Getting Error while calling BERTClassifier

[Gaurav7004]

import transformers
import aspect_based_sentiment_analysis as absa

name = 'absa/classifier-rest-0.2'
model = absa.BertABSClassifier.from_pretrained(name)
tokenizer = absa.BertTokenizer.from_pretrained(name)

---

ValueError Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_40004/473772509.py in
5
6 name = 'absa/classifier-rest-0.2'
----> 7 model = absa.BertABSClassifier.from_pretrained(name)
8
9 tokenizer = BertTokenizer.from_pretrained(model)

~\Anaconda3\lib\site-packages\transformers\modeling_tf_utils.py in from_pretrained(cls, pretrained_model_name_or_path, *model_args, **kwargs)
1593 model(model.dummy_inputs) # build the network with dummy inputs
1594 else:
-> 1595 model(model.dummy_inputs) # build the network with dummy inputs
1596
1597 assert os.path.isfile(resolved_archive_file), f"Error retrieving file {resolved_archive_file}"

~\Anaconda3\lib\site-packages\keras\utils\traceback_utils.py in error_handler(*args, **kwargs)
65 except Exception as e: # pylint: disable=broad-except
66 filtered_tb = _process_traceback_frames(e.traceback)
---> 67 raise e.with_traceback(filtered_tb) from None
68 finally:
69 del filtered_tb

~\Anaconda3\lib\site-packages\aspect_based_sentiment_analysis\models.py in call(self, token_ids, attention_mask, token_type_ids, training, **bert_kwargs)
139 **bert_kwargs
140 ) -> Tuple[tf.Tensor, Tuple[tf.Tensor, ...], Tuple[tf.Tensor, ...]]:
--> 141 outputs = self.bert(
142 inputs=token_ids,
143 attention_mask=attention_mask,

ValueError: Exception encountered when calling layer "bert_abs_classifier_10" (type BertABSClassifier).

The first argument to Layer.call must always be passed.

Call arguments received:
• token_ids={'input_ids': 'tf.Tensor(shape=(3, 5), dtype=int32)'}
• attention_mask=None
• token_type_ids=None
• training=False
• bert_kwargs=<class 'inspect._empty'>

[caballeto]

Getting same error.

[liorc-git]

I have the same problem with python 3.9 (worked well with 3.8).
Also - can anyone shed some light about the difference between the models absa/bert-rest-0.2 and absa/classifiers-rest-0.2 ? As in some of my attempts I noticed that the first managed to load and the second didn't.
In https://huggingface.co/absa it seems they were all updated at the same date.

[pmenonlek]

Getting the same error when working in python 3.9

[linayc]

Facing this issue as well..