search

allenai / bilm-tf

Tensorflow implementation of contextualized word representations from bi-directional language models
Apache License 2.0
1.62k stars 452 forks source link

Not able to start training #180

Closed 008karan closed 5 years ago

008karan commented 5 years ago

Traceback (most recent call last): File "bin/train_elmo.py", line 6, in <module> from bilm.training import train, load_options_latest_checkpoint, load_vocab File "/home/gamut/Downloads/bilm-tf-master/bin/bilm/__init__.py", line 2, in <module> from .data import Batcher, TokenBatcher File "/home/gamut/Downloads/bilm-tf-master/bin/bilm/data.py", line 197 def __init__(self, lm_vocab_file: str, max_token_length: int): SyntaxError: invalid syntax

huhuigou commented 5 years ago

have you solved the problem?

008karan commented 5 years ago

Yes. It was issue of cudnn version mismatch with TensorFlow.