lang.Long getAutoArimaMaxOrder() {
return autoArimaMaxOrder;
}
/**
* The max value of non-seasonal p and q.
* @param autoArimaMaxOrder autoArimaMaxOrder or {@code null} for none
*/
public TrainingOptions setAutoArimaMaxOrder(java.lang.Long autoArimaMaxOrder) {
this.autoArimaMaxOrder = autoArimaMaxOrder;
return this;
}
/**
* Batch size for dnn models.
* @return value or {@code null} for none
*/
public java.lang.Long getBatchSize() {
return batchSize;
}
/**
* Batch size for dnn models.
* @param batchSize batchSize or {@code null} for none
*/
public TrainingOptions setBatchSize(java.lang.Long batchSize) {
this.batchSize = batchSize;
return this;
}
/**
* The data frequency of a time series.
* @return value or {@code null} for none
*/
public java.lang.String getDataFrequency() {
return dataFrequency;
}
/**
* The data frequency of a time series.
* @param dataFrequency dataFrequency or {@code null} for none
*/
public TrainingOptions setDataFrequency(java.lang.String dataFrequency) {
this.dataFrequency = dataFrequency;
return this;
}
/**
* 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
* @return value or {@code null} for none
*/
public java.lang.String getDataSplitColumn() {
return dataSplitColumn;
}
/**
* 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
* @param dataSplitColumn dataSplitColumn or {@code null} for none
*/
public TrainingOptions setDataSplitColumn(java.lang.String dataSplitColumn) {
this.dataSplitColumn = dataSplitColumn;
return this;
}
/**
* 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.
* @return value or {@code null} for none
*/
public java.lang.Double getDataSplitEvalFraction() {
return dataSplitEvalFraction;
}
/**
* 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.
* @param dataSplitEvalFraction dataSplitEvalFraction or {@code null} for none
*/
public TrainingOptions setDataSplitEvalFraction(java.lang.Double dataSplitEvalFraction) {
this.dataSplitEvalFraction = dataSplitEvalFraction;
return this;
}
/**
* The data split type for training and evaluation, e.g. RANDOM.
* @return value or {@code null} for none
*/
public java.lang.String getDataSplitMethod() {
return dataSplitMethod;
}
/**
* The data split type for training and evaluation, e.g. RANDOM.
* @param dataSplitMethod dataSplitMethod or {@code null} for none
*/
public TrainingOptions setDataSplitMethod(java.lang.String dataSplitMethod) {
this.dataSplitMethod = dataSplitMethod;
return this;
}
/**
* Distance type for clustering models.
* @return value or {@code null} for none
*/
public java.lang.String getDistanceType() {
return distanceType;
}
/**
* Distance type for clustering models.
* @param distanceType distanceType or {@code null} for none
*/
public TrainingOptions setDistanceType(java.lang.String distanceType) {
this.distanceType = distanceType;
return this;
}
/**
* Dropout probability for dnn models.
* @return value or {@code null} for none
*/
public java.lang.Double getDropout() {
return dropout;
}
/**
* Dropout probability for dnn models.
* @param dropout dropout or {@code null} for none
*/
public TrainingOptions setDropout(java.lang.Double dropout) {
this.dropout = dropout;
return this;
}
/**
* Whether to stop early when the loss doesn't improve significantly any more (compared to
* min_relative_progress). Used only for iterative training algorithms.
* @return value or {@code null} for none
*/
public java.lang.Boolean getEarlyStop() {
return earlyStop;
}
/**
* Whether to stop early when the loss doesn't improve significantly any more (compared to
* min_relative_progress). Used only for iterative training algorithms.
* @param earlyStop earlyStop or {@code null} for none
*/
public TrainingOptions setEarlyStop(java.lang.Boolean earlyStop) {
this.earlyStop = earlyStop;
return this;
}
/**
* Feedback type that specifies which algorithm to run for matrix factorization.
* @return value or {@code null} for none
*/
public java.lang.String getFeedbackType() {
return feedbackType;
}
/**
* Feedback type that specifies which algorithm to run for matrix factorization.
* @param feedbackType feedbackType or {@code null} for none
*/
public TrainingOptions setFeedbackType(java.lang.String feedbackType) {
this.feedbackType = feedbackType;
return this;
}
/**
* Hidden units for dnn models.
* @return value or {@code null} for none
*/
public java.util.List getHiddenUnits() {
return hiddenUnits;
}
/**
* Hidden units for dnn models.
* @param hiddenUnits hiddenUnits or {@code null} for none
*/
public TrainingOptions setHiddenUnits(java.util.List hiddenUnits) {
this.hiddenUnits = hiddenUnits;
return this;
}
/**
* The geographical region based on which the holidays are considered in time series modeling. If
* a valid value is specified, then holiday effects modeling is enabled.
* @return value or {@code null} for none
*/
public java.lang.String getHolidayRegion() {
return holidayRegion;
}
/**
* The geographical region based on which the holidays are considered in time series modeling. If
* a valid value is specified, then holiday effects modeling is enabled.
* @param holidayRegion holidayRegion or {@code null} for none
*/
public TrainingOptions setHolidayRegion(java.lang.String holidayRegion) {
this.holidayRegion = holidayRegion;
return this;
}
/**
* The number of periods ahead that need to be forecasted.
* @return value or {@code null} for none
*/
public java.lang.Long getHorizon() {
return horizon;
}
/**
* The number of periods ahead that need to be forecasted.
* @param horizon horizon or {@code null} for none
*/
public TrainingOptions setHorizon(java.lang.Long horizon) {
this.horizon = horizon;
return this;
}
/**
* Include drift when fitting an ARIMA model.
* @return value or {@code null} for none
*/
public java.lang.Boolean getIncludeDrift() {
return includeDrift;
}
/**
* Include drift when fitting an ARIMA model.
* @param includeDrift includeDrift or {@code null} for none
*/
public TrainingOptions setIncludeDrift(java.lang.Boolean includeDrift) {
this.includeDrift = includeDrift;
return this;
}
/**
* Specifies the initial learning rate for the line search learn rate strategy.
* @return value or {@code null} for none
*/
public java.lang.Double getInitialLearnRate() {
return initialLearnRate;
}
/**
* Specifies the initial learning rate for the line search learn rate strategy.
* @param initialLearnRate initialLearnRate or {@code null} for none
*/
public TrainingOptions setInitialLearnRate(java.lang.Double initialLearnRate) {
this.initialLearnRate = initialLearnRate;
return this;
}
/**
* Name of input label columns in training data.
* @return value or {@code null} for none
*/
public java.util.List getInputLabelColumns() {
return inputLabelColumns;
}
/**
* Name of input label columns in training data.
* @param inputLabelColumns inputLabelColumns or {@code null} for none
*/
public TrainingOptions setInputLabelColumns(java.util.List inputLabelColumns) {
this.inputLabelColumns = inputLabelColumns;
return this;
}
/**
* Item column specified for matrix factorization models.
* @return value or {@code null} for none
*/
public java.lang.String getItemColumn() {
return itemColumn;
}
/**
* Item column specified for matrix factorization models.
* @param itemColumn itemColumn or {@code null} for none
*/
public TrainingOptions setItemColumn(java.lang.String itemColumn) {
this.itemColumn = itemColumn;
return this;
}
/**
* The column used to provide the initial centroids for kmeans algorithm when
* kmeans_initialization_method is CUSTOM.
* @return value or {@code null} for none
*/
public java.lang.String getKmeansInitializationColumn() {
return kmeansInitializationColumn;
}
/**
* The column used to provide the initial centroids for kmeans algorithm when
* kmeans_initialization_method is CUSTOM.
* @param kmeansInitializationColumn kmeansInitializationColumn or {@code null} for none
*/
public TrainingOptions setKmeansInitializationColumn(java.lang.String kmeansInitializationColumn) {
this.kmeansInitializationColumn = kmeansInitializationColumn;
return this;
}
/**
* The method used to initialize the centroids for kmeans algorithm.
* @return value or {@code null} for none
*/
public java.lang.String getKmeansInitializationMethod() {
return kmeansInitializationMethod;
}
/**
* The method used to initialize the centroids for kmeans algorithm.
* @param kmeansInitializationMethod kmeansInitializationMethod or {@code null} for none
*/
public TrainingOptions setKmeansInitializationMethod(java.lang.String kmeansInitializationMethod) {
this.kmeansInitializationMethod = kmeansInitializationMethod;
return this;
}
/**
* L1 regularization coefficient.
* @return value or {@code null} for none
*/
public java.lang.Double getL1Regularization() {
return l1Regularization;
}
/**
* L1 regularization coefficient.
* @param l1Regularization l1Regularization or {@code null} for none
*/
public TrainingOptions setL1Regularization(java.lang.Double l1Regularization) {
this.l1Regularization = l1Regularization;
return this;
}
/**
* L2 regularization coefficient.
* @return value or {@code null} for none
*/
public java.lang.Double getL2Regularization() {
return l2Regularization;
}
/**
* L2 regularization coefficient.
* @param l2Regularization l2Regularization or {@code null} for none
*/
public TrainingOptions setL2Regularization(java.lang.Double l2Regularization) {
this.l2Regularization = l2Regularization;
return this;
}
/**
* Weights associated with each label class, for rebalancing the training data. Only applicable
* for classification models.
* @return value or {@code null} for none
*/
public java.util.Map getLabelClassWeights() {
return labelClassWeights;
}
/**
* Weights associated with each label class, for rebalancing the training data. Only applicable
* for classification models.
* @param labelClassWeights labelClassWeights or {@code null} for none
*/
public TrainingOptions setLabelClassWeights(java.util.Map labelClassWeights) {
this.labelClassWeights = labelClassWeights;
return this;
}
/**
* Learning rate in training. Used only for iterative training algorithms.
* @return value or {@code null} for none
*/
public java.lang.Double getLearnRate() {
return learnRate;
}
/**
* Learning rate in training. Used only for iterative training algorithms.
* @param learnRate learnRate or {@code null} for none
*/
public TrainingOptions setLearnRate(java.lang.Double learnRate) {
this.learnRate = learnRate;
return this;
}
/**
* The strategy to determine learn rate for the current iteration.
* @return value or {@code null} for none
*/
public java.lang.String getLearnRateStrategy() {
return learnRateStrategy;
}
/**
* The strategy to determine learn rate for the current iteration.
* @param learnRateStrategy learnRateStrategy or {@code null} for none
*/
public TrainingOptions setLearnRateStrategy(java.lang.String learnRateStrategy) {
this.learnRateStrategy = learnRateStrategy;
return this;
}
/**
* Type of loss function used during training run.
* @return value or {@code null} for none
*/
public java.lang.String getLossType() {
return lossType;
}
/**
* Type of loss function used during training run.
* @param lossType lossType or {@code null} for none
*/
public TrainingOptions setLossType(java.lang.String lossType) {
this.lossType = lossType;
return this;
}
/**
* The maximum number of iterations in training. Used only for iterative training algorithms.
* @return value or {@code null} for none
*/
public java.lang.Long getMaxIterations() {
return maxIterations;
}
/**
* The maximum number of iterations in training. Used only for iterative training algorithms.
* @param maxIterations maxIterations or {@code null} for none
*/
public TrainingOptions setMaxIterations(java.lang.Long maxIterations) {
this.maxIterations = maxIterations;
return this;
}
/**
* Maximum depth of a tree for boosted tree models.
* @return value or {@code null} for none
*/
public java.lang.Long getMaxTreeDepth() {
return maxTreeDepth;
}
/**
* Maximum depth of a tree for boosted tree models.
* @param maxTreeDepth maxTreeDepth or {@code null} for none
*/
public TrainingOptions setMaxTreeDepth(java.lang.Long maxTreeDepth) {
this.maxTreeDepth = maxTreeDepth;
return this;
}
/**
* When early_stop is true, stops training when accuracy improvement is less than
* 'min_relative_progress'. Used only for iterative training algorithms.
* @return value or {@code null} for none
*/
public java.lang.Double getMinRelativeProgress() {
return minRelativeProgress;
}
/**
* When early_stop is true, stops training when accuracy improvement is less than
* 'min_relative_progress'. Used only for iterative training algorithms.
* @param minRelativeProgress minRelativeProgress or {@code null} for none
*/
public TrainingOptions setMinRelativeProgress(java.lang.Double minRelativeProgress) {
this.minRelativeProgress = minRelativeProgress;
return this;
}
/**
* Minimum split loss for boosted tree models.
* @return value or {@code null} for none
*/
public java.lang.Double getMinSplitLoss() {
return minSplitLoss;
}
/**
* Minimum split loss for boosted tree models.
* @param minSplitLoss minSplitLoss or {@code null} for none
*/
public TrainingOptions setMinSplitLoss(java.lang.Double minSplitLoss) {
this.minSplitLoss = minSplitLoss;
return this;
}
/**
* Google Cloud Storage URI from which the model was imported. Only applicable for imported
* models.
* @return value or {@code null} for none
*/
public java.lang.String getModelUri() {
return modelUri;
}
/**
* Google Cloud Storage URI from which the model was imported. Only applicable for imported
* models.
* @param modelUri modelUri or {@code null} for none
*/
public TrainingOptions setModelUri(java.lang.String modelUri) {
this.modelUri = modelUri;
return this;
}
/**
* A specification of the non-seasonal part of the ARIMA model: the three components (p, d, q) are
* the AR order, the degree of differencing, and the MA order.
* @return value or {@code null} for none
*/
public ArimaOrder getNonSeasonalOrder() {
return nonSeasonalOrder;
}
/**
* A specification of the non-seasonal part of the ARIMA model: the three components (p, d, q) are
* the AR order, the degree of differencing, and the MA order.
* @param nonSeasonalOrder nonSeasonalOrder or {@code null} for none
*/
public TrainingOptions setNonSeasonalOrder(ArimaOrder nonSeasonalOrder) {
this.nonSeasonalOrder = nonSeasonalOrder;
return this;
}
/**
* Number of clusters for clustering models.
* @return value or {@code null} for none
*/
public java.lang.Long getNumClusters() {
return numClusters;
}
/**
* Number of clusters for clustering models.
* @param numClusters numClusters or {@code null} for none
*/
public TrainingOptions setNumClusters(java.lang.Long numClusters) {
this.numClusters = numClusters;
return this;
}
/**
* Num factors specified for matrix factorization models.
* @return value or {@code null} for none
*/
public java.lang.Long getNumFactors() {
return numFactors;
}
/**
* Num factors specified for matrix factorization models.
* @param numFactors numFactors or {@code null} for none
*/
public TrainingOptions setNumFactors(java.lang.Long numFactors) {
this.numFactors = numFactors;
return this;
}
/**
* Optimization strategy for training linear regression models.
* @return value or {@code null} for none
*/
public java.lang.String getOptimizationStrategy() {
return optimizationStrategy;
}
/**
* Optimization strategy for training linear regression models.
* @param optimizationStrategy optimizationStrategy or {@code null} for none
*/
public TrainingOptions setOptimizationStrategy(java.lang.String optimizationStrategy) {
this.optimizationStrategy = optimizationStrategy;
return this;
}
/**
* Whether to preserve the input structs in output feature names. Suppose there is a struct A with
* field b. When false (default), the output feature name is A_b. When true, the output feature
* name is A.b.
* @return value or {@code null} for none
*/
public java.lang.Boolean getPreserveInputStructs() {
return preserveInputStructs;
}
/**
* Whether to preserve the input structs in output feature names. Suppose there is a struct A with
* field b. When false (default), the output feature name is A_b. When true, the output feature
* name is A.b.
* @param preserveInputStructs preserveInputStructs or {@code null} for none
*/
public TrainingOptions setPreserveInputStructs(java.lang.Boolean preserveInputStructs) {
this.preserveInputStructs = preserveInputStructs;
return this;
}
/**
* Subsample fraction of the training data to grow tree to prevent overfitting for boosted tree
* models.
* @return value or {@code null} for none
*/
public java.lang.Double getSubsample() {
return subsample;
}
/**
* Subsample fraction of the training data to grow tree to prevent overfitting for boosted tree
* models.
* @param subsample subsample or {@code null} for none
*/
public TrainingOptions setSubsample(java.lang.Double subsample) {
this.subsample = subsample;
return this;
}
/**
* Column to be designated as time series data for ARIMA model.
* @return value or {@code null} for none
*/
public java.lang.String getTimeSeriesDataColumn() {
return timeSeriesDataColumn;
}
/**
* Column to be designated as time series data for ARIMA model.
* @param timeSeriesDataColumn timeSeriesDataColumn or {@code null} for none
*/
public TrainingOptions setTimeSeriesDataColumn(java.lang.String timeSeriesDataColumn) {
this.timeSeriesDataColumn = timeSeriesDataColumn;
return this;
}
/**
* The time series id column that was used during ARIMA model training.
* @return value or {@code null} for none
*/
public java.lang.String getTimeSeriesIdColumn() {
return timeSeriesIdColumn;
}
/**
* The time series id column that was used during ARIMA model training.
* @param timeSeriesIdColumn timeSeriesIdColumn or {@code null} for none
*/
public TrainingOptions setTimeSeriesIdColumn(java.lang.String timeSeriesIdColumn) {
this.timeSeriesIdColumn = timeSeriesIdColumn;
return this;
}
/**
* Column to be designated as time series timestamp for ARIMA model.
* @return value or {@code null} for none
*/
public java.lang.String getTimeSeriesTimestampColumn() {
return timeSeriesTimestampColumn;
}
/**
* Column to be designated as time series timestamp for ARIMA model.
* @param timeSeriesTimestampColumn timeSeriesTimestampColumn or {@code null} for none
*/
public TrainingOptions setTimeSeriesTimestampColumn(java.lang.String timeSeriesTimestampColumn) {
this.timeSeriesTimestampColumn = timeSeriesTimestampColumn;
return this;
}
/**
* User column specified for matrix factorization models.
* @return value or {@code null} for none
*/
public java.lang.String getUserColumn() {
return userColumn;
}
/**
* User column specified for matrix factorization models.
* @param userColumn userColumn or {@code null} for none
*/
public TrainingOptions setUserColumn(java.lang.String userColumn) {
this.userColumn = userColumn;
return this;
}
/**
* Hyperparameter for matrix factoration when implicit feedback type is specified.
* @return value or {@code null} for none
*/
public java.lang.Double getWalsAlpha() {
return walsAlpha;
}
/**
* Hyperparameter for matrix factoration when implicit feedback type is specified.
* @param walsAlpha walsAlpha or {@code null} for none
*/
public TrainingOptions setWalsAlpha(java.lang.Double walsAlpha) {
this.walsAlpha = walsAlpha;
return this;
}
/**
* Whether to train a model from the last checkpoint.
* @return value or {@code null} for none
*/
public java.lang.Boolean getWarmStart() {
return warmStart;
}
/**
* Whether to train a model from the last checkpoint.
* @param warmStart warmStart or {@code null} for none
*/
public TrainingOptions setWarmStart(java.lang.Boolean warmStart) {
this.warmStart = warmStart;
return this;
}
@Override
public TrainingOptions set(String fieldName, Object value) {
return (TrainingOptions) super.set(fieldName, value);
}
@Override
public TrainingOptions clone() {
return (TrainingOptions) super.clone();
}
}
© 2015 - 2025 Weber Informatics LLC | Privacy Policy