REST Resource: models

string

Output only. A hash of this resource.

object (ModelReference)

Required. Unique identifier for this model.

string (int64 format)

Output only. The time when this model was created, in millisecs since the epoch.

string (int64 format)

Output only. The time when this model was last modified, in millisecs since the epoch.

string

Optional. A user-friendly description of this model.

string

Optional. A descriptive name for this model.

map (key: string, value: string)

The labels associated with this model. You can use these to organize and group your models. Label keys and values can be no longer than 63 characters, can only contain lowercase letters, numeric characters, underscores and dashes. International characters are allowed. Label values are optional. Label keys must start with a letter and each label in the list must have a different key.

string (int64 format)

Optional. The time when this model expires, in milliseconds since the epoch. If not present, the model will persist indefinitely. Expired models will be deleted and their storage reclaimed. The defaultTableExpirationMs property of the encapsulating dataset can be used to set a default expirationTime on newly created models.

string

Output only. The geographic location where the model resides. This value is inherited from the dataset.

object (EncryptionConfiguration)

Custom encryption configuration (e.g., Cloud KMS keys). This shows the encryption configuration of the model data while stored in BigQuery storage. This field can be used with models.patch to update encryption key for an already encrypted model.

enum (ModelType)

Output only. Type of the model resource.

object (TrainingRun)

Information for all training runs in increasing order of startTime.

object (StandardSqlField)

Output only. Input feature columns for the model inference. If the model is trained with TRANSFORM clause, these are the input of the TRANSFORM clause.

object (StandardSqlField)

Output only. Label columns that were used to train this model. The output of the model will have a "predicted_" prefix to these columns.

object (TransformColumn)

Output only. This field will be populated if a TRANSFORM clause was used to train a model. TRANSFORM clause (if used) takes featureColumns as input and outputs transformColumns. transformColumns then are used to train the model.

object (HparamSearchSpaces)

Output only. All hyperparameter search spaces in this model.

string (int64 format)

The best trialId across all training runs.

string (int64 format)

Output only. The default trialId to use in TVFs when the trialId is not passed in. For single-objective hyperparameter tuning models, this is the best trial ID. For multi-objective hyperparameter tuning models, this is the smallest trial ID among all Pareto optimal trials.

object (HparamTuningTrial)

Output only. Trials of a hyperparameter tuning model sorted by trialId.

string (int64 format)

Output only. For single-objective hyperparameter tuning models, it only contains the best trial. For multi-objective hyperparameter tuning models, it contains all Pareto optimal trials sorted by trialId.

object (RemoteModelInfo)

Output only. Remote model info

string

Required. The ID of the project containing this model.

string

Required. The ID of the dataset containing this model.

string

Required. The ID of the model. The ID must contain only letters (a-z, A-Z), numbers (0-9), or underscores (_). The maximum length is 1,024 characters.

object (TrainingOptions)

Output only. Options that were used for this training run, includes user specified and default options that were used.

string (int64 format)

Output only. The start time of this training run, in milliseconds since epoch.

string (Timestamp format)

Output only. The start time of this training run.

object (IterationResult)

Output only. Output of each iteration run, results.size() <= maxIterations.

object (EvaluationMetrics)

Output only. The evaluation metrics over training/eval data that were computed at the end of training.

object (DataSplitResult)

Output only. Data split result of the training run. Only set when the input data is actually split.

object (GlobalExplanation)

Output only. Global explanation contains the explanation of top features on the model level. Applies to both regression and classification models.

object (GlobalExplanation)

Output only. Global explanation contains the explanation of top features on the class level. Applies to classification models only.

string

The model id in the Vertex AI Model Registry for this training run.

string

Output only. The model version in the Vertex AI Model Registry for this training run.

string (int64 format)

The maximum number of iterations in training. Used only for iterative training algorithms.

enum (LossType)

Type of loss function used during training run.

number

Learning rate in training. Used only for iterative training algorithms.

number

L1 regularization coefficient.

number

L2 regularization coefficient.

number

When earlyStop is true, stops training when accuracy improvement is less than 'minRelativeProgress'. Used only for iterative training algorithms.

boolean

Whether to train a model from the last checkpoint.

boolean

Whether to stop early when the loss doesn't improve significantly any more (compared to minRelativeProgress). Used only for iterative training algorithms.

string

Name of input label columns in training data.

enum (DataSplitMethod)

The data split type for training and evaluation, e.g. RANDOM.

number

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.

string

The column to split data with. This column won't be used as a feature. 1. When dataSplitMethod 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 dataSplitMethod 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

enum (LearnRateStrategy)

The strategy to determine learn rate for the current iteration.

number

Specifies the initial learning rate for the line search learn rate strategy.

map (key: string, value: number)

Weights associated with each label class, for rebalancing the training data. Only applicable for classification models.

string

User column specified for matrix factorization models.

string

Item column specified for matrix factorization models.

enum (DistanceType)

Distance type for clustering models.

string (int64 format)

Number of clusters for clustering models.

string

Google Cloud Storage URI from which the model was imported. Only applicable for imported models.

enum (OptimizationStrategy)

Optimization strategy for training linear regression models.

string (int64 format)

Hidden units for dnn models.

string (int64 format)

Batch size for dnn models.

number

Dropout probability for dnn models.

string (int64 format)

Maximum depth of a tree for boosted tree models.

number

Subsample fraction of the training data to grow tree to prevent overfitting for boosted tree models.

number

Minimum split loss for boosted tree models.

enum (BoosterType)

Booster type for boosted tree models.

string (Int64Value format)

Number of parallel trees constructed during each iteration for boosted tree models.

enum (DartNormalizeType)

Type of normalization algorithm for boosted tree models using dart booster.

enum (TreeMethod)

Tree construction algorithm for boosted tree models.

string (Int64Value format)

Minimum sum of instance weight needed in a child for boosted tree models.

number

Subsample ratio of columns when constructing each tree for boosted tree models.

number

Subsample ratio of columns for each level for boosted tree models.

number

Subsample ratio of columns for each node(split) for boosted tree models.

string (int64 format)

Num factors specified for matrix factorization models.

enum (FeedbackType)

Feedback type that specifies which algorithm to run for matrix factorization.

number

Hyperparameter for matrix factoration when implicit feedback type is specified.

enum (KmeansInitializationMethod)

The method used to initialize the centroids for kmeans algorithm.

string

The column used to provide the initial centroids for kmeans algorithm when kmeansInitializationMethod is CUSTOM.

string

Column to be designated as time series timestamp for ARIMA model.

string

Column to be designated as time series data for ARIMA model.

boolean

Whether to enable auto ARIMA or not.

object (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.

enum (DataFrequency)

The data frequency of a time series.

boolean

Whether or not p-value test should be computed for this model. Only available for linear and logistic regression models.

boolean

Include drift when fitting an ARIMA model.

enum (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.

enum (HolidayRegion)

A list of geographical regions that are used for time series modeling.

string

The time series id column that was used during ARIMA model training.

string

The time series id columns that were used during ARIMA model training.

string (int64 format)

The number of periods ahead that need to be forecasted.

string (int64 format)

The max value of the sum of non-seasonal p and q.

string (int64 format)

The min value of the sum of non-seasonal p and q.

string (int64 format)

Number of trials to run this hyperparameter tuning job.

string (int64 format)

Maximum number of trials to run in parallel.

enum (HparamTuningObjective)

The target evaluation metrics to optimize the hyperparameters for.

boolean

If true, perform decompose time series and save the results.

boolean

If true, clean spikes and dips in the input time series.

boolean

If true, detect step changes and make data adjustment in the input time series.

boolean

If true, enable global explanation during training.

string (int64 format)

Number of paths for the sampled Shapley explain method.

string (int64 format)

Number of integral steps for the integrated gradients explain method.

enum (EncodingMethod)

Categorical feature encoding method.

string

Based on the selected TF version, the corresponding docker image is used to train external models.

string

Name of the instance weight column for training data. This column isn't be used as a feature.

string (int64 format)

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.

number

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.

string (int64 format)

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.

string (int64 format)

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.

string

User-selected XGBoost versions for training of XGBoost models.

boolean

Whether to use approximate feature contribution method in XGBoost model explanation for global explain.

boolean

Whether the model should include intercept during model training.

string (int64 format)

Number of principal components to keep in the PCA model. Must be <= the number of features.

number

The minimum ratio of cumulative explained variance that needs to be given by the PCA model.

boolean

If true, scale the feature values by dividing the feature standard deviation. Currently only apply to PCA.

enum (PcaSolver)

The solver for PCA.

boolean

Whether to calculate class weights automatically based on the popularity of each label.

string

Activation function of the neural nets.

string

Optimizer used for training the neural nets.

number

Budget in hours for AutoML training.

boolean

Whether to standardize numerical features. Default to true.

number

L1 regularization coefficient to activations.

enum (ModelRegistry)

The model registry.

string

The version aliases to apply in Vertex AI model registry. Always overwrite if the version aliases exists in a existing model.

string

Optional. Names of the columns to slice on. Applies to contribution analysis models.

string

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.

string

Name of the column used to determine the rows corresponding to control and test. Applies to contribution analysis models.

number

The apriori support minimum. Applies to contribution analysis models.

string (Int64Value format)

Order of the autoregressive part.

string (Int64Value format)

Order of the differencing part.

string (Int64Value format)

Order of the moving-average part.

integer

Index of the iteration, 0 based.

string (Int64Value format)

Time taken to run the iteration in milliseconds.

number

Loss computed on the training data at the end of iteration.

number

Loss computed on the eval data at the end of iteration.

number

Learn rate used for this iteration.

object (ClusterInfo)

Information about top clusters for clustering models.

object (ArimaResult)

Arima result.

object (PrincipalComponentInfo)

The information of the principal components.

string (int64 format)

Centroid id.

number

Cluster radius, the average distance from centroid to each point assigned to the cluster.

string (Int64Value format)

Cluster size, the total number of points assigned to the cluster.

object (ArimaModelInfo)

This message is repeated because there are multiple arima models fitted in auto-arima. For non-auto-arima model, its size is one.

enum (SeasonalPeriodType)

Seasonal periods. Repeated because multiple periods are supported for one time series.

object (ArimaOrder)

Non-seasonal order.

object (ArimaCoefficients)

Arima coefficients.

object (ArimaFittingMetrics)

Arima fitting metrics.

boolean

Whether Arima model fitted with drift or not. It is always false when d is not 1.

string

The timeSeriesId value for this time series. It will be one of the unique values from the timeSeriesIdColumn specified during ARIMA model training. Only present when timeSeriesIdColumn training option was used.

string

The tuple of timeSeriesIds identifying this time series. It will be one of the unique tuples of values present in the timeSeriesIdColumns specified during ARIMA model training. Only present when timeSeriesIdColumns training option was used and the order of values here are same as the order of timeSeriesIdColumns.

enum (SeasonalPeriodType)

Seasonal periods. Repeated because multiple periods are supported for one time series.

boolean

If true, holiday_effect is a part of time series decomposition result.

boolean

If true, spikes_and_dips is a part of time series decomposition result.

boolean

If true, step_changes is a part of time series decomposition result.

number

Auto-regressive coefficients, an array of double.

number

Moving-average coefficients, an array of double.

number

Intercept coefficient, just a double not an array.

number

Log-likelihood.

number

AIC.

number

Variance.

string (Int64Value format)

Id of the principal component.

number

Explained variance by this principal component, which is simply the eigenvalue.

number

Explained_variance over the total explained variance.

number

The explainedVariance is pre-ordered in the descending order to compute the cumulative explained variance ratio.

object (RegressionMetrics)

Populated for regression models and explicit feedback type matrix factorization models.

object (BinaryClassificationMetrics)

Populated for binary classification/classifier models.

object (MultiClassClassificationMetrics)

Populated for multi-class classification/classifier models.

object (ClusteringMetrics)

Populated for clustering models.

object (RankingMetrics)

Populated for implicit feedback type matrix factorization models.

object (ArimaForecastingMetrics)

Populated for ARIMA models.

object (DimensionalityReductionMetrics)

Evaluation metrics when the model is a dimensionality reduction model, which currently includes PCA.

number

Mean absolute error.

number

Mean squared error.

number

Mean squared log error.

number

Median absolute error.

number

R^2 score. This corresponds to r2_score in ML.EVALUATE.

object (AggregateClassificationMetrics)

Aggregate classification metrics.

object (BinaryConfusionMatrix)

Binary confusion matrix at multiple thresholds.

string

Label representing the positive class.

string

Label representing the negative class.

number

Precision is the fraction of actual positive predictions that had positive actual labels. For multiclass this is a macro-averaged metric treating each class as a binary classifier.

number

Recall is the fraction of actual positive labels that were given a positive prediction. For multiclass this is a macro-averaged metric.

number

Accuracy is the fraction of predictions given the correct label. For multiclass this is a micro-averaged metric.

number

Threshold at which the metrics are computed. For binary classification models this is the positive class threshold. For multi-class classfication models this is the confidence threshold.

number

The F1 score is an average of recall and precision. For multiclass this is a macro-averaged metric.

number

Logarithmic Loss. For multiclass this is a macro-averaged metric.

number

Area Under a ROC Curve. For multiclass this is a macro-averaged metric.

number

Threshold value used when computing each of the following metric.

string (Int64Value format)

Number of true samples predicted as true.

string (Int64Value format)

Number of false samples predicted as true.

string (Int64Value format)

Number of true samples predicted as false.

string (Int64Value format)

Number of false samples predicted as false.

number

The fraction of actual positive predictions that had positive actual labels.

number

The fraction of actual positive labels that were given a positive prediction.

number

The equally weighted average of recall and precision.

number

The fraction of predictions given the correct label.

object (AggregateClassificationMetrics)

Aggregate classification metrics.

object (ConfusionMatrix)

Confusion matrix at different thresholds.

number

Confidence threshold used when computing the entries of the confusion matrix.

object (Row)

One row per actual label.

string

The original label of this row.

object (Entry)

Info describing predicted label distribution.

string

The predicted label. For confidenceThreshold > 0, we will also add an entry indicating the number of items under the confidence threshold.

string (Int64Value format)

Number of items being predicted as this label.

number

Davies-Bouldin index.

number

Mean of squared distances between each sample to its cluster centroid.

object (Cluster)

Information for all clusters.

string (int64 format)

Centroid id.

object (FeatureValue)

Values of highly variant features for this cluster.

string (Int64Value format)

Count of training data rows that were assigned to this cluster.

string

The feature column name.

number

The numerical feature value. This is the centroid value for this feature.

object (CategoricalValue)

The categorical feature value.

object (CategoryCount)

Counts of all categories for the categorical feature. If there are more than ten categories, we return top ten (by count) and return one more CategoryCount with category "_OTHER_" and count as aggregate counts of remaining categories.

string

The name of category.

string (Int64Value format)

The count of training samples matching the category within the cluster.

number

Calculates a precision per user for all the items by ranking them and then averages all the precisions across all the users.

number

Similar to the mean squared error computed in regression and explicit recommendation models except instead of computing the rating directly, the output from evaluate is computed against a preference which is 1 or 0 depending on if the rating exists or not.

number

A metric to determine the goodness of a ranking calculated from the predicted confidence by comparing it to an ideal rank measured by the original ratings.

number

Determines the goodness of a ranking by computing the percentile rank from the predicted confidence and dividing it by the original rank.

object (ArimaOrder)

Non-seasonal order.

object (ArimaFittingMetrics)

Arima model fitting metrics.

enum (SeasonalPeriodType)

Seasonal periods. Repeated because multiple periods are supported for one time series.

boolean

Whether Arima model fitted with drift or not. It is always false when d is not 1.

string

Id to differentiate different time series for the large-scale case.

object (ArimaSingleModelForecastingMetrics)

Repeated as there can be many metric sets (one for each model) in auto-arima and the large-scale case.

object (ArimaOrder)

Non-seasonal order.

object (ArimaFittingMetrics)

Arima fitting metrics.

boolean

Is arima model fitted with drift or not. It is always false when d is not 1.

string

The timeSeriesId value for this time series. It will be one of the unique values from the timeSeriesIdColumn specified during ARIMA model training. Only present when timeSeriesIdColumn training option was used.

string

The tuple of timeSeriesIds identifying this time series. It will be one of the unique tuples of values present in the timeSeriesIdColumns specified during ARIMA model training. Only present when timeSeriesIdColumns training option was used and the order of values here are same as the order of timeSeriesIdColumns.

enum (SeasonalPeriodType)

Seasonal periods. Repeated because multiple periods are supported for one time series.

boolean

If true, holiday_effect is a part of time series decomposition result.

boolean

If true, spikes_and_dips is a part of time series decomposition result.

boolean

If true, step_changes is a part of time series decomposition result.

number

Total percentage of variance explained by the selected principal components.

object (TableReference)

Table reference of the training data after split.

object (TableReference)

Table reference of the evaluation data after split.

object (TableReference)

Table reference of the test data after split.

object (Explanation)

A list of the top global explanations. Sorted by absolute value of attribution in descending order.

string

Class label for this set of global explanations. Will be empty/null for binary logistic and linear regression models. Sorted alphabetically in descending order.

string

The full feature name. For non-numerical features, will be formatted like <column_name>.<encoded_feature_name>. Overall size of feature name will always be truncated to first 120 characters.

number

Attribution of feature.

string

Output only. Name of the column.

object (StandardSqlDataType)

Output only. Data type of the column after the transform.

string

Output only. The SQL expression used in the column transform.

object (DoubleHparamSearchSpace)

Learning rate of training jobs.

object (DoubleHparamSearchSpace)

L1 regularization coefficient.

object (DoubleHparamSearchSpace)

L2 regularization coefficient.

object (IntHparamSearchSpace)

Number of clusters for k-means.

object (IntHparamSearchSpace)

Number of latent factors to train on.

object (IntArrayHparamSearchSpace)

Hidden units for neural network models.

object (IntHparamSearchSpace)

Mini batch sample size.

object (DoubleHparamSearchSpace)

Dropout probability for dnn model training and boosted tree models using dart booster.

object (IntHparamSearchSpace)

Maximum depth of a tree for boosted tree models.

object (DoubleHparamSearchSpace)

Subsample the training data to grow tree to prevent overfitting for boosted tree models.

object (DoubleHparamSearchSpace)

Minimum split loss for boosted tree models.

object (DoubleHparamSearchSpace)

Hyperparameter for matrix factoration when implicit feedback type is specified.

object (StringHparamSearchSpace)

Booster type for boosted tree models.

object (IntHparamSearchSpace)

Number of parallel trees for boosted tree models.

object (StringHparamSearchSpace)

Dart normalization type for boosted tree models.

object (StringHparamSearchSpace)

Tree construction algorithm for boosted tree models.

object (IntHparamSearchSpace)

Minimum sum of instance weight needed in a child for boosted tree models.

object (DoubleHparamSearchSpace)

Subsample ratio of columns when constructing each tree for boosted tree models.

object (DoubleHparamSearchSpace)

Subsample ratio of columns for each level for boosted tree models.

object (DoubleHparamSearchSpace)

Subsample ratio of columns for each node(split) for boosted tree models.

object (StringHparamSearchSpace)

Activation functions of neural network models.

object (StringHparamSearchSpace)

Optimizer of TF models.

object (DoubleRange)

Range of the double hyperparameter.

object (DoubleCandidates)

Candidates of the double hyperparameter.

number

Min value of the double parameter.

number

Max value of the double parameter.

number

Candidates for the double parameter in increasing order.

object (IntRange)

Range of the int hyperparameter.

object (IntCandidates)

Candidates of the int hyperparameter.

string (Int64Value format)

Min value of the int parameter.

string (Int64Value format)

Max value of the int parameter.

string (Int64Value format)

Candidates for the int parameter in increasing order.

object (IntArray)

Candidates for the int array parameter.

string (int64 format)

Elements in the int array.

string

Canididates for the string or enum parameter in lower case.

string (int64 format)

1-based index of the trial.

string (int64 format)

Starting time of the trial.

string (int64 format)

Ending time of the trial.

object (TrainingOptions)

The hyperprameters selected for this trial.

object (EvaluationMetrics)

Evaluation metrics of this trial calculated on the test data. Empty in Job API.

enum (TrialStatus)

The status of the trial.

string

Error message for FAILED and INFEASIBLE trial.

number

Loss computed on the training data at the end of trial.

number

Loss computed on the eval data at the end of trial.

object (EvaluationMetrics)

Hyperparameter tuning evaluation metrics of this trial calculated on the eval data. Unlike evaluationMetrics, only the fields corresponding to the hparamTuningObjectives are set.

string

Output only. Fully qualified name of the user-provided connection object of the remote model. Format: "projects/{projectId}/locations/{locationId}/connections/{connectionId}"

string (int64 format)

Output only. Max number of rows in each batch sent to the remote service. If unset, the number of rows in each batch is set dynamically.

string

Output only. The model version for LLM.

string

Output only. The endpoint for remote model.

enum (RemoteServiceType)

Output only. The remote service type for remote model.