Spacy: Exception when training on GPU

Python version: 3.5.2
Models: en
spaCy version: 2.0.5
Platform: Linux-4.10.0-42-generic-x86_64-with-Ubuntu-16.04-xenial

Created on 30 Dec 2017 · 5Comments · Source: explosion/spaCy

Optimizer created like the following:

optimizer = nlp.begin_training(device=0)

call to update

    nlp.update(
                    [text],       
                    [annotations], 
                    drop=drop,    
                    sgd=optimizer,  
                    losses=losses)

throws exception

  File "/usr/local/lib/python3.5/dist-packages/plac_core.py", line 328, in call
    cmd, result = parser.consume(arglist)
  File "/usr/local/lib/python3.5/dist-packages/plac_core.py", line 207, in consume
    return cmd, self.func(*(args + varargs + extraopts), **kwargs)
  File "/home/andrey/Work/Synqq/skyNLP/src/train_spacy.py", line 119, in main
    losses=losses)
  File "/usr/local/lib/python3.5/dist-packages/spacy/language.py", line 407, in update
    proc.update(docs, golds, drop=drop, sgd=get_grads, losses=losses)
  File "nn_parser.pyx", line 609, in spacy.syntax.nn_parser.Parser.update
  File "nn_parser.pyx", line 712, in spacy.syntax.nn_parser.Parser._make_updates
  File "/usr/local/lib/python3.5/dist-packages/thinc/api.py", line 67, in continue_update
    gradient = callback(gradient, sgd)
  File "/usr/local/lib/python3.5/dist-packages/spacy/_ml.py", line 552, in finish_update
    return ops.unflatten(d_X, lengths, pad=0)
  File "ops.pyx", line 140, in thinc.neural.ops.Ops.unflatten
TypeError: slice indices must be integers or None or have an __index__ method