'torch._C.PyTorchFileReader' object has no attribute'seek'

Hello,

I am using the following model for sentence similarity

https://huggingface.co/sentence-transformers/stsb-xlm-r-multilingual/tree/main

word_embedding_model = models.Transformer(bert_model_dir)  # , max_seq_length=512
pooling_model = models.Pooling(word_embedding_model.get_word_embedding_dimension())
model = SentenceTransformer(modules=[word_embedding_model, pooling_model], device=device_str)

But, I get this error:

Traceback (most recent call last):

  File "/home/work/anaconda/lib/python3.6/site-packages/torch/serialization.py", line 306, in _check_seekable

    f.seek(f.tell())

AttributeError:'torch._C.PyTorchFileReader' object has no attribute'seek'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):

  File "/home/work/anaconda/lib/python3.6/site-packages/transformers/modeling_utils.py", line 1205, in from_pretrained

    state_dict = torch.load(resolved_archive_file, map_location="cpu")

  File "/home/work/anaconda/lib/python3.6/site-packages/torch/serialization.py", line 584, in load

    return _load(opened_zipfile, map_location, pickle_module, **pickle_load_args)

  File "/home/work/anaconda/lib/python3.6/site-packages/moxing/framework/file/file_io_patch.py", line 200, in _load

    _check_seekable(f)

  File "/home/work/anaconda/lib/python3.6/site-packages/torch/serialization.py", line 309, in _check_seekable

    raise_err_msg(["seek", "tell"], e)

  File "/home/work/anaconda/lib/python3.6/site-packages/torch/serialization.py", line 302, in raise_err_msg

    raise type(e)(msg)

AttributeError:'torch._C.PyTorchFileReader' object has no attribute'seek'. You can only torch.load from a file that is seekable. Please pre-load the data into a buffer like io.BytesIO and try to load from it instead .

During handling of the above exception, another exception occurred:

Traceback (most recent call last):

  File "code/similarity.py", line 118, in <module>

    word_embedding_model = models.Transformer(bert_model_dir) #, max_seq_length=512

  File "/home/work/anaconda/lib/python3.6/site-packages/sentence_transformers/models/Transformer.py", line 30, in __init__

    self.auto_model = AutoModel.from_pretrained(model_name_or_path, config=config, cache_dir=cache_dir)

  File "/home/work/anaconda/lib/python3.6/site-packages/transformers/models/auto/auto_factory.py", line 381, in from_pretrained

    return model_class.from_pretrained(pretrained_model_name_or_path, *model_args, config=config, **kwargs)

  File "/home/work/anaconda/lib/python3.6/site-packages/transformers/modeling_utils.py", line 1208, in from_pretrained

    f"Unable to load weights from pytorch checkpoint file for'{pretrained_model_name_or_path}' "

OSError: Unable to load weights from pytorch checkpoint file for'/home/work/user-job-dir/input/pretrained_models/stsb-xlm-r-multilingual/' at'/home/work/user-job-dir/input /pretrained_models/stsb-xlm-r-multilingual/pytorch_model.bin'If you tried to load a PyTorch model from a TF 2.0 checkpoint, please set from_tf=True. 

I checked on web but could not find any solution. What could be the problem? Thank you.

[nreimers]

Ah ok, that is the problem.

The torch load function requires that the file system supports seek (e.g. see https://www.tutorialspoint.com/python/file_seek.htm)

Apparently, your company cloud file system does not support this elementary file system operation. Hence, torch.load cannot load any file.

Real solution would be to get a better company cloud with a file system that support basic I/O operations.

In PyTorch 1.6, they changed the file format when models are saved. Maybe on your company cloud the old file format works? You can try it with this model: https://public.ukp.informatik.tu-darmstadt.de/reimers/sentence-transformers/v0.2/bert-base-nli-cls-token.zip

It has still the pre PyTorch 1.6. file format.

[deadsoul44]

torch.save(model, 'pytorch_model.bin', _use_new_zipfile_serialization=False)

This is working. Thank you.