aws sagemaker create-training-job

Starts a model training job. AWS CLI reference

28 options Its page in the AWS CLI reference

Options

28 in the AWS CLI reference

Show its 28 options

--algorithm-specificationThe registry path of the Docker image that contains the training algorithm and algorithm-specific metadata, including the input mode.

--checkpoint-configContains information about the output location for managed spot training checkpoint data.

--debug-hook-configConfiguration information for the Amazon SageMaker Debugger hook parameters, metric and tensor collections, and storage paths.

--debug-rule-configurationsConfiguration information for Amazon SageMaker Debugger rules for debugging output tensors.

--enable-inter-container-traffic-encryptionTo encrypt all communications between ML compute instances in distributed training, choose True.

--enable-managed-spot-trainingTo train models using managed spot training, choose True.

--enable-network-isolationIsolates the training container.

--environmentThe environment variables to set in the Docker container.

--experiment-configAssociates a SageMaker job as a trial component with an experiment and trial.

--hyper-parametersAlgorithm-specific parameters that influence the quality of the model.

--infra-check-configContains information about the infrastructure health check configuration for the training job.

--input-data-configAn array of Channel objects.

--mlflow-configThe MLflow configuration using SageMaker managed MLflow.

--model-package-configThe configuration for the model package.

--output-data-configSpecifies the path to the S3 location where you want to store model artifacts.

--profiler-configConfiguration information for Amazon SageMaker Debugger system monitoring, framework profiling, and storage paths.

--profiler-rule-configurationsConfiguration information for Amazon SageMaker Debugger rules for profiling system and framework metrics.

--remote-debug-configConfiguration for remote debugging.

--resource-configThe resources, including the ML compute instances and ML storage volumes, to use for model training.

--retry-strategyThe number of times to retry the job when the job fails due to an InternalServerError.

--role-arnThe Amazon Resource Name (ARN) of an IAM role that SageMaker can assume to perform tasks on your behalf.

--serverless-job-configThe configuration for serverless training jobs.

--session-chaining-configContains information about attribute-based access control (ABAC) for the training job.

--stopping-conditionSpecifies a limit to how long a model training job can run.

--tagsAn array of key-value pairs.

--tensor-board-output-configConfiguration of storage locations for the Amazon SageMaker Debugger TensorBoard output data.

--training-job-nameThe name of the training job.

--vpc-configA VpcConfig object that specifies the VPC that you want your training job to connect to.