ValueError: could not broadcast input array from shape (86935,256) into shape (87008,256) about adversarial_text HOT 5 OPEN

Thank you for your report!

Could you try to use following command?

$ cd data/imdb/
$ wget http://sato-motoki.com/research/vat/imdb_list.zip
$ unzip imdb_list.zip

from adversarial_text.

Hi,
Still got the same error. To help reproduce the error, i upload a notebook here
Thanks!

from adversarial_text.

Thank you for your notebook.

$ cd data/imdb/
$ wget http://sato-motoki.com/research/vat/imdb_list.zip
$ unzip imdb_list.zip

Then add encoding='utf-8' at every with open() in preprocessing.py

Please let me know the result!

from adversarial_text.

Hi,
Still got the same error. To help reproduce the error, i upload a notebook here
Thanks!

Did you fix it. And may I know how ?!

from adversarial_text.

No progress on this one? I can train my own LM than it seems to load the weights alright, but couldn't make it with the pretrained weights.

from adversarial_text.