One-hot vocab

[erika-alden]

I'd like to replicate the one-hot results, but when I run the command tape-train onehot fluorescence I get the error TypeError: __init__() missing 1 required positional argument: 'vocab_size'

How do I fix this/should I be using the --vocab_file flag? Not sure what file I would specify.

Thanks!

[rmrao]

Hmm I'm not entirely sure I've ever tested the onehot model in the pytorch version... you might need to play around with the model itself to get things working.