FileNotFoundError: [Errno 2] No such file or directory: 'cached_train_BertTokenizer_180.lock'

[Stimmot]

I want to train the model bert-base-german-cased on some documents, but when I try to run run_ner.py with the config.json it tells me, that it can't find the file mentioned above. I don't quite know what's the issue here, because it worked the last time I tried. Do I have to tell the model it shouldn't use any cached files? I tried that with the overwrite_cache flag. Does anyone have a clue what could be the problem?

[LysandreJik]

Hi! Could you provide the information related to your environment, as well as the command that you used to launch the script, as it's requested in the issue template? Thank you.

[Stimmot]

Yes sure!

• transformers version: 3.5.1
• Platform: Linux-5.9.1-kd-cluster-x86_64-with-glibc2.10
• Python version: 3.8.0
• PyTorch version (GPU?): 1.7.0 (True)
• Tensorflow version (GPU?): not installed (NA)
• Using GPU in script?: Yes
• Using distributed or parallel set-up in script?: No

Model I am using: BERT, specifically "bert-base-german-cased"

The problem arises when using:

• [x] the official example scripts: (give details below)
• [ ] my own modified scripts: (give details below)

The tasks I am working on is:

• [ ] an official GLUE/SQUaD task: (give the name)
• [x] my own task or dataset: (give details below)

Traceback: `Traceback (most recent call last): File "run_ner.py", line 324, in main() File "run_ner.py", line 187, in main TokenClassificationDataset( File "/home/IAIS/tschmude/bert_remote/examples/token-classification/utils_ner.py", line 240, in init with FileLock(lock_path): File "/home/IAIS/tschmude/anaconda3/envs/bert_env_remote/lib/python3.8/site-packages/filelock.py", line 323, in enter self.acquire() File "/home/IAIS/tschmude/anaconda3/envs/bert_env_remote/lib/python3.8/site-packages/filelock.py", line 271, in acquire self._acquire() File "/home/IAIS/tschmude/anaconda3/envs/bert_env_remote/lib/python3.8/site-packages/filelock.py", line 384, in _acquire fd = os.open(self._lock_file, open_mode) FileNotFoundError: [Errno 2] No such file or directory: '/home/tschmude/PycharmProjects/smart-sentencing/examples/token-classification/Data processing scripts/Data_Preprocessed/cached_train_BertTokenizer_180.lock'

Expected behavior

I'm running python run_ner.py Data/config.json to train the model for custom NER recognition. I have a couple self defined labels. It has worked before, but I can't quite tell what has changed since then. I already deleted cached .lock files that I could find.

[LysandreJik]

Would you mind providing the config.json as well, given that it contains your launch command? Thank you!

[Stimmot]

Sure, this is my config.json:

{ "data_dir": "/home/tschmude/PycharmProjects/smart-sentencing/examples/token-classification/Data processing scripts/Data_Preprocessed", "labels": "./Data/labels.txt", "model_name_or_path": "bert-base-german-cased", "output_dir": "./Data/Models", "task_type": "NER", "max_seq_length": 180, "num_train_epochs": 6, "per_device_train_batch_size": 48, "learning_rate": 0.001, "seed": 1, "overwrite_cache": true, "fp16": true, "do_train": true, "do_predict": true, "do_eval": true }

[Stimmot]

Issue solved... it had to do with a dumb typo in the path, sorry for the confusion!

[LysandreJik]

No problem, glad you solved your issue!