Spaces:

evaluate-metric
/

comet

Runtime error

Different scores from the three examples (Full match, Partial match, No match) on the App page

by wentau - opened Mar 18, 2025

I am using unbabel-comet==2.2.4, with the wmt22-comet-da model, getting scores

Full match [1.0, 1.0]
Partial match [0.84, 0.97]
No match [0.47, 0.43]

Is the example using a different model?

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment