Settings Input / Output Advanced History

Sampling

Evaluation

Evaluation id

If empty, defaults to a random value. Limited to a-z, 0-9 and _. Specify same id as other Model Evaluation to overwrite.

Evaluation name

If empty, defaults to the date and time of evaluation

Labels

Can contain variables in ${} notation, which are expanded when the recipe is run.
A Model Evaluation will include dynamically generated evaluation:date, evaluation:model-name and evaluationDataset:dataset-name labels, that you may override there.

Model

Prediction type

Prediction column

Skip performance metrics computation

If the ground truth is not available, it's still possible to build the model evaluation with only data statistics.

Labels column

Weights column

One-vs-all metrics class averaging

Proba aware

If the probabilities are not available, decision charts, lift charts and calibration curves will not be computed.

If the probabilities are not available, calibration curves will not be computed.

Probabilities

Classes

Interpret empty prediction as

Leave empty if you want to exclude the empty prediction rows from performance computation

Threshold

See

Cost matrix

If the model predicts that {{desc.targetVariable || 'the target'}} is true
- and it is indeed true, the gain is
- but it is not true, the gain is
If the model predicts that {{desc.targetVariable || 'the target'}} is false
- and it is indeed false the gain is
- but it is actually true, the gain is

Sampling

Evaluation

Model

Custom Evaluation Metrics

Container configuration

Spark

Python environment

Concurrent activities

Feature type