bert_chinese_pytorch icon indicating copy to clipboard operation
bert_chinese_pytorch copied to clipboard

bert for chinese text classification

Results 8 bert_chinese_pytorch issues
Sort by recently updated
recently updated
newest added

我使用example转feature的时候加载时间很长,经常卡在那里不动,请问你有遇到过吗

Now, for my dataset, Error:'model' not found, How to solve it? modify: ![image](https://user-images.githubusercontent.com/38399851/67007394-dc0fa500-f119-11e9-88f2-64023e9041da.png)

您好,请问‘’Weights of BertForSequenceClassification not initialized from pretrained model: ['classifier.weight', 'classifier.bias']‘’这种情况您碰到过吗,这样权重无法加载,预训练模型的意义也发挥不出来。

直接把do_train参数取消,提示: Output directory (path) already exists and is not empty.

Traceback (most recent call last): File "bert.py", line 651, in main() File "bert.py", line 541, in main cache_dir=PYTORCH_PRETRAINED_BERT_CACHE / 'distributed_{}'.format(args.local_rank)) File "/home/lindseywang/.local/lib/python3.6/site-packages/pytorch_pretrained_bert/modeling.py", line 581, in from_pretrained model = cls(config, *inputs,...

抱歉打扰您。我在运行过程中又遇到了模型出错问题。 这个问题我在pytor_pretrained_bert的issue里看到过,但作者说2018年11月就已经解决了这个问题,我看我现在运行的代码中也包含了修正后的部分。不知道我这现在是什么情况?

.json的文件怎么生成的啊。。。我现在有每行的格式是:label\t文本的.txt文件,不知道怎么变成代码里可以用的json文件。。。 求教

应该在哪添加啊,我看from_pretrained代码里没找到