The output of train textcat only contains Loss and F1-scores. How can I print the precision and recall? I can't find a way to change the code in train.py and didn't find any configuration in the prodigy.json file. Thanks. Here is the command I used -
prodigy train textcat demo_pattern_train,demo_neg,demo_neg_2,demo_train_2,demo_eval_neg blank:en -o /tmp/textcat_demo/ --n-iter 20Yes, I think at the moment the prodigy train command only reports the final score because the textcat scores differ depending on the evaluation metric used (which again, depends on the data, labels and config). You can read more about it here under textcat_score: https://spacy.io/api/scorer#properties
In the meantime, spacy train gives you more metrics, or you can write your own script and use the Language.evaluate directly (which returns an instance of the Scorer).