COMET
COMET copied to clipboard
[QUESTION] Train UnifiedMetric/XCOMET with word level predictions.
What is your question?
I was trying to train a model both at sentence and word level, but was unable to find any examples of it. I believe XCOMET is able to do so, since it has a word_level_training
parameter, but I could not find any examples on how to feed the word level data to the model.
Would you have some examples available for this?
I still was unable to get word level predictions. Would it be possible to add an small example to reproduce?