File "TrainingOptions.php"
Full Path: /home/palsarh/web/palsarh.in/public_html/vendor/google/apiclient-services/src/Bigquery/TrainingOptions.php
File size: 65.48 KB
MIME-type: text/x-php
Charset: utf-8
<?php
/*
* Copyright 2014 Google Inc.
*
* 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.
*/
namespace Google\Service\Bigquery;
class TrainingOptions extends \Google\Collection
{
/**
* Unspecified booster type.
*/
public const BOOSTER_TYPE_BOOSTER_TYPE_UNSPECIFIED = 'BOOSTER_TYPE_UNSPECIFIED';
/**
* Gbtree booster.
*/
public const BOOSTER_TYPE_GBTREE = 'GBTREE';
/**
* Dart booster.
*/
public const BOOSTER_TYPE_DART = 'DART';
/**
* Unspecified encoding method.
*/
public const CATEGORY_ENCODING_METHOD_ENCODING_METHOD_UNSPECIFIED = 'ENCODING_METHOD_UNSPECIFIED';
/**
* Applies one-hot encoding.
*/
public const CATEGORY_ENCODING_METHOD_ONE_HOT_ENCODING = 'ONE_HOT_ENCODING';
/**
* Applies label encoding.
*/
public const CATEGORY_ENCODING_METHOD_LABEL_ENCODING = 'LABEL_ENCODING';
/**
* Applies dummy encoding.
*/
public const CATEGORY_ENCODING_METHOD_DUMMY_ENCODING = 'DUMMY_ENCODING';
/**
* Unspecified color space
*/
public const COLOR_SPACE_COLOR_SPACE_UNSPECIFIED = 'COLOR_SPACE_UNSPECIFIED';
/**
* RGB
*/
public const COLOR_SPACE_RGB = 'RGB';
/**
* HSV
*/
public const COLOR_SPACE_HSV = 'HSV';
/**
* YIQ
*/
public const COLOR_SPACE_YIQ = 'YIQ';
/**
* YUV
*/
public const COLOR_SPACE_YUV = 'YUV';
/**
* GRAYSCALE
*/
public const COLOR_SPACE_GRAYSCALE = 'GRAYSCALE';
/**
* Unspecified dart normalize type.
*/
public const DART_NORMALIZE_TYPE_DART_NORMALIZE_TYPE_UNSPECIFIED = 'DART_NORMALIZE_TYPE_UNSPECIFIED';
/**
* New trees have the same weight of each of dropped trees.
*/
public const DART_NORMALIZE_TYPE_TREE = 'TREE';
/**
* New trees have the same weight of sum of dropped trees.
*/
public const DART_NORMALIZE_TYPE_FOREST = 'FOREST';
/**
* Default value.
*/
public const DATA_FREQUENCY_DATA_FREQUENCY_UNSPECIFIED = 'DATA_FREQUENCY_UNSPECIFIED';
/**
* Automatically inferred from timestamps.
*/
public const DATA_FREQUENCY_AUTO_FREQUENCY = 'AUTO_FREQUENCY';
/**
* Yearly data.
*/
public const DATA_FREQUENCY_YEARLY = 'YEARLY';
/**
* Quarterly data.
*/
public const DATA_FREQUENCY_QUARTERLY = 'QUARTERLY';
/**
* Monthly data.
*/
public const DATA_FREQUENCY_MONTHLY = 'MONTHLY';
/**
* Weekly data.
*/
public const DATA_FREQUENCY_WEEKLY = 'WEEKLY';
/**
* Daily data.
*/
public const DATA_FREQUENCY_DAILY = 'DAILY';
/**
* Hourly data.
*/
public const DATA_FREQUENCY_HOURLY = 'HOURLY';
/**
* Per-minute data.
*/
public const DATA_FREQUENCY_PER_MINUTE = 'PER_MINUTE';
/**
* Default value.
*/
public const DATA_SPLIT_METHOD_DATA_SPLIT_METHOD_UNSPECIFIED = 'DATA_SPLIT_METHOD_UNSPECIFIED';
/**
* Splits data randomly.
*/
public const DATA_SPLIT_METHOD_RANDOM = 'RANDOM';
/**
* Splits data with the user provided tags.
*/
public const DATA_SPLIT_METHOD_CUSTOM = 'CUSTOM';
/**
* Splits data sequentially.
*/
public const DATA_SPLIT_METHOD_SEQUENTIAL = 'SEQUENTIAL';
/**
* Data split will be skipped.
*/
public const DATA_SPLIT_METHOD_NO_SPLIT = 'NO_SPLIT';
/**
* Splits data automatically: Uses NO_SPLIT if the data size is small.
* Otherwise uses RANDOM.
*/
public const DATA_SPLIT_METHOD_AUTO_SPLIT = 'AUTO_SPLIT';
/**
* Default value.
*/
public const DISTANCE_TYPE_DISTANCE_TYPE_UNSPECIFIED = 'DISTANCE_TYPE_UNSPECIFIED';
/**
* Eculidean distance.
*/
public const DISTANCE_TYPE_EUCLIDEAN = 'EUCLIDEAN';
/**
* Cosine distance.
*/
public const DISTANCE_TYPE_COSINE = 'COSINE';
/**
* Default value.
*/
public const FEEDBACK_TYPE_FEEDBACK_TYPE_UNSPECIFIED = 'FEEDBACK_TYPE_UNSPECIFIED';
/**
* Use weighted-als for implicit feedback problems.
*/
public const FEEDBACK_TYPE_IMPLICIT = 'IMPLICIT';
/**
* Use nonweighted-als for explicit feedback problems.
*/
public const FEEDBACK_TYPE_EXPLICIT = 'EXPLICIT';
/**
* Holiday region unspecified.
*/
public const HOLIDAY_REGION_HOLIDAY_REGION_UNSPECIFIED = 'HOLIDAY_REGION_UNSPECIFIED';
/**
* Global.
*/
public const HOLIDAY_REGION_GLOBAL = 'GLOBAL';
/**
* North America.
*/
public const HOLIDAY_REGION_NA = 'NA';
/**
* Japan and Asia Pacific: Korea, Greater China, India, Australia, and New
* Zealand.
*/
public const HOLIDAY_REGION_JAPAC = 'JAPAC';
/**
* Europe, the Middle East and Africa.
*/
public const HOLIDAY_REGION_EMEA = 'EMEA';
/**
* Latin America and the Caribbean.
*/
public const HOLIDAY_REGION_LAC = 'LAC';
/**
* United Arab Emirates
*/
public const HOLIDAY_REGION_AE = 'AE';
/**
* Argentina
*/
public const HOLIDAY_REGION_AR = 'AR';
/**
* Austria
*/
public const HOLIDAY_REGION_AT = 'AT';
/**
* Australia
*/
public const HOLIDAY_REGION_AU = 'AU';
/**
* Belgium
*/
public const HOLIDAY_REGION_BE = 'BE';
/**
* Brazil
*/
public const HOLIDAY_REGION_BR = 'BR';
/**
* Canada
*/
public const HOLIDAY_REGION_CA = 'CA';
/**
* Switzerland
*/
public const HOLIDAY_REGION_CH = 'CH';
/**
* Chile
*/
public const HOLIDAY_REGION_CL = 'CL';
/**
* China
*/
public const HOLIDAY_REGION_CN = 'CN';
/**
* Colombia
*/
public const HOLIDAY_REGION_CO = 'CO';
/**
* Czechoslovakia
*/
public const HOLIDAY_REGION_CS = 'CS';
/**
* Czech Republic
*/
public const HOLIDAY_REGION_CZ = 'CZ';
/**
* Germany
*/
public const HOLIDAY_REGION_DE = 'DE';
/**
* Denmark
*/
public const HOLIDAY_REGION_DK = 'DK';
/**
* Algeria
*/
public const HOLIDAY_REGION_DZ = 'DZ';
/**
* Ecuador
*/
public const HOLIDAY_REGION_EC = 'EC';
/**
* Estonia
*/
public const HOLIDAY_REGION_EE = 'EE';
/**
* Egypt
*/
public const HOLIDAY_REGION_EG = 'EG';
/**
* Spain
*/
public const HOLIDAY_REGION_ES = 'ES';
/**
* Finland
*/
public const HOLIDAY_REGION_FI = 'FI';
/**
* France
*/
public const HOLIDAY_REGION_FR = 'FR';
/**
* Great Britain (United Kingdom)
*/
public const HOLIDAY_REGION_GB = 'GB';
/**
* Greece
*/
public const HOLIDAY_REGION_GR = 'GR';
/**
* Hong Kong
*/
public const HOLIDAY_REGION_HK = 'HK';
/**
* Hungary
*/
public const HOLIDAY_REGION_HU = 'HU';
/**
* Indonesia
*/
public const HOLIDAY_REGION_ID = 'ID';
/**
* Ireland
*/
public const HOLIDAY_REGION_IE = 'IE';
/**
* Israel
*/
public const HOLIDAY_REGION_IL = 'IL';
/**
* India
*/
public const HOLIDAY_REGION_IN = 'IN';
/**
* Iran
*/
public const HOLIDAY_REGION_IR = 'IR';
/**
* Italy
*/
public const HOLIDAY_REGION_IT = 'IT';
/**
* Japan
*/
public const HOLIDAY_REGION_JP = 'JP';
/**
* Korea (South)
*/
public const HOLIDAY_REGION_KR = 'KR';
/**
* Latvia
*/
public const HOLIDAY_REGION_LV = 'LV';
/**
* Morocco
*/
public const HOLIDAY_REGION_MA = 'MA';
/**
* Mexico
*/
public const HOLIDAY_REGION_MX = 'MX';
/**
* Malaysia
*/
public const HOLIDAY_REGION_MY = 'MY';
/**
* Nigeria
*/
public const HOLIDAY_REGION_NG = 'NG';
/**
* Netherlands
*/
public const HOLIDAY_REGION_NL = 'NL';
/**
* Norway
*/
public const HOLIDAY_REGION_NO = 'NO';
/**
* New Zealand
*/
public const HOLIDAY_REGION_NZ = 'NZ';
/**
* Peru
*/
public const HOLIDAY_REGION_PE = 'PE';
/**
* Philippines
*/
public const HOLIDAY_REGION_PH = 'PH';
/**
* Pakistan
*/
public const HOLIDAY_REGION_PK = 'PK';
/**
* Poland
*/
public const HOLIDAY_REGION_PL = 'PL';
/**
* Portugal
*/
public const HOLIDAY_REGION_PT = 'PT';
/**
* Romania
*/
public const HOLIDAY_REGION_RO = 'RO';
/**
* Serbia
*/
public const HOLIDAY_REGION_RS = 'RS';
/**
* Russian Federation
*/
public const HOLIDAY_REGION_RU = 'RU';
/**
* Saudi Arabia
*/
public const HOLIDAY_REGION_SA = 'SA';
/**
* Sweden
*/
public const HOLIDAY_REGION_SE = 'SE';
/**
* Singapore
*/
public const HOLIDAY_REGION_SG = 'SG';
/**
* Slovenia
*/
public const HOLIDAY_REGION_SI = 'SI';
/**
* Slovakia
*/
public const HOLIDAY_REGION_SK = 'SK';
/**
* Thailand
*/
public const HOLIDAY_REGION_TH = 'TH';
/**
* Turkey
*/
public const HOLIDAY_REGION_TR = 'TR';
/**
* Taiwan
*/
public const HOLIDAY_REGION_TW = 'TW';
/**
* Ukraine
*/
public const HOLIDAY_REGION_UA = 'UA';
/**
* United States
*/
public const HOLIDAY_REGION_US = 'US';
/**
* Venezuela
*/
public const HOLIDAY_REGION_VE = 'VE';
/**
* Vietnam
*/
public const HOLIDAY_REGION_VN = 'VN';
/**
* South Africa
*/
public const HOLIDAY_REGION_ZA = 'ZA';
/**
* Unspecified initialization method.
*/
public const KMEANS_INITIALIZATION_METHOD_KMEANS_INITIALIZATION_METHOD_UNSPECIFIED = 'KMEANS_INITIALIZATION_METHOD_UNSPECIFIED';
/**
* Initializes the centroids randomly.
*/
public const KMEANS_INITIALIZATION_METHOD_RANDOM = 'RANDOM';
/**
* Initializes the centroids using data specified in
* kmeans_initialization_column.
*/
public const KMEANS_INITIALIZATION_METHOD_CUSTOM = 'CUSTOM';
/**
* Initializes with kmeans++.
*/
public const KMEANS_INITIALIZATION_METHOD_KMEANS_PLUS_PLUS = 'KMEANS_PLUS_PLUS';
/**
* Default value.
*/
public const LEARN_RATE_STRATEGY_LEARN_RATE_STRATEGY_UNSPECIFIED = 'LEARN_RATE_STRATEGY_UNSPECIFIED';
/**
* Use line search to determine learning rate.
*/
public const LEARN_RATE_STRATEGY_LINE_SEARCH = 'LINE_SEARCH';
/**
* Use a constant learning rate.
*/
public const LEARN_RATE_STRATEGY_CONSTANT = 'CONSTANT';
/**
* Default value.
*/
public const LOSS_TYPE_LOSS_TYPE_UNSPECIFIED = 'LOSS_TYPE_UNSPECIFIED';
/**
* Mean squared loss, used for linear regression.
*/
public const LOSS_TYPE_MEAN_SQUARED_LOSS = 'MEAN_SQUARED_LOSS';
/**
* Mean log loss, used for logistic regression.
*/
public const LOSS_TYPE_MEAN_LOG_LOSS = 'MEAN_LOG_LOSS';
/**
* Default value.
*/
public const MODEL_REGISTRY_MODEL_REGISTRY_UNSPECIFIED = 'MODEL_REGISTRY_UNSPECIFIED';
/**
* Vertex AI.
*/
public const MODEL_REGISTRY_VERTEX_AI = 'VERTEX_AI';
/**
* Default value.
*/
public const OPTIMIZATION_STRATEGY_OPTIMIZATION_STRATEGY_UNSPECIFIED = 'OPTIMIZATION_STRATEGY_UNSPECIFIED';
/**
* Uses an iterative batch gradient descent algorithm.
*/
public const OPTIMIZATION_STRATEGY_BATCH_GRADIENT_DESCENT = 'BATCH_GRADIENT_DESCENT';
/**
* Uses a normal equation to solve linear regression problem.
*/
public const OPTIMIZATION_STRATEGY_NORMAL_EQUATION = 'NORMAL_EQUATION';
/**
* Default value.
*/
public const PCA_SOLVER_UNSPECIFIED = 'UNSPECIFIED';
/**
* Full eigen-decoposition.
*/
public const PCA_SOLVER_FULL = 'FULL';
/**
* Randomized SVD.
*/
public const PCA_SOLVER_RANDOMIZED = 'RANDOMIZED';
/**
* Auto.
*/
public const PCA_SOLVER_AUTO = 'AUTO';
/**
* Default value.
*/
public const RESERVATION_AFFINITY_TYPE_RESERVATION_AFFINITY_TYPE_UNSPECIFIED = 'RESERVATION_AFFINITY_TYPE_UNSPECIFIED';
/**
* No reservation.
*/
public const RESERVATION_AFFINITY_TYPE_NO_RESERVATION = 'NO_RESERVATION';
/**
* Any reservation.
*/
public const RESERVATION_AFFINITY_TYPE_ANY_RESERVATION = 'ANY_RESERVATION';
/**
* Specific reservation.
*/
public const RESERVATION_AFFINITY_TYPE_SPECIFIC_RESERVATION = 'SPECIFIC_RESERVATION';
/**
* Unspecified tree method.
*/
public const TREE_METHOD_TREE_METHOD_UNSPECIFIED = 'TREE_METHOD_UNSPECIFIED';
/**
* Use heuristic to choose the fastest method.
*/
public const TREE_METHOD_AUTO = 'AUTO';
/**
* Exact greedy algorithm.
*/
public const TREE_METHOD_EXACT = 'EXACT';
/**
* Approximate greedy algorithm using quantile sketch and gradient histogram.
*/
public const TREE_METHOD_APPROX = 'APPROX';
/**
* Fast histogram optimized approximate greedy algorithm.
*/
public const TREE_METHOD_HIST = 'HIST';
protected $collection_key = 'vertexAiModelVersionAliases';
/**
* Activation function of the neural nets.
*
* @var string
*/
public $activationFn;
/**
* If true, detect step changes and make data adjustment in the input time
* series.
*
* @var bool
*/
public $adjustStepChanges;
/**
* Whether to use approximate feature contribution method in XGBoost model
* explanation for global explain.
*
* @var bool
*/
public $approxGlobalFeatureContrib;
/**
* Whether to enable auto ARIMA or not.
*
* @var bool
*/
public $autoArima;
/**
* The max value of the sum of non-seasonal p and q.
*
* @var string
*/
public $autoArimaMaxOrder;
/**
* The min value of the sum of non-seasonal p and q.
*
* @var string
*/
public $autoArimaMinOrder;
/**
* Whether to calculate class weights automatically based on the popularity of
* each label.
*
* @var bool
*/
public $autoClassWeights;
/**
* Batch size for dnn models.
*
* @var string
*/
public $batchSize;
/**
* Booster type for boosted tree models.
*
* @var string
*/
public $boosterType;
/**
* Budget in hours for AutoML training.
*
* @var
*/
public $budgetHours;
/**
* Whether or not p-value test should be computed for this model. Only
* available for linear and logistic regression models.
*
* @var bool
*/
public $calculatePValues;
/**
* Categorical feature encoding method.
*
* @var string
*/
public $categoryEncodingMethod;
/**
* If true, clean spikes and dips in the input time series.
*
* @var bool
*/
public $cleanSpikesAndDips;
/**
* Enums for color space, used for processing images in Object Table. See more
* details at https://www.tensorflow.org/io/tutorials/colorspace.
*
* @var string
*/
public $colorSpace;
/**
* Subsample ratio of columns for each level for boosted tree models.
*
* @var
*/
public $colsampleBylevel;
/**
* Subsample ratio of columns for each node(split) for boosted tree models.
*
* @var
*/
public $colsampleBynode;
/**
* Subsample ratio of columns when constructing each tree for boosted tree
* models.
*
* @var
*/
public $colsampleBytree;
/**
* The contribution metric. Applies to contribution analysis models. Allowed
* formats supported are for summable and summable ratio contribution metrics.
* These include expressions such as `SUM(x)` or `SUM(x)/SUM(y)`, where x and
* y are column names from the base table.
*
* @var string
*/
public $contributionMetric;
/**
* Type of normalization algorithm for boosted tree models using dart booster.
*
* @var string
*/
public $dartNormalizeType;
/**
* The data frequency of a time series.
*
* @var string
*/
public $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
*
* @var string
*/
public $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.
*
* @var
*/
public $dataSplitEvalFraction;
/**
* The data split type for training and evaluation, e.g. RANDOM.
*
* @var string
*/
public $dataSplitMethod;
/**
* If true, perform decompose time series and save the results.
*
* @var bool
*/
public $decomposeTimeSeries;
/**
* Optional. Names of the columns to slice on. Applies to contribution
* analysis models.
*
* @var string[]
*/
public $dimensionIdColumns;
/**
* Distance type for clustering models.
*
* @var string
*/
public $distanceType;
/**
* Dropout probability for dnn models.
*
* @var
*/
public $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.
*
* @var bool
*/
public $earlyStop;
/**
* If true, enable global explanation during training.
*
* @var bool
*/
public $enableGlobalExplain;
/**
* The idle TTL of the endpoint before the resources get destroyed. The
* default value is 6.5 hours.
*
* @var string
*/
public $endpointIdleTtl;
/**
* Feedback type that specifies which algorithm to run for matrix
* factorization.
*
* @var string
*/
public $feedbackType;
/**
* Whether the model should include intercept during model training.
*
* @var bool
*/
public $fitIntercept;
/**
* The forecast limit lower bound that was used during ARIMA model training
* with limits. To see more details of the algorithm:
* https://otexts.com/fpp2/limits.html
*
* @var
*/
public $forecastLimitLowerBound;
/**
* The forecast limit upper bound that was used during ARIMA model training
* with limits.
*
* @var
*/
public $forecastLimitUpperBound;
/**
* Hidden units for dnn models.
*
* @var string[]
*/
public $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.
*
* @var string
*/
public $holidayRegion;
/**
* A list of geographical regions that are used for time series modeling.
*
* @var string[]
*/
public $holidayRegions;
/**
* The number of periods ahead that need to be forecasted.
*
* @var string
*/
public $horizon;
/**
* The target evaluation metrics to optimize the hyperparameters for.
*
* @var string[]
*/
public $hparamTuningObjectives;
/**
* The id of a Hugging Face model. For example, `google/gemma-2-2b-it`.
*
* @var string
*/
public $huggingFaceModelId;
/**
* Include drift when fitting an ARIMA model.
*
* @var bool
*/
public $includeDrift;
/**
* Specifies the initial learning rate for the line search learn rate
* strategy.
*
* @var
*/
public $initialLearnRate;
/**
* Name of input label columns in training data.
*
* @var string[]
*/
public $inputLabelColumns;
/**
* Name of the instance weight column for training data. This column isn't be
* used as a feature.
*
* @var string
*/
public $instanceWeightColumn;
/**
* Number of integral steps for the integrated gradients explain method.
*
* @var string
*/
public $integratedGradientsNumSteps;
/**
* Name of the column used to determine the rows corresponding to control and
* test. Applies to contribution analysis models.
*
* @var string
*/
public $isTestColumn;
/**
* Item column specified for matrix factorization models.
*
* @var string
*/
public $itemColumn;
/**
* The column used to provide the initial centroids for kmeans algorithm when
* kmeans_initialization_method is CUSTOM.
*
* @var string
*/
public $kmeansInitializationColumn;
/**
* The method used to initialize the centroids for kmeans algorithm.
*
* @var string
*/
public $kmeansInitializationMethod;
/**
* L1 regularization coefficient to activations.
*
* @var
*/
public $l1RegActivation;
/**
* L1 regularization coefficient.
*
* @var
*/
public $l1Regularization;
/**
* L2 regularization coefficient.
*
* @var
*/
public $l2Regularization;
/**
* Weights associated with each label class, for rebalancing the training
* data. Only applicable for classification models.
*
* @var []
*/
public $labelClassWeights;
/**
* Learning rate in training. Used only for iterative training algorithms.
*
* @var
*/
public $learnRate;
/**
* The strategy to determine learn rate for the current iteration.
*
* @var string
*/
public $learnRateStrategy;
/**
* Type of loss function used during training run.
*
* @var string
*/
public $lossType;
/**
* The type of the machine used to deploy and serve the model.
*
* @var string
*/
public $machineType;
/**
* The maximum number of iterations in training. Used only for iterative
* training algorithms.
*
* @var string
*/
public $maxIterations;
/**
* Maximum number of trials to run in parallel.
*
* @var string
*/
public $maxParallelTrials;
/**
* The maximum number of machine replicas that will be deployed on an
* endpoint. The default value is equal to min_replica_count.
*
* @var string
*/
public $maxReplicaCount;
/**
* The maximum number of time points in a time series that can be used in
* modeling the trend component of the time series. Don't use this option with
* the `timeSeriesLengthFraction` or `minTimeSeriesLength` options.
*
* @var string
*/
public $maxTimeSeriesLength;
/**
* Maximum depth of a tree for boosted tree models.
*
* @var string
*/
public $maxTreeDepth;
/**
* The apriori support minimum. Applies to contribution analysis models.
*
* @var
*/
public $minAprioriSupport;
/**
* When early_stop is true, stops training when accuracy improvement is less
* than 'min_relative_progress'. Used only for iterative training algorithms.
*
* @var
*/
public $minRelativeProgress;
/**
* The minimum number of machine replicas that will be always deployed on an
* endpoint. This value must be greater than or equal to 1. The default value
* is 1.
*
* @var string
*/
public $minReplicaCount;
/**
* Minimum split loss for boosted tree models.
*
* @var
*/
public $minSplitLoss;
/**
* The minimum number of time points in a time series that are used in
* modeling the trend component of the time series. If you use this option you
* must also set the `timeSeriesLengthFraction` option. This training option
* ensures that enough time points are available when you use
* `timeSeriesLengthFraction` in trend modeling. This is particularly
* important when forecasting multiple time series in a single query using
* `timeSeriesIdColumn`. If the total number of time points is less than the
* `minTimeSeriesLength` value, then the query uses all available time points.
*
* @var string
*/
public $minTimeSeriesLength;
/**
* Minimum sum of instance weight needed in a child for boosted tree models.
*
* @var string
*/
public $minTreeChildWeight;
/**
* The name of a Vertex model garden publisher model. Format is
* `publishers/{publisher}/models/{model}@{optional_version_id}`.
*
* @var string
*/
public $modelGardenModelName;
/**
* The model registry.
*
* @var string
*/
public $modelRegistry;
/**
* Google Cloud Storage URI from which the model was imported. Only applicable
* for imported models.
*
* @var string
*/
public $modelUri;
protected $nonSeasonalOrderType = ArimaOrder::class;
protected $nonSeasonalOrderDataType = '';
/**
* Number of clusters for clustering models.
*
* @var string
*/
public $numClusters;
/**
* Num factors specified for matrix factorization models.
*
* @var string
*/
public $numFactors;
/**
* Number of parallel trees constructed during each iteration for boosted tree
* models.
*
* @var string
*/
public $numParallelTree;
/**
* Number of principal components to keep in the PCA model. Must be <= the
* number of features.
*
* @var string
*/
public $numPrincipalComponents;
/**
* Number of trials to run this hyperparameter tuning job.
*
* @var string
*/
public $numTrials;
/**
* Optimization strategy for training linear regression models.
*
* @var string
*/
public $optimizationStrategy;
/**
* Optimizer used for training the neural nets.
*
* @var string
*/
public $optimizer;
/**
* The minimum ratio of cumulative explained variance that needs to be given
* by the PCA model.
*
* @var
*/
public $pcaExplainedVarianceRatio;
/**
* The solver for PCA.
*
* @var string
*/
public $pcaSolver;
/**
* Corresponds to the label key of a reservation resource used by Vertex AI.
* To target a SPECIFIC_RESERVATION by name, use
* `compute.googleapis.com/reservation-name` as the key and specify the name
* of your reservation as its value.
*
* @var string
*/
public $reservationAffinityKey;
/**
* Specifies the reservation affinity type used to configure a Vertex AI
* resource. The default value is `NO_RESERVATION`.
*
* @var string
*/
public $reservationAffinityType;
/**
* Corresponds to the label values of a reservation resource used by Vertex
* AI. This must be the full resource name of the reservation or reservation
* block.
*
* @var string[]
*/
public $reservationAffinityValues;
/**
* Number of paths for the sampled Shapley explain method.
*
* @var string
*/
public $sampledShapleyNumPaths;
/**
* If true, scale the feature values by dividing the feature standard
* deviation. Currently only apply to PCA.
*
* @var bool
*/
public $scaleFeatures;
/**
* Whether to standardize numerical features. Default to true.
*
* @var bool
*/
public $standardizeFeatures;
/**
* Subsample fraction of the training data to grow tree to prevent overfitting
* for boosted tree models.
*
* @var
*/
public $subsample;
/**
* Based on the selected TF version, the corresponding docker image is used to
* train external models.
*
* @var string
*/
public $tfVersion;
/**
* Column to be designated as time series data for ARIMA model.
*
* @var string
*/
public $timeSeriesDataColumn;
/**
* The time series id column that was used during ARIMA model training.
*
* @var string
*/
public $timeSeriesIdColumn;
/**
* The time series id columns that were used during ARIMA model training.
*
* @var string[]
*/
public $timeSeriesIdColumns;
/**
* The fraction of the interpolated length of the time series that's used to
* model the time series trend component. All of the time points of the time
* series are used to model the non-trend component. This training option
* accelerates modeling training without sacrificing much forecasting
* accuracy. You can use this option with `minTimeSeriesLength` but not with
* `maxTimeSeriesLength`.
*
* @var
*/
public $timeSeriesLengthFraction;
/**
* Column to be designated as time series timestamp for ARIMA model.
*
* @var string
*/
public $timeSeriesTimestampColumn;
/**
* Tree construction algorithm for boosted tree models.
*
* @var string
*/
public $treeMethod;
/**
* Smoothing window size for the trend component. When a positive value is
* specified, a center moving average smoothing is applied on the history
* trend. When the smoothing window is out of the boundary at the beginning or
* the end of the trend, the first element or the last element is padded to
* fill the smoothing window before the average is applied.
*
* @var string
*/
public $trendSmoothingWindowSize;
/**
* User column specified for matrix factorization models.
*
* @var string
*/
public $userColumn;
/**
* The version aliases to apply in Vertex AI model registry. Always overwrite
* if the version aliases exists in a existing model.
*
* @var string[]
*/
public $vertexAiModelVersionAliases;
/**
* Hyperparameter for matrix factoration when implicit feedback type is
* specified.
*
* @var
*/
public $walsAlpha;
/**
* Whether to train a model from the last checkpoint.
*
* @var bool
*/
public $warmStart;
/**
* User-selected XGBoost versions for training of XGBoost models.
*
* @var string
*/
public $xgboostVersion;
/**
* Activation function of the neural nets.
*
* @param string $activationFn
*/
public function setActivationFn($activationFn)
{
$this->activationFn = $activationFn;
}
/**
* @return string
*/
public function getActivationFn()
{
return $this->activationFn;
}
/**
* If true, detect step changes and make data adjustment in the input time
* series.
*
* @param bool $adjustStepChanges
*/
public function setAdjustStepChanges($adjustStepChanges)
{
$this->adjustStepChanges = $adjustStepChanges;
}
/**
* @return bool
*/
public function getAdjustStepChanges()
{
return $this->adjustStepChanges;
}
/**
* Whether to use approximate feature contribution method in XGBoost model
* explanation for global explain.
*
* @param bool $approxGlobalFeatureContrib
*/
public function setApproxGlobalFeatureContrib($approxGlobalFeatureContrib)
{
$this->approxGlobalFeatureContrib = $approxGlobalFeatureContrib;
}
/**
* @return bool
*/
public function getApproxGlobalFeatureContrib()
{
return $this->approxGlobalFeatureContrib;
}
/**
* Whether to enable auto ARIMA or not.
*
* @param bool $autoArima
*/
public function setAutoArima($autoArima)
{
$this->autoArima = $autoArima;
}
/**
* @return bool
*/
public function getAutoArima()
{
return $this->autoArima;
}
/**
* The max value of the sum of non-seasonal p and q.
*
* @param string $autoArimaMaxOrder
*/
public function setAutoArimaMaxOrder($autoArimaMaxOrder)
{
$this->autoArimaMaxOrder = $autoArimaMaxOrder;
}
/**
* @return string
*/
public function getAutoArimaMaxOrder()
{
return $this->autoArimaMaxOrder;
}
/**
* The min value of the sum of non-seasonal p and q.
*
* @param string $autoArimaMinOrder
*/
public function setAutoArimaMinOrder($autoArimaMinOrder)
{
$this->autoArimaMinOrder = $autoArimaMinOrder;
}
/**
* @return string
*/
public function getAutoArimaMinOrder()
{
return $this->autoArimaMinOrder;
}
/**
* Whether to calculate class weights automatically based on the popularity of
* each label.
*
* @param bool $autoClassWeights
*/
public function setAutoClassWeights($autoClassWeights)
{
$this->autoClassWeights = $autoClassWeights;
}
/**
* @return bool
*/
public function getAutoClassWeights()
{
return $this->autoClassWeights;
}
/**
* Batch size for dnn models.
*
* @param string $batchSize
*/
public function setBatchSize($batchSize)
{
$this->batchSize = $batchSize;
}
/**
* @return string
*/
public function getBatchSize()
{
return $this->batchSize;
}
/**
* Booster type for boosted tree models.
*
* Accepted values: BOOSTER_TYPE_UNSPECIFIED, GBTREE, DART
*
* @param self::BOOSTER_TYPE_* $boosterType
*/
public function setBoosterType($boosterType)
{
$this->boosterType = $boosterType;
}
/**
* @return self::BOOSTER_TYPE_*
*/
public function getBoosterType()
{
return $this->boosterType;
}
public function setBudgetHours($budgetHours)
{
$this->budgetHours = $budgetHours;
}
public function getBudgetHours()
{
return $this->budgetHours;
}
/**
* Whether or not p-value test should be computed for this model. Only
* available for linear and logistic regression models.
*
* @param bool $calculatePValues
*/
public function setCalculatePValues($calculatePValues)
{
$this->calculatePValues = $calculatePValues;
}
/**
* @return bool
*/
public function getCalculatePValues()
{
return $this->calculatePValues;
}
/**
* Categorical feature encoding method.
*
* Accepted values: ENCODING_METHOD_UNSPECIFIED, ONE_HOT_ENCODING,
* LABEL_ENCODING, DUMMY_ENCODING
*
* @param self::CATEGORY_ENCODING_METHOD_* $categoryEncodingMethod
*/
public function setCategoryEncodingMethod($categoryEncodingMethod)
{
$this->categoryEncodingMethod = $categoryEncodingMethod;
}
/**
* @return self::CATEGORY_ENCODING_METHOD_*
*/
public function getCategoryEncodingMethod()
{
return $this->categoryEncodingMethod;
}
/**
* If true, clean spikes and dips in the input time series.
*
* @param bool $cleanSpikesAndDips
*/
public function setCleanSpikesAndDips($cleanSpikesAndDips)
{
$this->cleanSpikesAndDips = $cleanSpikesAndDips;
}
/**
* @return bool
*/
public function getCleanSpikesAndDips()
{
return $this->cleanSpikesAndDips;
}
/**
* Enums for color space, used for processing images in Object Table. See more
* details at https://www.tensorflow.org/io/tutorials/colorspace.
*
* Accepted values: COLOR_SPACE_UNSPECIFIED, RGB, HSV, YIQ, YUV, GRAYSCALE
*
* @param self::COLOR_SPACE_* $colorSpace
*/
public function setColorSpace($colorSpace)
{
$this->colorSpace = $colorSpace;
}
/**
* @return self::COLOR_SPACE_*
*/
public function getColorSpace()
{
return $this->colorSpace;
}
public function setColsampleBylevel($colsampleBylevel)
{
$this->colsampleBylevel = $colsampleBylevel;
}
public function getColsampleBylevel()
{
return $this->colsampleBylevel;
}
public function setColsampleBynode($colsampleBynode)
{
$this->colsampleBynode = $colsampleBynode;
}
public function getColsampleBynode()
{
return $this->colsampleBynode;
}
public function setColsampleBytree($colsampleBytree)
{
$this->colsampleBytree = $colsampleBytree;
}
public function getColsampleBytree()
{
return $this->colsampleBytree;
}
/**
* The contribution metric. Applies to contribution analysis models. Allowed
* formats supported are for summable and summable ratio contribution metrics.
* These include expressions such as `SUM(x)` or `SUM(x)/SUM(y)`, where x and
* y are column names from the base table.
*
* @param string $contributionMetric
*/
public function setContributionMetric($contributionMetric)
{
$this->contributionMetric = $contributionMetric;
}
/**
* @return string
*/
public function getContributionMetric()
{
return $this->contributionMetric;
}
/**
* Type of normalization algorithm for boosted tree models using dart booster.
*
* Accepted values: DART_NORMALIZE_TYPE_UNSPECIFIED, TREE, FOREST
*
* @param self::DART_NORMALIZE_TYPE_* $dartNormalizeType
*/
public function setDartNormalizeType($dartNormalizeType)
{
$this->dartNormalizeType = $dartNormalizeType;
}
/**
* @return self::DART_NORMALIZE_TYPE_*
*/
public function getDartNormalizeType()
{
return $this->dartNormalizeType;
}
/**
* The data frequency of a time series.
*
* Accepted values: DATA_FREQUENCY_UNSPECIFIED, AUTO_FREQUENCY, YEARLY,
* QUARTERLY, MONTHLY, WEEKLY, DAILY, HOURLY, PER_MINUTE
*
* @param self::DATA_FREQUENCY_* $dataFrequency
*/
public function setDataFrequency($dataFrequency)
{
$this->dataFrequency = $dataFrequency;
}
/**
* @return self::DATA_FREQUENCY_*
*/
public function getDataFrequency()
{
return $this->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
*
* @param string $dataSplitColumn
*/
public function setDataSplitColumn($dataSplitColumn)
{
$this->dataSplitColumn = $dataSplitColumn;
}
/**
* @return string
*/
public function getDataSplitColumn()
{
return $this->dataSplitColumn;
}
public function setDataSplitEvalFraction($dataSplitEvalFraction)
{
$this->dataSplitEvalFraction = $dataSplitEvalFraction;
}
public function getDataSplitEvalFraction()
{
return $this->dataSplitEvalFraction;
}
/**
* The data split type for training and evaluation, e.g. RANDOM.
*
* Accepted values: DATA_SPLIT_METHOD_UNSPECIFIED, RANDOM, CUSTOM, SEQUENTIAL,
* NO_SPLIT, AUTO_SPLIT
*
* @param self::DATA_SPLIT_METHOD_* $dataSplitMethod
*/
public function setDataSplitMethod($dataSplitMethod)
{
$this->dataSplitMethod = $dataSplitMethod;
}
/**
* @return self::DATA_SPLIT_METHOD_*
*/
public function getDataSplitMethod()
{
return $this->dataSplitMethod;
}
/**
* If true, perform decompose time series and save the results.
*
* @param bool $decomposeTimeSeries
*/
public function setDecomposeTimeSeries($decomposeTimeSeries)
{
$this->decomposeTimeSeries = $decomposeTimeSeries;
}
/**
* @return bool
*/
public function getDecomposeTimeSeries()
{
return $this->decomposeTimeSeries;
}
/**
* Optional. Names of the columns to slice on. Applies to contribution
* analysis models.
*
* @param string[] $dimensionIdColumns
*/
public function setDimensionIdColumns($dimensionIdColumns)
{
$this->dimensionIdColumns = $dimensionIdColumns;
}
/**
* @return string[]
*/
public function getDimensionIdColumns()
{
return $this->dimensionIdColumns;
}
/**
* Distance type for clustering models.
*
* Accepted values: DISTANCE_TYPE_UNSPECIFIED, EUCLIDEAN, COSINE
*
* @param self::DISTANCE_TYPE_* $distanceType
*/
public function setDistanceType($distanceType)
{
$this->distanceType = $distanceType;
}
/**
* @return self::DISTANCE_TYPE_*
*/
public function getDistanceType()
{
return $this->distanceType;
}
public function setDropout($dropout)
{
$this->dropout = $dropout;
}
public function getDropout()
{
return $this->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.
*
* @param bool $earlyStop
*/
public function setEarlyStop($earlyStop)
{
$this->earlyStop = $earlyStop;
}
/**
* @return bool
*/
public function getEarlyStop()
{
return $this->earlyStop;
}
/**
* If true, enable global explanation during training.
*
* @param bool $enableGlobalExplain
*/
public function setEnableGlobalExplain($enableGlobalExplain)
{
$this->enableGlobalExplain = $enableGlobalExplain;
}
/**
* @return bool
*/
public function getEnableGlobalExplain()
{
return $this->enableGlobalExplain;
}
/**
* The idle TTL of the endpoint before the resources get destroyed. The
* default value is 6.5 hours.
*
* @param string $endpointIdleTtl
*/
public function setEndpointIdleTtl($endpointIdleTtl)
{
$this->endpointIdleTtl = $endpointIdleTtl;
}
/**
* @return string
*/
public function getEndpointIdleTtl()
{
return $this->endpointIdleTtl;
}
/**
* Feedback type that specifies which algorithm to run for matrix
* factorization.
*
* Accepted values: FEEDBACK_TYPE_UNSPECIFIED, IMPLICIT, EXPLICIT
*
* @param self::FEEDBACK_TYPE_* $feedbackType
*/
public function setFeedbackType($feedbackType)
{
$this->feedbackType = $feedbackType;
}
/**
* @return self::FEEDBACK_TYPE_*
*/
public function getFeedbackType()
{
return $this->feedbackType;
}
/**
* Whether the model should include intercept during model training.
*
* @param bool $fitIntercept
*/
public function setFitIntercept($fitIntercept)
{
$this->fitIntercept = $fitIntercept;
}
/**
* @return bool
*/
public function getFitIntercept()
{
return $this->fitIntercept;
}
public function setForecastLimitLowerBound($forecastLimitLowerBound)
{
$this->forecastLimitLowerBound = $forecastLimitLowerBound;
}
public function getForecastLimitLowerBound()
{
return $this->forecastLimitLowerBound;
}
public function setForecastLimitUpperBound($forecastLimitUpperBound)
{
$this->forecastLimitUpperBound = $forecastLimitUpperBound;
}
public function getForecastLimitUpperBound()
{
return $this->forecastLimitUpperBound;
}
/**
* Hidden units for dnn models.
*
* @param string[] $hiddenUnits
*/
public function setHiddenUnits($hiddenUnits)
{
$this->hiddenUnits = $hiddenUnits;
}
/**
* @return string[]
*/
public function getHiddenUnits()
{
return $this->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.
*
* Accepted values: HOLIDAY_REGION_UNSPECIFIED, GLOBAL, NA, JAPAC, EMEA, LAC,
* AE, AR, AT, AU, BE, BR, CA, CH, CL, CN, CO, CS, CZ, DE, DK, DZ, EC, EE, EG,
* ES, FI, FR, GB, GR, HK, HU, ID, IE, IL, IN, IR, IT, JP, KR, LV, MA, MX, MY,
* NG, NL, NO, NZ, PE, PH, PK, PL, PT, RO, RS, RU, SA, SE, SG, SI, SK, TH, TR,
* TW, UA, US, VE, VN, ZA
*
* @param self::HOLIDAY_REGION_* $holidayRegion
*/
public function setHolidayRegion($holidayRegion)
{
$this->holidayRegion = $holidayRegion;
}
/**
* @return self::HOLIDAY_REGION_*
*/
public function getHolidayRegion()
{
return $this->holidayRegion;
}
/**
* A list of geographical regions that are used for time series modeling.
*
* @param string[] $holidayRegions
*/
public function setHolidayRegions($holidayRegions)
{
$this->holidayRegions = $holidayRegions;
}
/**
* @return string[]
*/
public function getHolidayRegions()
{
return $this->holidayRegions;
}
/**
* The number of periods ahead that need to be forecasted.
*
* @param string $horizon
*/
public function setHorizon($horizon)
{
$this->horizon = $horizon;
}
/**
* @return string
*/
public function getHorizon()
{
return $this->horizon;
}
/**
* The target evaluation metrics to optimize the hyperparameters for.
*
* @param string[] $hparamTuningObjectives
*/
public function setHparamTuningObjectives($hparamTuningObjectives)
{
$this->hparamTuningObjectives = $hparamTuningObjectives;
}
/**
* @return string[]
*/
public function getHparamTuningObjectives()
{
return $this->hparamTuningObjectives;
}
/**
* The id of a Hugging Face model. For example, `google/gemma-2-2b-it`.
*
* @param string $huggingFaceModelId
*/
public function setHuggingFaceModelId($huggingFaceModelId)
{
$this->huggingFaceModelId = $huggingFaceModelId;
}
/**
* @return string
*/
public function getHuggingFaceModelId()
{
return $this->huggingFaceModelId;
}
/**
* Include drift when fitting an ARIMA model.
*
* @param bool $includeDrift
*/
public function setIncludeDrift($includeDrift)
{
$this->includeDrift = $includeDrift;
}
/**
* @return bool
*/
public function getIncludeDrift()
{
return $this->includeDrift;
}
public function setInitialLearnRate($initialLearnRate)
{
$this->initialLearnRate = $initialLearnRate;
}
public function getInitialLearnRate()
{
return $this->initialLearnRate;
}
/**
* Name of input label columns in training data.
*
* @param string[] $inputLabelColumns
*/
public function setInputLabelColumns($inputLabelColumns)
{
$this->inputLabelColumns = $inputLabelColumns;
}
/**
* @return string[]
*/
public function getInputLabelColumns()
{
return $this->inputLabelColumns;
}
/**
* Name of the instance weight column for training data. This column isn't be
* used as a feature.
*
* @param string $instanceWeightColumn
*/
public function setInstanceWeightColumn($instanceWeightColumn)
{
$this->instanceWeightColumn = $instanceWeightColumn;
}
/**
* @return string
*/
public function getInstanceWeightColumn()
{
return $this->instanceWeightColumn;
}
/**
* Number of integral steps for the integrated gradients explain method.
*
* @param string $integratedGradientsNumSteps
*/
public function setIntegratedGradientsNumSteps($integratedGradientsNumSteps)
{
$this->integratedGradientsNumSteps = $integratedGradientsNumSteps;
}
/**
* @return string
*/
public function getIntegratedGradientsNumSteps()
{
return $this->integratedGradientsNumSteps;
}
/**
* Name of the column used to determine the rows corresponding to control and
* test. Applies to contribution analysis models.
*
* @param string $isTestColumn
*/
public function setIsTestColumn($isTestColumn)
{
$this->isTestColumn = $isTestColumn;
}
/**
* @return string
*/
public function getIsTestColumn()
{
return $this->isTestColumn;
}
/**
* Item column specified for matrix factorization models.
*
* @param string $itemColumn
*/
public function setItemColumn($itemColumn)
{
$this->itemColumn = $itemColumn;
}
/**
* @return string
*/
public function getItemColumn()
{
return $this->itemColumn;
}
/**
* The column used to provide the initial centroids for kmeans algorithm when
* kmeans_initialization_method is CUSTOM.
*
* @param string $kmeansInitializationColumn
*/
public function setKmeansInitializationColumn($kmeansInitializationColumn)
{
$this->kmeansInitializationColumn = $kmeansInitializationColumn;
}
/**
* @return string
*/
public function getKmeansInitializationColumn()
{
return $this->kmeansInitializationColumn;
}
/**
* The method used to initialize the centroids for kmeans algorithm.
*
* Accepted values: KMEANS_INITIALIZATION_METHOD_UNSPECIFIED, RANDOM, CUSTOM,
* KMEANS_PLUS_PLUS
*
* @param self::KMEANS_INITIALIZATION_METHOD_* $kmeansInitializationMethod
*/
public function setKmeansInitializationMethod($kmeansInitializationMethod)
{
$this->kmeansInitializationMethod = $kmeansInitializationMethod;
}
/**
* @return self::KMEANS_INITIALIZATION_METHOD_*
*/
public function getKmeansInitializationMethod()
{
return $this->kmeansInitializationMethod;
}
public function setL1RegActivation($l1RegActivation)
{
$this->l1RegActivation = $l1RegActivation;
}
public function getL1RegActivation()
{
return $this->l1RegActivation;
}
public function setL1Regularization($l1Regularization)
{
$this->l1Regularization = $l1Regularization;
}
public function getL1Regularization()
{
return $this->l1Regularization;
}
public function setL2Regularization($l2Regularization)
{
$this->l2Regularization = $l2Regularization;
}
public function getL2Regularization()
{
return $this->l2Regularization;
}
public function setLabelClassWeights($labelClassWeights)
{
$this->labelClassWeights = $labelClassWeights;
}
public function getLabelClassWeights()
{
return $this->labelClassWeights;
}
public function setLearnRate($learnRate)
{
$this->learnRate = $learnRate;
}
public function getLearnRate()
{
return $this->learnRate;
}
/**
* The strategy to determine learn rate for the current iteration.
*
* Accepted values: LEARN_RATE_STRATEGY_UNSPECIFIED, LINE_SEARCH, CONSTANT
*
* @param self::LEARN_RATE_STRATEGY_* $learnRateStrategy
*/
public function setLearnRateStrategy($learnRateStrategy)
{
$this->learnRateStrategy = $learnRateStrategy;
}
/**
* @return self::LEARN_RATE_STRATEGY_*
*/
public function getLearnRateStrategy()
{
return $this->learnRateStrategy;
}
/**
* Type of loss function used during training run.
*
* Accepted values: LOSS_TYPE_UNSPECIFIED, MEAN_SQUARED_LOSS, MEAN_LOG_LOSS
*
* @param self::LOSS_TYPE_* $lossType
*/
public function setLossType($lossType)
{
$this->lossType = $lossType;
}
/**
* @return self::LOSS_TYPE_*
*/
public function getLossType()
{
return $this->lossType;
}
/**
* The type of the machine used to deploy and serve the model.
*
* @param string $machineType
*/
public function setMachineType($machineType)
{
$this->machineType = $machineType;
}
/**
* @return string
*/
public function getMachineType()
{
return $this->machineType;
}
/**
* The maximum number of iterations in training. Used only for iterative
* training algorithms.
*
* @param string $maxIterations
*/
public function setMaxIterations($maxIterations)
{
$this->maxIterations = $maxIterations;
}
/**
* @return string
*/
public function getMaxIterations()
{
return $this->maxIterations;
}
/**
* Maximum number of trials to run in parallel.
*
* @param string $maxParallelTrials
*/
public function setMaxParallelTrials($maxParallelTrials)
{
$this->maxParallelTrials = $maxParallelTrials;
}
/**
* @return string
*/
public function getMaxParallelTrials()
{
return $this->maxParallelTrials;
}
/**
* The maximum number of machine replicas that will be deployed on an
* endpoint. The default value is equal to min_replica_count.
*
* @param string $maxReplicaCount
*/
public function setMaxReplicaCount($maxReplicaCount)
{
$this->maxReplicaCount = $maxReplicaCount;
}
/**
* @return string
*/
public function getMaxReplicaCount()
{
return $this->maxReplicaCount;
}
/**
* The maximum number of time points in a time series that can be used in
* modeling the trend component of the time series. Don't use this option with
* the `timeSeriesLengthFraction` or `minTimeSeriesLength` options.
*
* @param string $maxTimeSeriesLength
*/
public function setMaxTimeSeriesLength($maxTimeSeriesLength)
{
$this->maxTimeSeriesLength = $maxTimeSeriesLength;
}
/**
* @return string
*/
public function getMaxTimeSeriesLength()
{
return $this->maxTimeSeriesLength;
}
/**
* Maximum depth of a tree for boosted tree models.
*
* @param string $maxTreeDepth
*/
public function setMaxTreeDepth($maxTreeDepth)
{
$this->maxTreeDepth = $maxTreeDepth;
}
/**
* @return string
*/
public function getMaxTreeDepth()
{
return $this->maxTreeDepth;
}
public function setMinAprioriSupport($minAprioriSupport)
{
$this->minAprioriSupport = $minAprioriSupport;
}
public function getMinAprioriSupport()
{
return $this->minAprioriSupport;
}
public function setMinRelativeProgress($minRelativeProgress)
{
$this->minRelativeProgress = $minRelativeProgress;
}
public function getMinRelativeProgress()
{
return $this->minRelativeProgress;
}
/**
* The minimum number of machine replicas that will be always deployed on an
* endpoint. This value must be greater than or equal to 1. The default value
* is 1.
*
* @param string $minReplicaCount
*/
public function setMinReplicaCount($minReplicaCount)
{
$this->minReplicaCount = $minReplicaCount;
}
/**
* @return string
*/
public function getMinReplicaCount()
{
return $this->minReplicaCount;
}
public function setMinSplitLoss($minSplitLoss)
{
$this->minSplitLoss = $minSplitLoss;
}
public function getMinSplitLoss()
{
return $this->minSplitLoss;
}
/**
* The minimum number of time points in a time series that are used in
* modeling the trend component of the time series. If you use this option you
* must also set the `timeSeriesLengthFraction` option. This training option
* ensures that enough time points are available when you use
* `timeSeriesLengthFraction` in trend modeling. This is particularly
* important when forecasting multiple time series in a single query using
* `timeSeriesIdColumn`. If the total number of time points is less than the
* `minTimeSeriesLength` value, then the query uses all available time points.
*
* @param string $minTimeSeriesLength
*/
public function setMinTimeSeriesLength($minTimeSeriesLength)
{
$this->minTimeSeriesLength = $minTimeSeriesLength;
}
/**
* @return string
*/
public function getMinTimeSeriesLength()
{
return $this->minTimeSeriesLength;
}
/**
* Minimum sum of instance weight needed in a child for boosted tree models.
*
* @param string $minTreeChildWeight
*/
public function setMinTreeChildWeight($minTreeChildWeight)
{
$this->minTreeChildWeight = $minTreeChildWeight;
}
/**
* @return string
*/
public function getMinTreeChildWeight()
{
return $this->minTreeChildWeight;
}
/**
* The name of a Vertex model garden publisher model. Format is
* `publishers/{publisher}/models/{model}@{optional_version_id}`.
*
* @param string $modelGardenModelName
*/
public function setModelGardenModelName($modelGardenModelName)
{
$this->modelGardenModelName = $modelGardenModelName;
}
/**
* @return string
*/
public function getModelGardenModelName()
{
return $this->modelGardenModelName;
}
/**
* The model registry.
*
* Accepted values: MODEL_REGISTRY_UNSPECIFIED, VERTEX_AI
*
* @param self::MODEL_REGISTRY_* $modelRegistry
*/
public function setModelRegistry($modelRegistry)
{
$this->modelRegistry = $modelRegistry;
}
/**
* @return self::MODEL_REGISTRY_*
*/
public function getModelRegistry()
{
return $this->modelRegistry;
}
/**
* Google Cloud Storage URI from which the model was imported. Only applicable
* for imported models.
*
* @param string $modelUri
*/
public function setModelUri($modelUri)
{
$this->modelUri = $modelUri;
}
/**
* @return string
*/
public function getModelUri()
{
return $this->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.
*
* @param ArimaOrder $nonSeasonalOrder
*/
public function setNonSeasonalOrder(ArimaOrder $nonSeasonalOrder)
{
$this->nonSeasonalOrder = $nonSeasonalOrder;
}
/**
* @return ArimaOrder
*/
public function getNonSeasonalOrder()
{
return $this->nonSeasonalOrder;
}
/**
* Number of clusters for clustering models.
*
* @param string $numClusters
*/
public function setNumClusters($numClusters)
{
$this->numClusters = $numClusters;
}
/**
* @return string
*/
public function getNumClusters()
{
return $this->numClusters;
}
/**
* Num factors specified for matrix factorization models.
*
* @param string $numFactors
*/
public function setNumFactors($numFactors)
{
$this->numFactors = $numFactors;
}
/**
* @return string
*/
public function getNumFactors()
{
return $this->numFactors;
}
/**
* Number of parallel trees constructed during each iteration for boosted tree
* models.
*
* @param string $numParallelTree
*/
public function setNumParallelTree($numParallelTree)
{
$this->numParallelTree = $numParallelTree;
}
/**
* @return string
*/
public function getNumParallelTree()
{
return $this->numParallelTree;
}
/**
* Number of principal components to keep in the PCA model. Must be <= the
* number of features.
*
* @param string $numPrincipalComponents
*/
public function setNumPrincipalComponents($numPrincipalComponents)
{
$this->numPrincipalComponents = $numPrincipalComponents;
}
/**
* @return string
*/
public function getNumPrincipalComponents()
{
return $this->numPrincipalComponents;
}
/**
* Number of trials to run this hyperparameter tuning job.
*
* @param string $numTrials
*/
public function setNumTrials($numTrials)
{
$this->numTrials = $numTrials;
}
/**
* @return string
*/
public function getNumTrials()
{
return $this->numTrials;
}
/**
* Optimization strategy for training linear regression models.
*
* Accepted values: OPTIMIZATION_STRATEGY_UNSPECIFIED, BATCH_GRADIENT_DESCENT,
* NORMAL_EQUATION
*
* @param self::OPTIMIZATION_STRATEGY_* $optimizationStrategy
*/
public function setOptimizationStrategy($optimizationStrategy)
{
$this->optimizationStrategy = $optimizationStrategy;
}
/**
* @return self::OPTIMIZATION_STRATEGY_*
*/
public function getOptimizationStrategy()
{
return $this->optimizationStrategy;
}
/**
* Optimizer used for training the neural nets.
*
* @param string $optimizer
*/
public function setOptimizer($optimizer)
{
$this->optimizer = $optimizer;
}
/**
* @return string
*/
public function getOptimizer()
{
return $this->optimizer;
}
public function setPcaExplainedVarianceRatio($pcaExplainedVarianceRatio)
{
$this->pcaExplainedVarianceRatio = $pcaExplainedVarianceRatio;
}
public function getPcaExplainedVarianceRatio()
{
return $this->pcaExplainedVarianceRatio;
}
/**
* The solver for PCA.
*
* Accepted values: UNSPECIFIED, FULL, RANDOMIZED, AUTO
*
* @param self::PCA_SOLVER_* $pcaSolver
*/
public function setPcaSolver($pcaSolver)
{
$this->pcaSolver = $pcaSolver;
}
/**
* @return self::PCA_SOLVER_*
*/
public function getPcaSolver()
{
return $this->pcaSolver;
}
/**
* Corresponds to the label key of a reservation resource used by Vertex AI.
* To target a SPECIFIC_RESERVATION by name, use
* `compute.googleapis.com/reservation-name` as the key and specify the name
* of your reservation as its value.
*
* @param string $reservationAffinityKey
*/
public function setReservationAffinityKey($reservationAffinityKey)
{
$this->reservationAffinityKey = $reservationAffinityKey;
}
/**
* @return string
*/
public function getReservationAffinityKey()
{
return $this->reservationAffinityKey;
}
/**
* Specifies the reservation affinity type used to configure a Vertex AI
* resource. The default value is `NO_RESERVATION`.
*
* Accepted values: RESERVATION_AFFINITY_TYPE_UNSPECIFIED, NO_RESERVATION,
* ANY_RESERVATION, SPECIFIC_RESERVATION
*
* @param self::RESERVATION_AFFINITY_TYPE_* $reservationAffinityType
*/
public function setReservationAffinityType($reservationAffinityType)
{
$this->reservationAffinityType = $reservationAffinityType;
}
/**
* @return self::RESERVATION_AFFINITY_TYPE_*
*/
public function getReservationAffinityType()
{
return $this->reservationAffinityType;
}
/**
* Corresponds to the label values of a reservation resource used by Vertex
* AI. This must be the full resource name of the reservation or reservation
* block.
*
* @param string[] $reservationAffinityValues
*/
public function setReservationAffinityValues($reservationAffinityValues)
{
$this->reservationAffinityValues = $reservationAffinityValues;
}
/**
* @return string[]
*/
public function getReservationAffinityValues()
{
return $this->reservationAffinityValues;
}
/**
* Number of paths for the sampled Shapley explain method.
*
* @param string $sampledShapleyNumPaths
*/
public function setSampledShapleyNumPaths($sampledShapleyNumPaths)
{
$this->sampledShapleyNumPaths = $sampledShapleyNumPaths;
}
/**
* @return string
*/
public function getSampledShapleyNumPaths()
{
return $this->sampledShapleyNumPaths;
}
/**
* If true, scale the feature values by dividing the feature standard
* deviation. Currently only apply to PCA.
*
* @param bool $scaleFeatures
*/
public function setScaleFeatures($scaleFeatures)
{
$this->scaleFeatures = $scaleFeatures;
}
/**
* @return bool
*/
public function getScaleFeatures()
{
return $this->scaleFeatures;
}
/**
* Whether to standardize numerical features. Default to true.
*
* @param bool $standardizeFeatures
*/
public function setStandardizeFeatures($standardizeFeatures)
{
$this->standardizeFeatures = $standardizeFeatures;
}
/**
* @return bool
*/
public function getStandardizeFeatures()
{
return $this->standardizeFeatures;
}
public function setSubsample($subsample)
{
$this->subsample = $subsample;
}
public function getSubsample()
{
return $this->subsample;
}
/**
* Based on the selected TF version, the corresponding docker image is used to
* train external models.
*
* @param string $tfVersion
*/
public function setTfVersion($tfVersion)
{
$this->tfVersion = $tfVersion;
}
/**
* @return string
*/
public function getTfVersion()
{
return $this->tfVersion;
}
/**
* Column to be designated as time series data for ARIMA model.
*
* @param string $timeSeriesDataColumn
*/
public function setTimeSeriesDataColumn($timeSeriesDataColumn)
{
$this->timeSeriesDataColumn = $timeSeriesDataColumn;
}
/**
* @return string
*/
public function getTimeSeriesDataColumn()
{
return $this->timeSeriesDataColumn;
}
/**
* The time series id column that was used during ARIMA model training.
*
* @param string $timeSeriesIdColumn
*/
public function setTimeSeriesIdColumn($timeSeriesIdColumn)
{
$this->timeSeriesIdColumn = $timeSeriesIdColumn;
}
/**
* @return string
*/
public function getTimeSeriesIdColumn()
{
return $this->timeSeriesIdColumn;
}
/**
* The time series id columns that were used during ARIMA model training.
*
* @param string[] $timeSeriesIdColumns
*/
public function setTimeSeriesIdColumns($timeSeriesIdColumns)
{
$this->timeSeriesIdColumns = $timeSeriesIdColumns;
}
/**
* @return string[]
*/
public function getTimeSeriesIdColumns()
{
return $this->timeSeriesIdColumns;
}
public function setTimeSeriesLengthFraction($timeSeriesLengthFraction)
{
$this->timeSeriesLengthFraction = $timeSeriesLengthFraction;
}
public function getTimeSeriesLengthFraction()
{
return $this->timeSeriesLengthFraction;
}
/**
* Column to be designated as time series timestamp for ARIMA model.
*
* @param string $timeSeriesTimestampColumn
*/
public function setTimeSeriesTimestampColumn($timeSeriesTimestampColumn)
{
$this->timeSeriesTimestampColumn = $timeSeriesTimestampColumn;
}
/**
* @return string
*/
public function getTimeSeriesTimestampColumn()
{
return $this->timeSeriesTimestampColumn;
}
/**
* Tree construction algorithm for boosted tree models.
*
* Accepted values: TREE_METHOD_UNSPECIFIED, AUTO, EXACT, APPROX, HIST
*
* @param self::TREE_METHOD_* $treeMethod
*/
public function setTreeMethod($treeMethod)
{
$this->treeMethod = $treeMethod;
}
/**
* @return self::TREE_METHOD_*
*/
public function getTreeMethod()
{
return $this->treeMethod;
}
/**
* Smoothing window size for the trend component. When a positive value is
* specified, a center moving average smoothing is applied on the history
* trend. When the smoothing window is out of the boundary at the beginning or
* the end of the trend, the first element or the last element is padded to
* fill the smoothing window before the average is applied.
*
* @param string $trendSmoothingWindowSize
*/
public function setTrendSmoothingWindowSize($trendSmoothingWindowSize)
{
$this->trendSmoothingWindowSize = $trendSmoothingWindowSize;
}
/**
* @return string
*/
public function getTrendSmoothingWindowSize()
{
return $this->trendSmoothingWindowSize;
}
/**
* User column specified for matrix factorization models.
*
* @param string $userColumn
*/
public function setUserColumn($userColumn)
{
$this->userColumn = $userColumn;
}
/**
* @return string
*/
public function getUserColumn()
{
return $this->userColumn;
}
/**
* The version aliases to apply in Vertex AI model registry. Always overwrite
* if the version aliases exists in a existing model.
*
* @param string[] $vertexAiModelVersionAliases
*/
public function setVertexAiModelVersionAliases($vertexAiModelVersionAliases)
{
$this->vertexAiModelVersionAliases = $vertexAiModelVersionAliases;
}
/**
* @return string[]
*/
public function getVertexAiModelVersionAliases()
{
return $this->vertexAiModelVersionAliases;
}
public function setWalsAlpha($walsAlpha)
{
$this->walsAlpha = $walsAlpha;
}
public function getWalsAlpha()
{
return $this->walsAlpha;
}
/**
* Whether to train a model from the last checkpoint.
*
* @param bool $warmStart
*/
public function setWarmStart($warmStart)
{
$this->warmStart = $warmStart;
}
/**
* @return bool
*/
public function getWarmStart()
{
return $this->warmStart;
}
/**
* User-selected XGBoost versions for training of XGBoost models.
*
* @param string $xgboostVersion
*/
public function setXgboostVersion($xgboostVersion)
{
$this->xgboostVersion = $xgboostVersion;
}
/**
* @return string
*/
public function getXgboostVersion()
{
return $this->xgboostVersion;
}
}
// Adding a class alias for backwards compatibility with the previous class name.
class_alias(TrainingOptions::class, 'Google_Service_Bigquery_TrainingOptions');