lemonhu / NER-BERT-pytorch

PyTorch solution of named entity recognition task Using Google AI's pre-trained BERT model.
MIT License
436 stars 110 forks source link

AttributeError: 'NoneType' object has no attribute 'tokenize' #26

Open wss-xm opened 4 years ago

wss-xm commented 4 years ago

tokens = self.tokenizer.tokenize(line.strip())

AttributeError: 'NoneType' object has no attribute 'tokenize'

tongchangD commented 4 years ago

How did you solve the problem

wss-xm commented 4 years ago

Sorry, I don't solve the problem

At 2020-08-04 14:05:48, "tongchangD" notifications@github.com wrote:

How did you solve the problem

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

tongchangD commented 4 years ago

Sorry, I don't solve the problem At 2020-08-04 14:05:48, "tongchangD" notifications@github.com wrote: How did you solve the problem — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

i change the tokenize,you can see in from pytorch_pretrained_bert import BertTokenizer tokenizer = BertTokenizer.from_pretrained('./dataset/bert-base-chinese')

yolanda182 commented 3 years ago

Sorry, I don't solve the problem At 2020-08-04 14:05:48, "tongchangD" notifications@github.com wrote: How did you solve the problem — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

i change the tokenize,you can see in from pytorch_pretrained_bert import BertTokenizer tokenizer = BertTokenizer.from_pretrained('./dataset/bert-base-chinese')

Could you please tell me where you change it?

tongchangD commented 3 years ago

Sorry, I don't solve the problem At 2020-08-04 14:05:48, "tongchangD" notifications@github.com wrote: How did you solve the problem — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

i change the tokenize,you can see in from pytorch_pretrained_bert import BertTokenizer tokenizer = BertTokenizer.from_pretrained('./dataset/bert-base-chinese')

Could you please tell me where you change it?

I modified it on the company computer