components finetune_common_validation

Common Validation Component

finetune_common_validation

Overview

Component to validate the finetune job against Validation Service

Version: 0.0.6

View in Studio: https://ml.azure.com/registries/azureml/components/finetune_common_validation/version/0.0.6

Inputs

component input: mlflow model path

Name	Description	Type	Default	Optional	Enum
mlflow_model_path	MLflow model asset path. Special characters like \ and ' are invalid in the parameter value.	mlflow_model		True

Data validation component input: training mltable

Name	Description	Type	Default	Optional	Enum
train_mltable_path	Path to the mltable of the training dataset.	mltable		False

optional component input: validation mltable

Name	Description	Type	Default	Optional	Enum
validation_mltable_path	Path to the mltable of the validation dataset.	mltable		True

component input: test mltable

Name	Description	Type	Default	Optional	Enum
test_mltable_path	Path to the mltable of the test dataset.	mltable		True
user_column_names	Comma separated list of column names to be used for training.	string		True

Compute validation

Name	Description	Type	Default	Optional	Enum
compute_preprocess	Compute to be used for preprocess eg. provide 'FT-Cluster' if your compute is named 'FT-Cluster'. Special characters like \ and ' are invalid in the parameter value. If compute cluster name is provided, instance_type field will be ignored and the respective cluster will be used.	string		True
instance_type_preprocess	Instance type to be used for preprocess component in case of serverless compute, eg. standard_d12_v2. The parameter compute_preprocess must be set to 'serverless' for instance_type to be used	string		True
compute_model_import	Compute to be used for model_import eg. provide 'FT-Cluster' if your compute is named 'FT-Cluster'	string		True
instance_type_model_import	Instance type to be used for model_import component in case of serverless compute, eg. standard_d12_v2. The parameter compute_model_import must be set to 'serverless' for instance_type to be used	string		True
compute_finetune	Compute to be used for finetuning eg. provide 'FT-Cluster' if your compute is named 'FT-Cluster'. Special characters like \ and ' are invalid in the parameter value. If compute cluster name is provided, instance_type field will be ignored and the respective cluster will be used	string		True
instance_type_finetune	Instance type to be used for finetune component in case of serverless compute, eg. standard_nc24rs_v3. The parameter compute_finetune must be set to 'serverless' for instance_type to be used	string		True
instance_count	Number of nodes to be used for finetuning (used for distributed training)	integer	1	True
process_count_per_instance	Number of gpus to be used per node for finetuning, should be equal to number of gpu per node in the compute SKU used for finetune	integer	1	True
compute_model_evaluation	Compute to be used for model evaluation eg. provide 'FT-Cluster' if your compute is named 'FT-Cluster'	string		True
instance_type_model_evaluation	Instance type to be used for model_evaluation components in case of serverless compute, eg. standard_nc24rs_v3. The parameter compute_model_evaluation must be set to 'serverless' for instance_type to be used	string		True
task_name	Which task the model is solving.	string			['tabular-classification', 'tabular-classification-multilabel', 'tabular-regression', 'text-classification', 'text-classification-multilabel', 'text-named-entity-recognition', 'text-summarization', 'question-answering', 'text-translation', 'text-generation', 'fill-mask', 'image-classification', 'image-classification-multilabel', 'image-object-detection', 'image-instance-segmentation', 'video-multi-object-tracking']

ME validation

Name	Description	Type	Default	Optional	Enum
test_batch_size	Test batch size.	integer	1	True
label_column_name	Label column name in provided test dataset, for example "label".	string	label	True
device		string	auto	False	['auto', 'cpu', 'gpu']
evaluation_config	Additional parameters for Computing Metrics.	uri_file		True
evaluation_config_params	Additional parameters as JSON serialized string.	string		True

Task Speciffic params validation

Name	Description	Type	Default	Optional	Enum
task_specific_extra_params	All extra params. The values should be key values pairs separated by semi-colon. For example "param1=value1;param2=value2"	string		True

Outputs

Name	Description	Type
validation_info	Validation status.	uri_file

Environment

azureml://registries/azureml/environments/acpt-pytorch-2.2-cuda12.1/versions/18