Class: Google::Apis::BigqueryV2::TrainingOptions

Inherits:
Object
  • Object
show all
Includes:
Core::Hashable, Core::JsonObjectSupport
Defined in:
lib/google/apis/bigquery_v2/classes.rb,
lib/google/apis/bigquery_v2/representations.rb,
lib/google/apis/bigquery_v2/representations.rb

Overview

Options used in model training.

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(**args) ⇒ TrainingOptions

Returns a new instance of TrainingOptions.



7170
7171
7172
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7170

def initialize(**args)
   update!(**args)
end

Instance Attribute Details

#adjust_step_changes ⇒ Boolean Also known as: adjust_step_changes?

If true, detect step changes and make data adjustment in the input time series. Corresponds to the JSON property adjustStepChanges

Returns:

  • (Boolean) 


6909
6910
6911
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6909

def adjust_step_changes
  @adjust_step_changes
end

#auto_arima ⇒ Boolean Also known as: auto_arima?

Whether to enable auto ARIMA or not. Corresponds to the JSON property autoArima

Returns:

  • (Boolean) 


6915
6916
6917
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6915

def auto_arima
  @auto_arima
end

#auto_arima_max_order ⇒ Fixnum

The max value of non-seasonal p and q. Corresponds to the JSON property autoArimaMaxOrder

Returns:

  • (Fixnum) 


6921
6922
6923
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6921

def auto_arima_max_order
  @auto_arima_max_order
end

#batch_size ⇒ Fixnum

Batch size for dnn models. Corresponds to the JSON property batchSize

Returns:

  • (Fixnum) 


6926
6927
6928
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6926

def batch_size
  @batch_size
end

#clean_spikes_and_dips ⇒ Boolean Also known as: clean_spikes_and_dips?

If true, clean spikes and dips in the input time series. Corresponds to the JSON property cleanSpikesAndDips

Returns:

  • (Boolean) 


6931
6932
6933
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6931

def clean_spikes_and_dips
  @clean_spikes_and_dips
end

#data_frequency ⇒ String

The data frequency of a time series. Corresponds to the JSON property dataFrequency

Returns:

  • (String) 


6937
6938
6939
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6937

def data_frequency
  @data_frequency
end

#data_split_column ⇒ String

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 Corresponds to the JSON property dataSplitColumn

Returns:

  • (String) 


6949
6950
6951
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6949

def data_split_column
  @data_split_column
end

#data_split_eval_fraction ⇒ Float

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. Corresponds to the JSON property dataSplitEvalFraction

Returns:

  • (Float) 


6956
6957
6958
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6956

def data_split_eval_fraction
  @data_split_eval_fraction
end

#data_split_method ⇒ String

The data split type for training and evaluation, e.g. RANDOM. Corresponds to the JSON property dataSplitMethod

Returns:

  • (String) 


6961
6962
6963
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6961

def data_split_method
  @data_split_method
end

#decompose_time_series ⇒ Boolean Also known as: decompose_time_series?

If true, perform decompose time series and save the results. Corresponds to the JSON property decomposeTimeSeries

Returns:

  • (Boolean) 


6966
6967
6968
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6966

def decompose_time_series
  @decompose_time_series
end

#distance_type ⇒ String

Distance type for clustering models. Corresponds to the JSON property distanceType

Returns:

  • (String) 


6972
6973
6974
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6972

def distance_type
  @distance_type
end

#dropout ⇒ Float

Dropout probability for dnn models. Corresponds to the JSON property dropout

Returns:

  • (Float) 


6977
6978
6979
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6977

def dropout
  @dropout
end

#early_stop ⇒ Boolean Also known as: early_stop?

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

Returns:

  • (Boolean) 


6984
6985
6986
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6984

def early_stop
  @early_stop
end

#feedback_type ⇒ String

Feedback type that specifies which algorithm to run for matrix factorization. Corresponds to the JSON property feedbackType

Returns:

  • (String) 


6990
6991
6992
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6990

def feedback_type
  @feedback_type
end

#hidden_units ⇒ Array<Fixnum>

Hidden units for dnn models. Corresponds to the JSON property hiddenUnits

Returns:

  • (Array<Fixnum>) 


6995
6996
6997
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 6995

def hidden_units
  @hidden_units
end

#holiday_region ⇒ String

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. Corresponds to the JSON property holidayRegion

Returns:

  • (String) 


7002
7003
7004
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7002

def holiday_region
  @holiday_region
end

#horizon ⇒ Fixnum

The number of periods ahead that need to be forecasted. Corresponds to the JSON property horizon

Returns:

  • (Fixnum) 


7007
7008
7009
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7007

def horizon
  @horizon
end

#include_drift ⇒ Boolean Also known as: include_drift?

Include drift when fitting an ARIMA model. Corresponds to the JSON property includeDrift

Returns:

  • (Boolean) 


7012
7013
7014
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7012

def include_drift
  @include_drift
end

#initial_learn_rate ⇒ Float

Specifies the initial learning rate for the line search learn rate strategy. Corresponds to the JSON property initialLearnRate

Returns:

  • (Float) 


7018
7019
7020
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7018

def initial_learn_rate
  @initial_learn_rate
end

#input_label_columns ⇒ Array<String>

Name of input label columns in training data. Corresponds to the JSON property inputLabelColumns

Returns:

  • (Array<String>) 


7023
7024
7025
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7023

def input_label_columns
  @input_label_columns
end

#item_column ⇒ String

Item column specified for matrix factorization models. Corresponds to the JSON property itemColumn

Returns:

  • (String) 


7028
7029
7030
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7028

def item_column
  @item_column
end

#kmeans_initialization_column ⇒ String

The column used to provide the initial centroids for kmeans algorithm when kmeans_initialization_method is CUSTOM. Corresponds to the JSON property kmeansInitializationColumn

Returns:

  • (String) 


7034
7035
7036
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7034

def kmeans_initialization_column
  @kmeans_initialization_column
end

#kmeans_initialization_method ⇒ String

The method used to initialize the centroids for kmeans algorithm. Corresponds to the JSON property kmeansInitializationMethod

Returns:

  • (String) 


7039
7040
7041
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7039

def kmeans_initialization_method
  @kmeans_initialization_method
end

#l1_regularization ⇒ Float

L1 regularization coefficient. Corresponds to the JSON property l1Regularization

Returns:

  • (Float) 


7044
7045
7046
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7044

def l1_regularization
  @l1_regularization
end

#l2_regularization ⇒ Float

L2 regularization coefficient. Corresponds to the JSON property l2Regularization

Returns:

  • (Float) 


7049
7050
7051
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7049

def l2_regularization
  @l2_regularization
end

#label_class_weights ⇒ Hash<String,Float>

Weights associated with each label class, for rebalancing the training data. Only applicable for classification models. Corresponds to the JSON property labelClassWeights

Returns:

  • (Hash<String,Float>) 


7055
7056
7057
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7055

def label_class_weights
  @label_class_weights
end

#learn_rate ⇒ Float

Learning rate in training. Used only for iterative training algorithms. Corresponds to the JSON property learnRate

Returns:

  • (Float) 


7060
7061
7062
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7060

def learn_rate
  @learn_rate
end

#learn_rate_strategy ⇒ String

The strategy to determine learn rate for the current iteration. Corresponds to the JSON property learnRateStrategy

Returns:

  • (String) 


7065
7066
7067
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7065

def learn_rate_strategy
  @learn_rate_strategy
end

#loss_type ⇒ String

Type of loss function used during training run. Corresponds to the JSON property lossType

Returns:

  • (String) 


7070
7071
7072
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7070

def loss_type
  @loss_type
end

#max_iterations ⇒ Fixnum

The maximum number of iterations in training. Used only for iterative training algorithms. Corresponds to the JSON property maxIterations

Returns:

  • (Fixnum) 


7076
7077
7078
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7076

def max_iterations
  @max_iterations
end

#max_tree_depth ⇒ Fixnum

Maximum depth of a tree for boosted tree models. Corresponds to the JSON property maxTreeDepth

Returns:

  • (Fixnum) 


7081
7082
7083
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7081

def max_tree_depth
  @max_tree_depth
end

#min_relative_progress ⇒ Float

When early_stop is true, stops training when accuracy improvement is less than 'min_relative_progress'. Used only for iterative training algorithms. Corresponds to the JSON property minRelativeProgress

Returns:

  • (Float) 


7087
7088
7089
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7087

def min_relative_progress
  @min_relative_progress
end

#min_split_loss ⇒ Float

Minimum split loss for boosted tree models. Corresponds to the JSON property minSplitLoss

Returns:

  • (Float) 


7092
7093
7094
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7092

def min_split_loss
  @min_split_loss
end

#model_uri ⇒ String

Google Cloud Storage URI from which the model was imported. Only applicable for imported models. Corresponds to the JSON property modelUri

Returns:

  • (String) 


7098
7099
7100
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7098

def model_uri
  @model_uri
end

#non_seasonal_order ⇒ Google::Apis::BigqueryV2::ArimaOrder

Arima order, can be used for both non-seasonal and seasonal parts. Corresponds to the JSON property nonSeasonalOrder

Returns:



7103
7104
7105
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7103

def non_seasonal_order
  @non_seasonal_order
end

#num_clusters ⇒ Fixnum

Number of clusters for clustering models. Corresponds to the JSON property numClusters

Returns:

  • (Fixnum) 


7108
7109
7110
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7108

def num_clusters
  @num_clusters
end

#num_factors ⇒ Fixnum

Num factors specified for matrix factorization models. Corresponds to the JSON property numFactors

Returns:

  • (Fixnum) 


7113
7114
7115
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7113

def num_factors
  @num_factors
end

#optimization_strategy ⇒ String

Optimization strategy for training linear regression models. Corresponds to the JSON property optimizationStrategy

Returns:

  • (String) 


7118
7119
7120
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7118

def optimization_strategy
  @optimization_strategy
end

#preserve_input_structs ⇒ Boolean Also known as: preserve_input_structs?

Whether to preserve the input structs in output feature names. Suppose there is a struct A with field b. When false (default), the output feature name is A_b. When true, the output feature name is A.b. Corresponds to the JSON property preserveInputStructs

Returns:

  • (Boolean) 


7125
7126
7127
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7125

def preserve_input_structs
  @preserve_input_structs
end

#subsample ⇒ Float

Subsample fraction of the training data to grow tree to prevent overfitting for boosted tree models. Corresponds to the JSON property subsample

Returns:

  • (Float) 


7132
7133
7134
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7132

def subsample
  @subsample
end

#time_series_data_column ⇒ String

Column to be designated as time series data for ARIMA model. Corresponds to the JSON property timeSeriesDataColumn

Returns:

  • (String) 


7137
7138
7139
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7137

def time_series_data_column
  @time_series_data_column
end

#time_series_id_column ⇒ String

The time series id column that was used during ARIMA model training. Corresponds to the JSON property timeSeriesIdColumn

Returns:

  • (String) 


7142
7143
7144
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7142

def time_series_id_column
  @time_series_id_column
end

#time_series_id_columns ⇒ Array<String>

The time series id columns that were used during ARIMA model training. Corresponds to the JSON property timeSeriesIdColumns

Returns:

  • (Array<String>) 


7147
7148
7149
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7147

def time_series_id_columns
  @time_series_id_columns
end

#time_series_timestamp_column ⇒ String

Column to be designated as time series timestamp for ARIMA model. Corresponds to the JSON property timeSeriesTimestampColumn

Returns:

  • (String) 


7152
7153
7154
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7152

def time_series_timestamp_column
  @time_series_timestamp_column
end

#user_column ⇒ String

User column specified for matrix factorization models. Corresponds to the JSON property userColumn

Returns:

  • (String) 


7157
7158
7159
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7157

def user_column
  @user_column
end

#wals_alpha ⇒ Float

Hyperparameter for matrix factoration when implicit feedback type is specified. Corresponds to the JSON property walsAlpha

Returns:

  • (Float) 


7162
7163
7164
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7162

def wals_alpha
  @wals_alpha
end

#warm_start ⇒ Boolean Also known as: warm_start?

Whether to train a model from the last checkpoint. Corresponds to the JSON property warmStart

Returns:

  • (Boolean) 


7167
7168
7169
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7167

def warm_start
  @warm_start
end

Instance Method Details

#update!(**args) ⇒ Object

Update properties of this object



7175
7176
7177
7178
7179
7180
7181
7182
7183
7184
7185
7186
7187
7188
7189
7190
7191
7192
7193
7194
7195
7196
7197
7198
7199
7200
7201
7202
7203
7204
7205
7206
7207
7208
7209
7210
7211
7212
7213
7214
7215
7216
7217
7218
7219
7220
7221
7222
7223
 
# File 'lib/google/apis/bigquery_v2/classes.rb', line 7175

def update!(**args)
  @adjust_step_changes = args[:adjust_step_changes] if args.key?(:adjust_step_changes)
  @auto_arima = args[:auto_arima] if args.key?(:auto_arima)
  @auto_arima_max_order = args[:auto_arima_max_order] if args.key?(:auto_arima_max_order)
  @batch_size = args[:batch_size] if args.key?(:batch_size)
  @clean_spikes_and_dips = args[:clean_spikes_and_dips] if args.key?(:clean_spikes_and_dips)
  @data_frequency = args[:data_frequency] if args.key?(:data_frequency)
  @data_split_column = args[:data_split_column] if args.key?(:data_split_column)
  @data_split_eval_fraction = args[:data_split_eval_fraction] if args.key?(:data_split_eval_fraction)
  @data_split_method = args[:data_split_method] if args.key?(:data_split_method)
  @decompose_time_series = args[:decompose_time_series] if args.key?(:decompose_time_series)
  @distance_type = args[:distance_type] if args.key?(:distance_type)
  @dropout = args[:dropout] if args.key?(:dropout)
  @early_stop = args[:early_stop] if args.key?(:early_stop)
  @feedback_type = args[:feedback_type] if args.key?(:feedback_type)
  @hidden_units = args[:hidden_units] if args.key?(:hidden_units)
  @holiday_region = args[:holiday_region] if args.key?(:holiday_region)
  @horizon = args[:horizon] if args.key?(:horizon)
  @include_drift = args[:include_drift] if args.key?(:include_drift)
  @initial_learn_rate = args[:initial_learn_rate] if args.key?(:initial_learn_rate)
  @input_label_columns = args[:input_label_columns] if args.key?(:input_label_columns)
  @item_column = args[:item_column] if args.key?(:item_column)
  @kmeans_initialization_column = args[:kmeans_initialization_column] if args.key?(:kmeans_initialization_column)
  @kmeans_initialization_method = args[:kmeans_initialization_method] if args.key?(:kmeans_initialization_method)
  @l1_regularization = args[:l1_regularization] if args.key?(:l1_regularization)
  @l2_regularization = args[:l2_regularization] if args.key?(:l2_regularization)
  @label_class_weights = args[:label_class_weights] if args.key?(:label_class_weights)
  @learn_rate = args[:learn_rate] if args.key?(:learn_rate)
  @learn_rate_strategy = args[:learn_rate_strategy] if args.key?(:learn_rate_strategy)
  @loss_type = args[:loss_type] if args.key?(:loss_type)
  @max_iterations = args[:max_iterations] if args.key?(:max_iterations)
  @max_tree_depth = args[:max_tree_depth] if args.key?(:max_tree_depth)
  @min_relative_progress = args[:min_relative_progress] if args.key?(:min_relative_progress)
  @min_split_loss = args[:min_split_loss] if args.key?(:min_split_loss)
  @model_uri = args[:model_uri] if args.key?(:model_uri)
  @non_seasonal_order = args[:non_seasonal_order] if args.key?(:non_seasonal_order)
  @num_clusters = args[:num_clusters] if args.key?(:num_clusters)
  @num_factors = args[:num_factors] if args.key?(:num_factors)
  @optimization_strategy = args[:optimization_strategy] if args.key?(:optimization_strategy)
  @preserve_input_structs = args[:preserve_input_structs] if args.key?(:preserve_input_structs)
  @subsample = args[:subsample] if args.key?(:subsample)
  @time_series_data_column = args[:time_series_data_column] if args.key?(:time_series_data_column)
  @time_series_id_column = args[:time_series_id_column] if args.key?(:time_series_id_column)
  @time_series_id_columns = args[:time_series_id_columns] if args.key?(:time_series_id_columns)
  @time_series_timestamp_column = args[:time_series_timestamp_column] if args.key?(:time_series_timestamp_column)
  @user_column = args[:user_column] if args.key?(:user_column)
  @wals_alpha = args[:wals_alpha] if args.key?(:wals_alpha)
  @warm_start = args[:warm_start] if args.key?(:warm_start)
end