Class: Aws::SageMaker::Types::StoppingCondition
- Inherits:
-
Struct
- Object
- Struct
- Aws::SageMaker::Types::StoppingCondition
- Includes:
- Aws::Structure
- Defined in:
- lib/aws-sdk-sagemaker/types.rb
Overview
When making an API call, you may pass StoppingCondition data as a hash:
{
max_runtime_in_seconds: 1,
}
Specifies how long model training can run. When model training reaches the limit, Amazon SageMaker ends the training job. Use this API to cap model training cost.
To stop a job, Amazon SageMaker sends the algorithm the ‘SIGTERM` signal, which delays job termination for120 seconds. Algorithms might use this 120-second window to save the model artifacts, so the results of training is not lost.
Training algorithms provided by Amazon SageMaker automatically saves the intermediate results of a model training job (it is best effort case, as model might not be ready to save as some stages, for example training just started). This intermediate data is a valid model artifact. You can use it to create a model (‘CreateModel`).
Instance Attribute Summary collapse
-
#max_runtime_in_seconds ⇒ Integer
The maximum length of time, in seconds, that the training job can run.
Instance Attribute Details
#max_runtime_in_seconds ⇒ Integer
The maximum length of time, in seconds, that the training job can run. If model training does not complete during this time, Amazon SageMaker ends the job. If value is not specified, default value is 1 day. Maximum value is 28 days.
10044 10045 10046 10047 |
# File 'lib/aws-sdk-sagemaker/types.rb', line 10044 class StoppingCondition < Struct.new( :max_runtime_in_seconds) include Aws::Structure end |