Class: Google::Apis::BigqueryV2::BqmlIterationResult

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

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(**args) ⇒ BqmlIterationResult

Returns a new instance of BqmlIterationResult.



895
896
897
 
# File 'generated/google/apis/bigquery_v2/classes.rb', line 895

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

Instance Attribute Details

#duration_ms ⇒ Fixnum

[Output-only, Beta] Time taken to run the training iteration in milliseconds. Corresponds to the JSON property durationMs

Returns:

  • (Fixnum) 


867
868
869
 
# File 'generated/google/apis/bigquery_v2/classes.rb', line 867

def duration_ms
  @duration_ms
end

#eval_loss ⇒ Float

[Output-only, Beta] Eval loss computed on the eval data at the end of the iteration. The eval loss is used for early stopping to avoid overfitting. No eval loss if eval_split_method option is specified as no_split or auto_split with input data size less than 500 rows. Corresponds to the JSON property evalLoss

Returns:

  • (Float) 


875
876
877
 
# File 'generated/google/apis/bigquery_v2/classes.rb', line 875

def eval_loss
  @eval_loss
end

#index ⇒ Fixnum

[Output-only, Beta] Index of the ML training iteration, starting from zero for each training run. Corresponds to the JSON property index

Returns:

  • (Fixnum) 


881
882
883
 
# File 'generated/google/apis/bigquery_v2/classes.rb', line 881

def index
  @index
end

#learn_rate ⇒ Float

[Output-only, Beta] Learning rate used for this iteration, it varies for different training iterations if learn_rate_strategy option is not constant. Corresponds to the JSON property learnRate

Returns:

  • (Float) 


887
888
889
 
# File 'generated/google/apis/bigquery_v2/classes.rb', line 887

def learn_rate
  @learn_rate
end

#training_loss ⇒ Float

[Output-only, Beta] Training loss computed on the training data at the end of the iteration. The training loss function is defined by model type. Corresponds to the JSON property trainingLoss

Returns:

  • (Float) 


893
894
895
 
# File 'generated/google/apis/bigquery_v2/classes.rb', line 893

def training_loss
  @training_loss
end

Instance Method Details

#update!(**args) ⇒ Object

Update properties of this object



900
901
902
903
904
905
906
 
# File 'generated/google/apis/bigquery_v2/classes.rb', line 900

def update!(**args)
  @duration_ms = args[:duration_ms] if args.key?(:duration_ms)
  @eval_loss = args[:eval_loss] if args.key?(:eval_loss)
  @index = args[:index] if args.key?(:index)
  @learn_rate = args[:learn_rate] if args.key?(:learn_rate)
  @training_loss = args[:training_loss] if args.key?(:training_loss)
end