Hi, What would be the easiest way for me to evaluate the models performance on data? Specifically, I would like to see how many correct vs incorrect predictions the model made on the training set after training. Is there a way for me to do this within prodigy or is it better to export the model and annotations, and evaluate via sklearn?
Hi @klopez, if your model was trained from Prodigy, it's compatible with spaCy. Given that, you can use spacy evaluate to get some metrics for your model. Of course, the metrics will differ based on the type of model you're checking.
Thank you