PrecisionRecallCurve
One vs rest precision against recall, with average precision annotated.
Preferred over ROC under class imbalance: the false positive rate that ROC plots is divided by the (large) number of true negatives, so a rare-positive problem can show an excellent ROC curve while the model is mostly wrong whenever it does predict the positive class. Precision has no such denominator, so this curve stays honest as the classes skew.
References
Methods
compute(self, y_true, y_pred, class_names: Optional[List[str]] = None) -> List[DashAI.back.core.artifacts.Artifact]
PrecisionRecallCurveBuild the one vs rest precision recall curves.
Parameters
- y_true : ndarray
- Encoded true class indexes.
- y_pred : ndarray
- Class probability matrix.
- class_names : Optional[List[str]]
- Class labels in encoded order.
Returns
- List[Artifact]
- A single figure holding one curve per class.
get_credential(self, name: str)
ConfigObjectResolve a registered credential component by name.
Parameters
- name : str
- Credential component class name (e.g. "HuggingFaceCredential").
Returns
- BaseCredential
- An instance of the requested credential component.
get_metadata(cls) -> Dict[str, Any]
BaseReportGet metadata values for the current report.
Returns
- Dict[str, Any]
- UI metadata, including whether the report needs a model that outputs class probabilities.
get_schema(cls) -> dict
ConfigObjectGenerates the component related Json Schema.
Returns
- dict
- Dictionary representing the Json Schema of the component.
validate_and_transform(self, raw_data: dict) -> dict
ConfigObjectIt takes the data given by the user to initialize the model and returns it with all the objects that the model needs to work.
Parameters
- raw_data : dict
- A dictionary with the data provided by the user to initialize the model.
Returns
- dict
- A validated dictionary with the necessary objects.