Joel Grus
Joel Grus
look at how `TokenCharactersIndexer.tokens_to_indices` works: https://github.com/allenai/allennlp/blob/master/allennlp/data/token_indexers/token_characters_indexer.py#L74 you'd basically just do that, except that you'd have to grab each `token.idx` and generate a second vector of offsets to return. in fact,...
I spent a few hours today investigating this, I couldn't get things 100% working, but I got some things working, here's what I did: https://gist.github.com/joelgrus/8e994d908816ff91cd3df3602389dc0b If you have success with...
we have AUC for binary classification https://github.com/allenai/allennlp/blob/master/allennlp/training/metrics/auc.py but I don't know of anyone who's implemented it for multi-label / multi-class classification, sorry.
we hope to get to this soon, but it's not an explicit priority right now. PR welcome.
next time I'm having you do my copy-editing. would you believe this one I did on purpose, for a couple of reasons 1. the "correct" way made the line much...
can you share the actual error message? it's hard to know what went wrong without it
most python ML people use `virtualenv` or `conda` to manage multiple python installations, and you should too!