Weird error regarding CUDA? tokenization? who knows
How to reproduce:
2и missed a call\nall Back
dis be the cursed message
("2и" is enough)
Error Log:
ERROR - tasks - error_logger - PID: 2798371 - TID: 140595448178496 - apply_tagger_to_index - index out of range in self - 2022-05-19 10:43:13,112
Traceback (most recent call last):
File "/opt/texta/texta-rest/toolkit/bert_tagger/tasks.py", line 263, in apply_tagger_to_index
for success, info in streaming_bulk(client=ec.es, actions=actions, refresh="wait_for", chunk_size=bulk_size, max_chunk_bytes=max_chunk_bytes, max_retries=3):
File "/opt/texta/.conda/envs/texta-rest-gpu/lib/python3.8/site-packages/elasticsearch/helpers/actions.py", line 319, in streaming_bulk
for bulk_data, bulk_actions in _chunk_actions(
File "/opt/texta/.conda/envs/texta-rest-gpu/lib/python3.8/site-packages/elasticsearch/helpers/actions.py", line 155, in _chunk_actions
for action, data in actions:
File "/opt/texta/texta-rest/toolkit/bert_tagger/tasks.py", line 213, in update_generator
result = tagger_object.apply_loaded_tagger(tagger, text, input_type="text", feedback=False)
File "/opt/texta/texta-rest/toolkit/bert_tagger/models.py", line 219, in apply_loaded_tagger
tagger_result = tagger.tag_text(tagger_input)
File "/opt/texta/.conda/envs/texta-rest-gpu/lib/python3.8/site-packages/texta_bert_tagger/tagger.py", line 895, in tag_text
outputs = self.model(b_input_ids, token_type_ids=None,
File "/opt/texta/.conda/envs/texta-rest-gpu/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/opt/texta/.conda/envs/texta-rest-gpu/lib/python3.8/site-packages/transformers/models/bert/modeling_bert.py", line 1524, in forward
outputs = self.bert(
File "/opt/texta/.conda/envs/texta-rest-gpu/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/opt/texta/.conda/envs/texta-rest-gpu/lib/python3.8/site-packages/transformers/models/bert/modeling_bert.py", line 983, in forward
embedding_output = self.embeddings(
File "/opt/texta/.conda/envs/texta-rest-gpu/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/opt/texta/.conda/envs/texta-rest-gpu/lib/python3.8/site-packages/transformers/models/bert/modeling_bert.py", line 215, in forward
inputs_embeds = self.word_embeddings(input_ids)
File "/opt/texta/.conda/envs/texta-rest-gpu/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/opt/texta/.conda/envs/texta-rest-gpu/lib/python3.8/site-packages/torch/nn/modules/sparse.py", line 124, in forward
return F.embedding(
File "/opt/texta/.conda/envs/texta-rest-gpu/lib/python3.8/site-packages/torch/nn/functional.py", line 1852, in embedding
return torch.embedding(weight, input, padding_idx, scale_grad_by_freq, sparse)
IndexError: index out of range in self