TrainingOptions.fromJson constructor

TrainingOptions.fromJson(Map json_)

Implementation

TrainingOptions.fromJson(core.Map json_)
    : this(
        activationFn: json_['activationFn'] as core.String?,
        adjustStepChanges: json_['adjustStepChanges'] as core.bool?,
        approxGlobalFeatureContrib:
            json_['approxGlobalFeatureContrib'] as core.bool?,
        autoArima: json_['autoArima'] as core.bool?,
        autoArimaMaxOrder: json_['autoArimaMaxOrder'] as core.String?,
        autoArimaMinOrder: json_['autoArimaMinOrder'] as core.String?,
        autoClassWeights: json_['autoClassWeights'] as core.bool?,
        batchSize: json_['batchSize'] as core.String?,
        boosterType: json_['boosterType'] as core.String?,
        budgetHours: (json_['budgetHours'] as core.num?)?.toDouble(),
        calculatePValues: json_['calculatePValues'] as core.bool?,
        categoryEncodingMethod:
            json_['categoryEncodingMethod'] as core.String?,
        cleanSpikesAndDips: json_['cleanSpikesAndDips'] as core.bool?,
        colorSpace: json_['colorSpace'] as core.String?,
        colsampleBylevel:
            (json_['colsampleBylevel'] as core.num?)?.toDouble(),
        colsampleBynode: (json_['colsampleBynode'] as core.num?)?.toDouble(),
        colsampleBytree: (json_['colsampleBytree'] as core.num?)?.toDouble(),
        contributionMetric: json_['contributionMetric'] as core.String?,
        dartNormalizeType: json_['dartNormalizeType'] as core.String?,
        dataFrequency: json_['dataFrequency'] as core.String?,
        dataSplitColumn: json_['dataSplitColumn'] as core.String?,
        dataSplitEvalFraction:
            (json_['dataSplitEvalFraction'] as core.num?)?.toDouble(),
        dataSplitMethod: json_['dataSplitMethod'] as core.String?,
        decomposeTimeSeries: json_['decomposeTimeSeries'] as core.bool?,
        dimensionIdColumns: (json_['dimensionIdColumns'] as core.List?)
            ?.map((value) => value as core.String)
            .toList(),
        distanceType: json_['distanceType'] as core.String?,
        dropout: (json_['dropout'] as core.num?)?.toDouble(),
        earlyStop: json_['earlyStop'] as core.bool?,
        enableGlobalExplain: json_['enableGlobalExplain'] as core.bool?,
        feedbackType: json_['feedbackType'] as core.String?,
        fitIntercept: json_['fitIntercept'] as core.bool?,
        forecastLimitLowerBound:
            (json_['forecastLimitLowerBound'] as core.num?)?.toDouble(),
        forecastLimitUpperBound:
            (json_['forecastLimitUpperBound'] as core.num?)?.toDouble(),
        hiddenUnits: (json_['hiddenUnits'] as core.List?)
            ?.map((value) => value as core.String)
            .toList(),
        holidayRegion: json_['holidayRegion'] as core.String?,
        holidayRegions: (json_['holidayRegions'] as core.List?)
            ?.map((value) => value as core.String)
            .toList(),
        horizon: json_['horizon'] as core.String?,
        hparamTuningObjectives:
            (json_['hparamTuningObjectives'] as core.List?)
                ?.map((value) => value as core.String)
                .toList(),
        includeDrift: json_['includeDrift'] as core.bool?,
        initialLearnRate:
            (json_['initialLearnRate'] as core.num?)?.toDouble(),
        inputLabelColumns: (json_['inputLabelColumns'] as core.List?)
            ?.map((value) => value as core.String)
            .toList(),
        instanceWeightColumn: json_['instanceWeightColumn'] as core.String?,
        integratedGradientsNumSteps:
            json_['integratedGradientsNumSteps'] as core.String?,
        isTestColumn: json_['isTestColumn'] as core.String?,
        itemColumn: json_['itemColumn'] as core.String?,
        kmeansInitializationColumn:
            json_['kmeansInitializationColumn'] as core.String?,
        kmeansInitializationMethod:
            json_['kmeansInitializationMethod'] as core.String?,
        l1RegActivation: (json_['l1RegActivation'] as core.num?)?.toDouble(),
        l1Regularization:
            (json_['l1Regularization'] as core.num?)?.toDouble(),
        l2Regularization:
            (json_['l2Regularization'] as core.num?)?.toDouble(),
        labelClassWeights: (json_['labelClassWeights']
                as core.Map<core.String, core.dynamic>?)
            ?.map(
          (key, value) => core.MapEntry(
            key,
            (value as core.num).toDouble(),
          ),
        ),
        learnRate: (json_['learnRate'] as core.num?)?.toDouble(),
        learnRateStrategy: json_['learnRateStrategy'] as core.String?,
        lossType: json_['lossType'] as core.String?,
        maxIterations: json_['maxIterations'] as core.String?,
        maxParallelTrials: json_['maxParallelTrials'] as core.String?,
        maxTimeSeriesLength: json_['maxTimeSeriesLength'] as core.String?,
        maxTreeDepth: json_['maxTreeDepth'] as core.String?,
        minAprioriSupport:
            (json_['minAprioriSupport'] as core.num?)?.toDouble(),
        minRelativeProgress:
            (json_['minRelativeProgress'] as core.num?)?.toDouble(),
        minSplitLoss: (json_['minSplitLoss'] as core.num?)?.toDouble(),
        minTimeSeriesLength: json_['minTimeSeriesLength'] as core.String?,
        minTreeChildWeight: json_['minTreeChildWeight'] as core.String?,
        modelRegistry: json_['modelRegistry'] as core.String?,
        modelUri: json_['modelUri'] as core.String?,
        nonSeasonalOrder: json_.containsKey('nonSeasonalOrder')
            ? ArimaOrder.fromJson(json_['nonSeasonalOrder']
                as core.Map<core.String, core.dynamic>)
            : null,
        numClusters: json_['numClusters'] as core.String?,
        numFactors: json_['numFactors'] as core.String?,
        numParallelTree: json_['numParallelTree'] as core.String?,
        numPrincipalComponents:
            json_['numPrincipalComponents'] as core.String?,
        numTrials: json_['numTrials'] as core.String?,
        optimizationStrategy: json_['optimizationStrategy'] as core.String?,
        optimizer: json_['optimizer'] as core.String?,
        pcaExplainedVarianceRatio:
            (json_['pcaExplainedVarianceRatio'] as core.num?)?.toDouble(),
        pcaSolver: json_['pcaSolver'] as core.String?,
        sampledShapleyNumPaths:
            json_['sampledShapleyNumPaths'] as core.String?,
        scaleFeatures: json_['scaleFeatures'] as core.bool?,
        standardizeFeatures: json_['standardizeFeatures'] as core.bool?,
        subsample: (json_['subsample'] as core.num?)?.toDouble(),
        tfVersion: json_['tfVersion'] as core.String?,
        timeSeriesDataColumn: json_['timeSeriesDataColumn'] as core.String?,
        timeSeriesIdColumn: json_['timeSeriesIdColumn'] as core.String?,
        timeSeriesIdColumns: (json_['timeSeriesIdColumns'] as core.List?)
            ?.map((value) => value as core.String)
            .toList(),
        timeSeriesLengthFraction:
            (json_['timeSeriesLengthFraction'] as core.num?)?.toDouble(),
        timeSeriesTimestampColumn:
            json_['timeSeriesTimestampColumn'] as core.String?,
        treeMethod: json_['treeMethod'] as core.String?,
        trendSmoothingWindowSize:
            json_['trendSmoothingWindowSize'] as core.String?,
        userColumn: json_['userColumn'] as core.String?,
        vertexAiModelVersionAliases:
            (json_['vertexAiModelVersionAliases'] as core.List?)
                ?.map((value) => value as core.String)
                .toList(),
        walsAlpha: (json_['walsAlpha'] as core.num?)?.toDouble(),
        warmStart: json_['warmStart'] as core.bool?,
        xgboostVersion: json_['xgboostVersion'] as core.String?,
      );