[Bug]: Please help to solve the bug in loading TARSClassifier

[Anya-wUw]

Describe the bug

When I try to do TARSClassifier.load('tars-base') I catch RuntimeError

To Reproduce

from flair.models import TARSClassifier
classifier = TARSClassifier.load('tars-base')

Expected behavior

Import model without any errors

Logs and Stack traces

/usr/local/lib/python3.10/dist-packages/flair/models/tars_model.py in load(cls, model_path)
    926         from typing import cast
    927 
--> 928         return cast("TARSClassifier", super().load(model_path=model_path))

/usr/local/lib/python3.10/dist-packages/flair/models/tars_model.py in load(cls, model_path)
    321         from typing import cast
    322 
--> 323         return cast("FewshotClassifier", super().load(model_path=model_path))
    324 
    325 

/usr/local/lib/python3.10/dist-packages/flair/nn/model.py in load(cls, model_path)
    557         from typing import cast
    558 
--> 559         return cast("Classifier", super().load(model_path=model_path))
    560 
    561 

/usr/local/lib/python3.10/dist-packages/flair/nn/model.py in load(cls, model_path)
    189             if not isinstance(model_path, dict):
    190                 model_file = cls._fetch_model(str(model_path))
--> 191                 state = load_torch_state(model_file)
    192             else:
    193                 state = model_path

/usr/local/lib/python3.10/dist-packages/flair/file_utils.py in load_torch_state(model_file)
    357         # see https://github.com/zalandoresearch/flair/issues/351
    358         f = load_big_file(model_file)
--> 359         return torch.load(f, map_location="cpu")

/usr/local/lib/python3.10/dist-packages/torch/serialization.py in load(f, map_location, pickle_module, weights_only, mmap, **pickle_load_args)
   1012                     except RuntimeError as e:
   1013                         raise pickle.UnpicklingError(UNSAFE_MESSAGE + str(e)) from None
-> 1014                 return _load(opened_zipfile,
   1015                              map_location,
   1016                              pickle_module,

/usr/local/lib/python3.10/dist-packages/torch/serialization.py in _load(zip_file, map_location, pickle_module, pickle_file, overall_storage, **pickle_load_args)
   1420     unpickler = UnpicklerWrapper(data_file, **pickle_load_args)
   1421     unpickler.persistent_load = persistent_load
-> 1422     result = unpickler.load()
   1423 
   1424     torch._utils._validate_loaded_sparse_tensors()

/usr/local/lib/python3.10/dist-packages/flair/embeddings/transformer.py in __setstate__(self, state)
   1167 
   1168         if model_state_dict:
-> 1169             self.model.load_state_dict(model_state_dict)
   1170 
   1171     @classmethod

/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py in load_state_dict(self, state_dict, strict, assign)
   2150 
   2151         if len(error_msgs) > 0:
-> 2152             raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
   2153                                self.__class__.__name__, "\n\t".join(error_msgs)))
   2154         return _IncompatibleKeys(missing_keys, unexpected_keys)

RuntimeError: Error(s) in loading state_dict for BertModel:
	Unexpected key(s) in state_dict: "embeddings.position_ids".

Screenshots

Additional Context

No response

Environment

Versions:

Flair

0.12.2

Pytorch

2.1.0+cu118

Transformers

4.34.1

GPU

False

[None-Such]

@Anya-wUw ,

Looks like you are getting the same error I did.

I just closed my ticket (#3305), but it explains the assorted fixes for that Tutorial thanks to @helpmefindaname .

[Anya-wUw]

@None-Such Thank you so much! You saved my last nerve cells!!

How to solve:
!pip install transformers==4.29.2
!pip install huggingface_hub==0.14.1