​
Key concepts

  • Evaluator pipeline - the pipeline evaluating the results
  • Dataset format – two fixed JSON objects: target and data, each with any keys.

​
Flow overview

High-level overview of the evaluation flow

What happens if data and target have overlapping keys? Values in target have higher priority over and overwrite the values in data.

​
Detailed flow

For every datapoint in the dataset, evaluation does the following:

  1. Merge the data and target objects. If keys clash, values in target are selected.
  2. Set all the values in the resulting merged object as values to the input nodes of the evaluator pipeline. This is done by matching* the keys of the object to the input node names of the evaluator pipeline.
  3. Evaluator pipeline is run. All required env variables must be set for this to succeed.
  4. Evaluator pipeline produces a single numeric output. This is stored in the results of the evaluation.

​
Requirements

  • Evaluator pipeline has at least one commit.
  • Object obtained by merging target and data at least contains keys matching* the names of the input nodes on the evaluator pipeline.
  • All required environment variables for the evaluator run are set in the Env vars page.
  • Evaluator pipeline must produce one output. This output must be parsable into a number (8-bit float).

* Exact case-sensitive match