A calibration run measures agreement between an evaluator and human annotations on a dataset. Runs execute asynchronously: a freshly created run has status: "pending"; poll get until it is completed or failed to read its metrics.
status: "pending"
get
completed
failed
metrics
A calibration run measures agreement between an evaluator and human annotations on a dataset. Runs execute asynchronously: a freshly created run has
status: "pending"; pollgetuntil it iscompletedorfailedto read itsmetrics.