Package | Description |
---|---|
com.google.api.services.bigquery.model |
Modifier and Type | Method and Description |
---|---|
TrainingOptions |
TrainingOptions.clone() |
TrainingOptions |
HparamTuningTrial.getHparams()
The hyperprameters selected for this trial.
|
TrainingOptions |
TrainingRun.getTrainingOptions()
Output only.
|
TrainingOptions |
TrainingOptions.set(String fieldName,
Object value) |
TrainingOptions |
TrainingOptions.setAdjustStepChanges(Boolean adjustStepChanges)
If true, detect step changes and make data adjustment in the input time series.
|
TrainingOptions |
TrainingOptions.setAutoArima(Boolean autoArima)
Whether to enable auto ARIMA or not.
|
TrainingOptions |
TrainingOptions.setAutoArimaMaxOrder(Long autoArimaMaxOrder)
The max value of non-seasonal p and q.
|
TrainingOptions |
TrainingOptions.setBatchSize(Long batchSize)
Batch size for dnn models.
|
TrainingOptions |
TrainingOptions.setBoosterType(String boosterType)
Booster type for boosted tree models.
|
TrainingOptions |
TrainingOptions.setCalculatePValues(Boolean calculatePValues)
Whether or not p-value test should be computed for this model.
|
TrainingOptions |
TrainingOptions.setCleanSpikesAndDips(Boolean cleanSpikesAndDips)
If true, clean spikes and dips in the input time series.
|
TrainingOptions |
TrainingOptions.setColorSpace(String colorSpace)
Enums for color space, used for processing images in Object Table.
|
TrainingOptions |
TrainingOptions.setColsampleBylevel(Double colsampleBylevel)
Subsample ratio of columns for each level for boosted tree models.
|
TrainingOptions |
TrainingOptions.setColsampleBynode(Double colsampleBynode)
Subsample ratio of columns for each node(split) for boosted tree models.
|
TrainingOptions |
TrainingOptions.setColsampleBytree(Double colsampleBytree)
Subsample ratio of columns when constructing each tree for boosted tree models.
|
TrainingOptions |
TrainingOptions.setDartNormalizeType(String dartNormalizeType)
Type of normalization algorithm for boosted tree models using dart booster.
|
TrainingOptions |
TrainingOptions.setDataFrequency(String dataFrequency)
The data frequency of a time series.
|
TrainingOptions |
TrainingOptions.setDataSplitColumn(String dataSplitColumn)
The column to split data with.
|
TrainingOptions |
TrainingOptions.setDataSplitEvalFraction(Double dataSplitEvalFraction)
The fraction of evaluation data over the whole input data.
|
TrainingOptions |
TrainingOptions.setDataSplitMethod(String dataSplitMethod)
The data split type for training and evaluation, e.g.
|
TrainingOptions |
TrainingOptions.setDecomposeTimeSeries(Boolean decomposeTimeSeries)
If true, perform decompose time series and save the results.
|
TrainingOptions |
TrainingOptions.setDistanceType(String distanceType)
Distance type for clustering models.
|
TrainingOptions |
TrainingOptions.setDropout(Double dropout)
Dropout probability for dnn models.
|
TrainingOptions |
TrainingOptions.setEarlyStop(Boolean earlyStop)
Whether to stop early when the loss doesn't improve significantly any more (compared to
min_relative_progress).
|
TrainingOptions |
TrainingOptions.setEnableGlobalExplain(Boolean enableGlobalExplain)
If true, enable global explanation during training.
|
TrainingOptions |
TrainingOptions.setFeedbackType(String feedbackType)
Feedback type that specifies which algorithm to run for matrix factorization.
|
TrainingOptions |
TrainingOptions.setHiddenUnits(List<Long> hiddenUnits)
Hidden units for dnn models.
|
TrainingOptions |
TrainingOptions.setHolidayRegion(String holidayRegion)
The geographical region based on which the holidays are considered in time series modeling.
|
TrainingOptions |
TrainingOptions.setHorizon(Long horizon)
The number of periods ahead that need to be forecasted.
|
TrainingOptions |
TrainingOptions.setHparamTuningObjectives(List<String> hparamTuningObjectives)
The target evaluation metrics to optimize the hyperparameters for.
|
TrainingOptions |
TrainingOptions.setIncludeDrift(Boolean includeDrift)
Include drift when fitting an ARIMA model.
|
TrainingOptions |
TrainingOptions.setInitialLearnRate(Double initialLearnRate)
Specifies the initial learning rate for the line search learn rate strategy.
|
TrainingOptions |
TrainingOptions.setInputLabelColumns(List<String> inputLabelColumns)
Name of input label columns in training data.
|
TrainingOptions |
TrainingOptions.setIntegratedGradientsNumSteps(Long integratedGradientsNumSteps)
Number of integral steps for the integrated gradients explain method.
|
TrainingOptions |
TrainingOptions.setItemColumn(String itemColumn)
Item column specified for matrix factorization models.
|
TrainingOptions |
TrainingOptions.setKmeansInitializationColumn(String kmeansInitializationColumn)
The column used to provide the initial centroids for kmeans algorithm when
kmeans_initialization_method is CUSTOM.
|
TrainingOptions |
TrainingOptions.setKmeansInitializationMethod(String kmeansInitializationMethod)
The method used to initialize the centroids for kmeans algorithm.
|
TrainingOptions |
TrainingOptions.setL1Regularization(Double l1Regularization)
L1 regularization coefficient.
|
TrainingOptions |
TrainingOptions.setL2Regularization(Double l2Regularization)
L2 regularization coefficient.
|
TrainingOptions |
TrainingOptions.setLabelClassWeights(Map<String,Double> labelClassWeights)
Weights associated with each label class, for rebalancing the training data.
|
TrainingOptions |
TrainingOptions.setLearnRate(Double learnRate)
Learning rate in training.
|
TrainingOptions |
TrainingOptions.setLearnRateStrategy(String learnRateStrategy)
The strategy to determine learn rate for the current iteration.
|
TrainingOptions |
TrainingOptions.setLossType(String lossType)
Type of loss function used during training run.
|
TrainingOptions |
TrainingOptions.setMaxIterations(Long maxIterations)
The maximum number of iterations in training.
|
TrainingOptions |
TrainingOptions.setMaxParallelTrials(Long maxParallelTrials)
Maximum number of trials to run in parallel.
|
TrainingOptions |
TrainingOptions.setMaxTimeSeriesLength(Long maxTimeSeriesLength)
Get truncated length by last n points in time series.
|
TrainingOptions |
TrainingOptions.setMaxTreeDepth(Long maxTreeDepth)
Maximum depth of a tree for boosted tree models.
|
TrainingOptions |
TrainingOptions.setMinRelativeProgress(Double minRelativeProgress)
When early_stop is true, stops training when accuracy improvement is less than
'min_relative_progress'.
|
TrainingOptions |
TrainingOptions.setMinSplitLoss(Double minSplitLoss)
Minimum split loss for boosted tree models.
|
TrainingOptions |
TrainingOptions.setMinTimeSeriesLength(Long minTimeSeriesLength)
Set fast trend ARIMA_PLUS model minimum training length.
|
TrainingOptions |
TrainingOptions.setMinTreeChildWeight(Long minTreeChildWeight)
Minimum sum of instance weight needed in a child for boosted tree models.
|
TrainingOptions |
TrainingOptions.setModelUri(String modelUri)
Google Cloud Storage URI from which the model was imported.
|
TrainingOptions |
TrainingOptions.setNonSeasonalOrder(ArimaOrder 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.
|
TrainingOptions |
TrainingOptions.setNumClusters(Long numClusters)
Number of clusters for clustering models.
|
TrainingOptions |
TrainingOptions.setNumFactors(Long numFactors)
Num factors specified for matrix factorization models.
|
TrainingOptions |
TrainingOptions.setNumParallelTree(Long numParallelTree)
Number of parallel trees constructed during each iteration for boosted tree models.
|
TrainingOptions |
TrainingOptions.setNumTrials(Long numTrials)
Number of trials to run this hyperparameter tuning job.
|
TrainingOptions |
TrainingOptions.setOptimizationStrategy(String optimizationStrategy)
Optimization strategy for training linear regression models.
|
TrainingOptions |
TrainingOptions.setPreserveInputStructs(Boolean preserveInputStructs)
Whether to preserve the input structs in output feature names.
|
TrainingOptions |
TrainingOptions.setSampledShapleyNumPaths(Long sampledShapleyNumPaths)
Number of paths for the sampled Shapley explain method.
|
TrainingOptions |
TrainingOptions.setSubsample(Double subsample)
Subsample fraction of the training data to grow tree to prevent overfitting for boosted tree
models.
|
TrainingOptions |
TrainingOptions.setTimeSeriesDataColumn(String timeSeriesDataColumn)
Column to be designated as time series data for ARIMA model.
|
TrainingOptions |
TrainingOptions.setTimeSeriesIdColumn(String timeSeriesIdColumn)
The time series id column that was used during ARIMA model training.
|
TrainingOptions |
TrainingOptions.setTimeSeriesIdColumns(List<String> timeSeriesIdColumns)
The time series id columns that were used during ARIMA model training.
|
TrainingOptions |
TrainingOptions.setTimeSeriesLengthFraction(Double timeSeriesLengthFraction)
Get truncated length by fraction in time series.
|
TrainingOptions |
TrainingOptions.setTimeSeriesTimestampColumn(String timeSeriesTimestampColumn)
Column to be designated as time series timestamp for ARIMA model.
|
TrainingOptions |
TrainingOptions.setTreeMethod(String treeMethod)
Tree construction algorithm for boosted tree models.
|
TrainingOptions |
TrainingOptions.setTrendSmoothingWindowSize(Long trendSmoothingWindowSize)
The smoothing window size for the trend component of the time series.
|
TrainingOptions |
TrainingOptions.setUserColumn(String userColumn)
User column specified for matrix factorization models.
|
TrainingOptions |
TrainingOptions.setWalsAlpha(Double walsAlpha)
Hyperparameter for matrix factoration when implicit feedback type is specified.
|
TrainingOptions |
TrainingOptions.setWarmStart(Boolean warmStart)
Whether to train a model from the last checkpoint.
|
Modifier and Type | Method and Description |
---|---|
HparamTuningTrial |
HparamTuningTrial.setHparams(TrainingOptions hparams)
The hyperprameters selected for this trial.
|
TrainingRun |
TrainingRun.setTrainingOptions(TrainingOptions trainingOptions)
Output only.
|
Copyright © 2011–2022 Google. All rights reserved.