Output

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

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.

The input dataset must contain the past data at least {{ prettyTimeSteps((modelDetails.iperf.minTimeseriesSizeForScoring + modelDetails.coreParams.predictionLength) * modelDetails.coreParams.timestepParams.numberOfTimeunits, modelDetails.coreParams.timestepParams.timeunit) }} ({{ prettyTimeSteps(modelDetails.iperf.minTimeseriesSizeForScoring * modelDetails.coreParams.timestepParams.numberOfTimeunits, modelDetails.coreParams.timestepParams.timeunit) }} to make predictions and {{ prettyTimeSteps(modelDetails.coreParams.predictionLength * modelDetails.coreParams.timestepParams.numberOfTimeunits, modelDetails.coreParams.timestepParams.timeunit) }} to evaluate them) of past data (for every evaluation window, up to {{ prettyTimeSteps(modelDetails.iperf.maxUsedTimestepsForScoring * modelDetails.coreParams.timestepParams.numberOfTimeunits, modelDetails.coreParams.timestepParams.timeunit) }} will be used for forecasting)

Forecast horizon

{{ prettyTimeSteps(modelDetails.coreParams.predictionLength * modelDetails.coreParams.timestepParams.numberOfTimeunits, modelDetails.coreParams.timestepParams.timeunit) }} ({{ modelDetails.coreParams.predictionLength }} {{ 'step' | plurify: modelDetails.coreParams.predictionLength }} of {{ prettyTimeSteps(modelDetails.coreParams.timestepParams.numberOfTimeunits, modelDetails.coreParams.timestepParams.timeunit) }}).

Forecast quantiles

Output quantiles for every forecast time step

Not supported by algorithm.

Nb. evaluation timesteps

Max. nb. timesteps for evaluation. 0 for as many as possible.

Warning: Changing the number of evaluation timesteps may make comparisons with existing model evaluations meaningless.

Metrics to compute

Aggregated

Aggregate metrics over all time series from the input dataset.

Per time series

Show one row per time series from the input dataset. Also add time series identifier columns.

Refit model

Refit model on input data (only relevant for statistical models, mandatory for Seasonal trend).

Treatment variable

Control value

Treatment values

Preferred outcome class

Compute propensity score and metrics

No propensity model has been trained along the main causal model

Skip performance metrics computation

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

Output probabilities

output probabilities for each class in addition to the prediction

Output percentile

percentile, among test set, of the proba of class 1

Note: metrics are computed with the variable {{modelDetails.coreParams.weight.sampleWeightVariable}} as sample weight.

Custom Evaluation Metrics

Add custom evaluation metrics to score model on.

Computation will run with the code-env used to train the model: {{ modelEnvName }}

Warning: You do not have permission to write arbitrary code.

No settings are available for regression scoring

Evaluation dataset

Model

Threshold

Output

Custom Evaluation Metrics

Spark

Container configuration