运行 infer_cpu.py 报错 built-in method enable_padding of Tokenizer object at 0x00000219C9F57130

[zchengzhong]

请提出你的问题

运行PaddleNlp 下的uie模型目录中的 infer_cpu.py 报错 built-in method enable_padding of Tokenizer object at 0x00000219C9F57130

G:\soft\Anaconda3\envs\my_paddlenlp\python.exe G:/myself/future_pro/nlp/PaddleNLP/model_zoo/uie/deploy/python/infer_cpu.py [2022-10-14 22:48:14,265] [ INFO] - We are using <class 'paddlenlp.transformers.ernie.faster_tokenizer.ErnieFasterTokenizer'> to load 'ernie-3.0-base-zh'. [2022-10-14 22:48:14,267] [ INFO] - Already cached C:\Users\dataexa.paddlenlp\models\ernie-3.0-base-zh\ernie_3.0_base_zh_vocab.txt

[InferBackend] Creating Engine ... [Paddle2ONNX] Start to parse PaddlePaddle model... [Paddle2ONNX] Model file path: G:\myself\future_pro\nlp\PaddleNLP\model_zoo\uie\checkpoint\model_600\static\inference.pdmodel [Paddle2ONNX] Paramters file path: G:\myself\future_pro\nlp\PaddleNLP\model_zoo\uie\checkpoint\model_600\static\inference.pdiparams [Paddle2ONNX] Start to parsing Paddle model... [Paddle2ONNX] Use opset_version = 13 for ONNX export. [Paddle2ONNX] PaddlePaddle model is exported as ONNX format now. [InferBackend] Use CPU to inference ... TypeError: 'pad_token_type_id' is an invalid keyword argument for this function

The above exception was the direct cause of the following exception:

Traceback (most recent call last): File "G:/myself/future_pro/nlp/PaddleNLP/model_zoo/uie/deploy/python/infer_cpu.py", line 86, in main() File "G:/myself/future_pro/nlp/PaddleNLP/model_zoo/uie/deploy/python/infer_cpu.py", line 62, in main outputs = predictor.predict(texts) File "G:\myself\future_pro\nlp\PaddleNLP\model_zoo\uie\deploy\python\uie_predictor.py", line 457, in predict results = self._multi_stage_predict(input_data) File "G:\myself\future_pro\nlp\PaddleNLP\model_zoo\uie\deploy\python\uie_predictor.py", line 398, in _multi_stage_predict result_list = self._single_stage_predict(examples) File "G:\myself\future_pro\nlp\PaddleNLP\model_zoo\uie\deploy\python\uie_predictor.py", line 166, in _single_stage_predict return_dict=False) File "G:\myself\future_pro\nlp\PaddleNLP\paddlenlp\transformers\tokenizer_utils_base.py", line 2231, in call **kwargs) File "G:\myself\future_pro\nlp\PaddleNLP\paddlenlp\transformers\tokenizer_utils_base.py", line 2428, in batch_encode **kwargs, ) File "G:\myself\future_pro\nlp\PaddleNLP\paddlenlp\transformers\tokenizer_utils_faster.py", line 361, in _batch_encode_plus pad_to_multiple_of=pad_to_multiple_of, ) File "G:\myself\future_pro\nlp\PaddleNLP\paddlenlp\transformers\tokenizer_utils_faster.py", line 322, in set_truncation_and_padding self._tokenizer.enable_padding(**target) SystemError: <built-in method enable_padding of Tokenizer object at 0x00000219C9F57130> returned a result with an error set

[zchengzhong]

uie_predictor.py 中 下面这段修改成 self._tokenizer = AutoTokenizer.from_pretrained( "ernie-3.0-base-zh")

[wj-Mcat]

Glad to see that you have found the solution which not use faster tokenizer.

[github-actions[bot]]

This issue is stale because it has been open for 60 days with no activity. 当前issue 60天内无活动，被标记为stale。