When running InferSent with the latest pytorch version from master, I'm getting this warning:
InferSent/models.py:222: UserWarning: volatile was removed and now has no effect. Use `with torch.no_grad():` instead.
sentences[stidx:stidx + bsize]), volatile=True)
When I changed the code of the encode method that caused the warning to use the with toch.grad() method, everything seemed to work as expected.
When running InferSent with the latest pytorch version from master, I'm getting this warning:
When I changed the code of the
encode
method that caused the warning to use thewith toch.grad()
method, everything seemed to work as expected.