A numeric detail generated by the model for each label.
A numeric detail generated by the model for each label. Typically, this is a value which is minimized or maximized to determine the predicted value. e.g. confidence probabilities (to be maximised) or distance to the labelled cluster (to be minimised).
This MUST be in the same order as the list returned by "labels".
A list of numeric details corresponding to the labels (order must be preserved).
Used in tests to compare results, within a given numerical tolerance range.
Used in tests to compare results, within a given numerical tolerance range.
The list of potential labels that can be generated by the model. e.g. ["yes", "no"]
The value is the arg max of the confidences.