Compute the health score metric for a metrics testing run.
Documentation Index
Fetch the complete documentation index at: https://docs.galileo.ai/llms.txt
Use this file to discover all available pages before exploring further.
The scorer's output type, used to dispatch the correct metric.
boolean, categorical, count, discrete, freeform, percentage, multilabel, retrieved_chunk_list_boolean, boolean_multilabel The scorer's scoreable_node_types. Determines which record type carries the score.
llm, retriever, tool, workflow, agent, control, trace, session Client-side pending MGT edits: {row_id: value}. Overrides committed dataset values.
Successful Response
macro_f1, micro_f1, mse, mae Primary health score metric value, or None if no valid rows.
Rows excluded because MGT or score could not be parsed.
Secondary metrics (MAE, RMSE, R², per-class F1, etc.).
Rows with a successful scorer result.
Rows with a non-null MGT value after overlay.
Rows with both a score and a MGT value (used for computation).