Evaluation
Loading run…
Jump to id
⌘K
Run reconciliation
Evaluation
How accurate is it, measured?
Always the demo corpus run — the only one with ground truth