Result Class

ml_debugger.result.base.result.Result

Bases: ABC

Base class for ML evaluation results.

Manages polling for evaluation completion, issue retrieval, and provides abstract methods for summarizing and viewing results.

Attributes:

Name	Type	Description
result_id	str	Unique evaluation job identifier.
result_name	str	Alias for the evaluation result.
model_name	str	Name of the evaluated model.
version_name	str	Version identifier of the evaluated model.
service_client	ServiceClient	Client interface for API calls.
evaluation_completed	bool	True once evaluation has finished.
issues_data	Optional[Dict[str, Any]]	Cached issues once retrieved.

__init__(result_id, result_name, model_name, version_name, service_client)

Initialize the Result and wait for evaluation completion.

Parameters:

Name	Type	Description	Default
result_id	str	Unique identifier of the evaluation job.	required
result_name	str	User-facing alias for the result.	required
model_name	str	Name of the evaluated model.	required
version_name	str	Version of the evaluated model.	required
service_client	ServiceClient	Client for service API interactions.	required

metrics_summary(dataset_type=None)

issue_category_summary(dataset_type=None)

get_summary(*args, **kwargs) abstractmethod

Print a formatted summary of evaluation results to stdout.

get_issues()

Retrieve list of issues.

Returns:

Type	Description
DataFrame	The final evaluation result from the server

get_view(*args, **kwargs) abstractmethod

Retrieve the final evaluation result.

Returns:

Type	Description
DataFrame	pd.DataFrame: summary of the evaluation result