Closed benieric closed 4 months ago
Description Getting error message like below in resource chaining flow when creating TrainingJob without resource chaining:
--------------------------------------------------------------------------- AttributeError Traceback (most recent call last) [/var/folders/2s/7lzvylzj7qn0s37k6q5pyx140000gr/T/ipykernel_5460/2324256432.py](https://file+.vscode-resource.vscode-cdn.net/var/folders/2s/7lzvylzj7qn0s37k6q5pyx140000gr/T/ipykernel_5460/2324256432.py) in ?() 5 OutputDataConfig, ResourceConfig, StoppingCondition 6 7 job_name_v3 = 'xgboost-iris-' + time.strftime("%Y-%m-%d-%H-%M-%S", time.gmtime()) 8 ----> 9 training_job = TrainingJob.create( 10 training_job_name=job_name_v3, 11 hyper_parameters={ 12 'objective': 'multi:softmax', [/Volumes/workplace/sagemaker-core/end_user_experience/../src/generated/resources.py](https://file+.vscode-resource.vscode-cdn.net/Volumes/workplace/sagemaker-core/src/generated/resources.py) in ?(*args, **kwargs) 9522 "type": "string" 9523 } 9524 } 9525 } -> 9526 return create_func(*args, **Base.get_updated_kwargs_with_configured_attributes(config_schema_for_resource, "TrainingJob", **kwargs)) [/Volumes/workplace/sagemaker-core/end_user_experience/../src/generated/resources.py](https://file+.vscode-resource.vscode-cdn.net/Volumes/workplace/sagemaker-core/src/generated/resources.py) in ?(cls, training_job_name, algorithm_specification, role_arn, output_data_config, resource_config, stopping_condition, hyper_parameters, input_data_config, vpc_config, tags, enable_network_isolation, enable_inter_container_traffic_encryption, enable_managed_spot_training, checkpoint_config, debug_hook_config, debug_rule_configurations, tensor_board_output_config, experiment_config, profiler_config, profiler_rule_configurations, environment, retry_strategy, remote_debug_config, infra_check_config, session, region) 9586 'RemoteDebugConfig': remote_debug_config, 9587 'InfraCheckConfig': infra_check_config, 9588 } 9589 -> 9590 operation_input_args = Base.populate_chained_attributes(resource_name='TrainingJob', operation_input_args=operation_input_args) ... --> 105 obj_dict = {snake_to_pascal(k): v for k, v in value.__dict__.items()} 106 updated_args[arg] = Base.populate_chained_attributes(resource_name=type(value).__name__, operation_input_args=obj_dict) 107 return updated_args AttributeError: 'dict' object has no attribute '__dict__'
Reproduce Run example notebook until this cell: https://github.com/aws/sagemaker-core/blob/main/end_user_experience/xgboost_training_demo.ipynb?short_path=db07d72#L225
Merged is fixed
Description Getting error message like below in resource chaining flow when creating TrainingJob without resource chaining:
Reproduce Run example notebook until this cell: https://github.com/aws/sagemaker-core/blob/main/end_user_experience/xgboost_training_demo.ipynb?short_path=db07d72#L225