Giskard ML Scanner

Options

Model type

The type of model that is being evaluated.

Available options:

• Classification: Whether to test the results of a classification model.
• Regression: Whether to test the results of a regression model.

Dataset name

The name of the dataset. Used in the generated report.

Model name

The model name. Used in the generated report

Target column

The ground truth column from the validation table.

Classification labels

Comma separated list of classification labels without spaces. The order must match the probability columns selected in the prediction columns.

Prediction columns

The columns in the output table of the workflow that represent the model prediction.

• For classification models these contain the class probabilities.
• For regression models it's a single column containing the values predicted by the model.

Feature columns

The columns of your dataset that are used as features by the model.

Views

Giskard report

The Giskard scan report as HTML.

Workflows

• No workflows found

Links

• No links available

Developers