components automl_many_models_training_setup_step - Azure/azureml-assets GitHub Wiki
Version: 0.0.7
View in Studio: https://ml.azure.com/registries/azureml/components/automl_many_models_training_setup_step/version/0.0.7
| Name | Description | Type | Default | Optional | Enum |
|---|---|---|---|---|---|
| raw_data | Folder URI with inferencing test data. | uri_folder | |||
| automl_config | The automl config yml/json used for training. | uri_file | |||
| early_validation_failure | Skip early validation failure. | boolean | True | ||
| instance_count | Number of compute instances that used for AutoML training step. | integer | |||
| max_concurrency_per_instance | Number of processes that will be run concurrently on any given node. This number should not be larger than 1/2 of the number of cores in an individual node in the specified cluster | integer | |||
| parallel_step_timeout_in_seconds | The PRS step time out settings in secs. | integer | |||
| internal_skip_data_access | Disable copying access data in setup step. If this is disabled, the data should be directly point to the train step. | boolean | False | True | |
| enable_event_logger | Enable event logger. | boolean | True | True |
| Name | Description | Type |
|---|---|---|
| metadata | The metadata from train setup run. | uri_folder |
| processed_data | The processed data from train setup run. | uri_folder |
azureml://registries/azureml/environments/automl-gpu/labels/latest