TrainingOptions class Null safety

Options used in model training.

Constructors

TrainingOptions({bool? adjustStepChanges, bool? autoArima, String? autoArimaMaxOrder, String? batchSize, String? boosterType, bool? calculatePValues, bool? cleanSpikesAndDips, String? colorSpace, double? colsampleBylevel, double? colsampleBynode, double? colsampleBytree, String? dartNormalizeType, String? dataFrequency, String? dataSplitColumn, double? dataSplitEvalFraction, String? dataSplitMethod, bool? decomposeTimeSeries, String? distanceType, double? dropout, bool? earlyStop, bool? enableGlobalExplain, String? feedbackType, List<String>? hiddenUnits, String? holidayRegion, String? horizon, List<String>? hparamTuningObjectives, bool? includeDrift, double? initialLearnRate, List<String>? inputLabelColumns, String? integratedGradientsNumSteps, String? itemColumn, String? kmeansInitializationColumn, String? kmeansInitializationMethod, double? l1Regularization, double? l2Regularization, Map<String, double>? labelClassWeights, double? learnRate, String? learnRateStrategy, String? lossType, String? maxIterations, String? maxParallelTrials, String? maxTimeSeriesLength, String? maxTreeDepth, double? minRelativeProgress, double? minSplitLoss, String? minTimeSeriesLength, String? minTreeChildWeight, String? modelUri, ArimaOrder? nonSeasonalOrder, String? numClusters, String? numFactors, String? numParallelTree, String? numTrials, String? optimizationStrategy, bool? preserveInputStructs, String? sampledShapleyNumPaths, double? subsample, String? timeSeriesDataColumn, String? timeSeriesIdColumn, List<String>? timeSeriesIdColumns, double? timeSeriesLengthFraction, String? timeSeriesTimestampColumn, String? treeMethod, String? trendSmoothingWindowSize, String? userColumn, double? walsAlpha, bool? warmStart})
TrainingOptions.fromJson(Map json_)

Properties

adjustStepChanges ↔ bool?
If true, detect step changes and make data adjustment in the input time series.
read / write
autoArima ↔ bool?
Whether to enable auto ARIMA or not.
read / write
autoArimaMaxOrder ↔ String?
The max value of non-seasonal p and q.
read / write
batchSize ↔ String?
Batch size for dnn models.
read / write
boosterType ↔ String?
Booster type for boosted tree models. Possible string values are:
read / write
calculatePValues ↔ bool?
Whether or not p-value test should be computed for this model.
read / write
cleanSpikesAndDips ↔ bool?
If true, clean spikes and dips in the input time series.
read / write
colorSpace ↔ String?
Enums for color space, used for processing images in Object Table.
read / write
colsampleBylevel ↔ double?
Subsample ratio of columns for each level for boosted tree models.
read / write
colsampleBynode ↔ double?
Subsample ratio of columns for each node(split) for boosted tree models.
read / write
colsampleBytree ↔ double?
Subsample ratio of columns when constructing each tree for boosted tree models.
read / write
dartNormalizeType ↔ String?
Type of normalization algorithm for boosted tree models using dart booster. Possible string values are:
read / write
dataFrequency ↔ String?
The data frequency of a time series. Possible string values are:
read / write
dataSplitColumn ↔ String?
The column to split data with.
read / write
dataSplitEvalFraction ↔ double?
The fraction of evaluation data over the whole input data.
read / write
dataSplitMethod ↔ String?
The data split type for training and evaluation, e.g. RANDOM. Possible string values are:
read / write
decomposeTimeSeries ↔ bool?
If true, perform decompose time series and save the results.
read / write
distanceType ↔ String?
Distance type for clustering models. Possible string values are:
read / write
dropout ↔ double?
Dropout probability for dnn models.
read / write
earlyStop ↔ bool?
Whether to stop early when the loss doesn't improve significantly any more (compared to min_relative_progress).
read / write
enableGlobalExplain ↔ bool?
If true, enable global explanation during training.
read / write
feedbackType ↔ String?
Feedback type that specifies which algorithm to run for matrix factorization. Possible string values are:
read / write
hashCode → int
The hash code for this object.
read-onlyinherited
hiddenUnits ↔ List<String>?
Hidden units for dnn models.
read / write
holidayRegion ↔ String?
The geographical region based on which the holidays are considered in time series modeling.
read / write
horizon ↔ String?
The number of periods ahead that need to be forecasted.
read / write
hparamTuningObjectives ↔ List<String>?
The target evaluation metrics to optimize the hyperparameters for.
read / write
includeDrift ↔ bool?
Include drift when fitting an ARIMA model.
read / write
initialLearnRate ↔ double?
Specifies the initial learning rate for the line search learn rate strategy.
read / write
inputLabelColumns ↔ List<String>?
Name of input label columns in training data.
read / write
integratedGradientsNumSteps ↔ String?
Number of integral steps for the integrated gradients explain method.
read / write
itemColumn ↔ String?
Item column specified for matrix factorization models.
read / write
kmeansInitializationColumn ↔ String?
The column used to provide the initial centroids for kmeans algorithm when kmeans_initialization_method is CUSTOM.
read / write
kmeansInitializationMethod ↔ String?
The method used to initialize the centroids for kmeans algorithm. Possible string values are:
read / write
l1Regularization ↔ double?
L1 regularization coefficient.
read / write
l2Regularization ↔ double?
L2 regularization coefficient.
read / write
labelClassWeights ↔ Map<String, double>?
Weights associated with each label class, for rebalancing the training data.
read / write
learnRate ↔ double?
Learning rate in training.
read / write
learnRateStrategy ↔ String?
The strategy to determine learn rate for the current iteration. Possible string values are:
read / write
lossType ↔ String?
Type of loss function used during training run. Possible string values are:
read / write
maxIterations ↔ String?
The maximum number of iterations in training.
read / write
maxParallelTrials ↔ String?
Maximum number of trials to run in parallel.
read / write
maxTimeSeriesLength ↔ String?
Get truncated length by last n points in time series.
read / write
maxTreeDepth ↔ String?
Maximum depth of a tree for boosted tree models.
read / write
minRelativeProgress ↔ double?
When early_stop is true, stops training when accuracy improvement is less than 'min_relative_progress'.
read / write
minSplitLoss ↔ double?
Minimum split loss for boosted tree models.
read / write
minTimeSeriesLength ↔ String?
Set fast trend ARIMA_PLUS model minimum training length.
read / write
minTreeChildWeight ↔ String?
Minimum sum of instance weight needed in a child for boosted tree models.
read / write
modelUri ↔ String?
Google Cloud Storage URI from which the model was imported.
read / write
nonSeasonalOrder ↔ ArimaOrder?
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.
read / write
numClusters ↔ String?
Number of clusters for clustering models.
read / write
numFactors ↔ String?
Num factors specified for matrix factorization models.
read / write
numParallelTree ↔ String?
Number of parallel trees constructed during each iteration for boosted tree models.
read / write
numTrials ↔ String?
Number of trials to run this hyperparameter tuning job.
read / write
optimizationStrategy ↔ String?
Optimization strategy for training linear regression models. Possible string values are:
read / write
preserveInputStructs ↔ bool?
Whether to preserve the input structs in output feature names.
read / write
runtimeType → Type
A representation of the runtime type of the object.
read-onlyinherited
sampledShapleyNumPaths ↔ String?
Number of paths for the sampled Shapley explain method.
read / write
subsample ↔ double?
Subsample fraction of the training data to grow tree to prevent overfitting for boosted tree models.
read / write
timeSeriesDataColumn ↔ String?
Column to be designated as time series data for ARIMA model.
read / write
timeSeriesIdColumn ↔ String?
The time series id column that was used during ARIMA model training.
read / write
timeSeriesIdColumns ↔ List<String>?
The time series id columns that were used during ARIMA model training.
read / write
timeSeriesLengthFraction ↔ double?
Get truncated length by fraction in time series.
read / write
timeSeriesTimestampColumn ↔ String?
Column to be designated as time series timestamp for ARIMA model.
read / write
treeMethod ↔ String?
Tree construction algorithm for boosted tree models. Possible string values are:
read / write
trendSmoothingWindowSize ↔ String?
The smoothing window size for the trend component of the time series.
read / write
userColumn ↔ String?
User column specified for matrix factorization models.
read / write
walsAlpha ↔ double?
Hyperparameter for matrix factoration when implicit feedback type is specified.
read / write
warmStart ↔ bool?
Whether to train a model from the last checkpoint.
read / write

Methods

noSuchMethod(Invocation invocation) → dynamic
Invoked when a non-existent method or property is accessed.
inherited
toJson() → Map<String, dynamic>
toString() → String
A string representation of this object.
inherited

Operators

operator ==(Object other) → bool
The equality operator.
inherited