steps.base.dataset.validator

validator

Functions:

Name	Description
validate_dataset	Validates a dataset or a dataset collection to ensure data integrity and correctness.

validate_dataset(dataset, fix_annotation=False)

Validates a dataset or a dataset collection to ensure data integrity and correctness.

This function checks each dataset in a collection or a single dataset for any issues. If annotation errors are found, it can attempt to fix them based on the provided fix_annotation flag. If validation fails for any dataset, an error is logged. The validation process is skipped for datasets without a validator.

Parameters:

Name	Type	Description	Default
dataset	Union[TBaseDataset, DatasetCollection]	The dataset or dataset collection to validate. If a DatasetCollection is provided, each individual dataset within the collection will be validated.	required
fix_annotation	bool	Flag to indicate whether to attempt fixing annotation errors. Defaults to False. If set to True, the function will try to correct any found annotation issues during validation.	False