Failing to run a training job in SageMaker - Unknown parameter in input: "ProfilerRuleConfigurations"

user_121956 0
Viewed 204

I tried to run a training job with the latest SageMaker SDK (2.24.5) in a clean virtual environment and am getting the error:

botocore.exceptions.ParamValidationError: Parameter validation failed: Unknown parameter in input: "ProfilerRuleConfigurations", must be one of: TrainingJobName, HyperParameters, AlgorithmSpecification, RoleArn, InputDataConfig, OutputDataConfig, ResourceConfig, VpcConfig, StoppingCondition, Tags, EnableNetworkIsolation, EnableInterContainerTrafficEncryption, EnableManagedSpotTraining, CheckpointConfig, DebugHookConfig, DebugRuleConfigurations, TensorBoardOutputConfig, ExperimentConfig, ProfilerConfig

amazon-sagemaker boto3
1 Answers

this issue sometimes occurs because a much older botocore version was first installed and old botocore models (where ProfilerRuleConfigurations is not defined) is cached somewhere, even if a new botocore version is installed. This caching doesn’t appear to be tied with the package itself, hence why setting up a new virtual environement doesn’t always fix the issue.

Check to see if a folder ~/.aws/models exists, and if it does, remove the folder and start a new virtual environment.

mv  ~/.aws/models /tmp
user_1219560
Related