<model_name>!SHOW_CONFUSION_MATRIX¶

Returns a table containing the number of instances of each combination of actual class and predicted class in models where evaluation was enabled at instantiation. You can use this dataset to plot a confusion matrix. This method takes no arguments. See Confusion Matrix in show_confusion_matrix.

Output¶

Column	Type	Description
`dataset_type`	VARCHAR	The name of the dataset used for metrics calculation, currently EVAL.
`actual_class`	VARCHAR	The actual class.
`predicted_class`	VARCHAR	The predicted class.
`count`	INTEGER	The number of instances of the given combination of actual and predicted class.
`logs`	VARIANT	Contains error or warning messages.