FastBERT

FastBERT copied to clipboard

pypi/fastbert/fastbert.py 一个参数名拼写错误

1

line 234: ```python self._self_distillation( sentences_train, batch_size, learning_rate, epochs_num, warmup, report_steps, model_saving_pathm, sentences_dev, labels_dev, dev_speed, verbose ) ``` `model_saving_pathm`应该是`model_saving_path`吧

liuyair

多标签分类是否可行？

7

请问如果我是多标签任务，在每个维度独立做二分类 1.可以直接用KL散度做蒸馏loss吗 2.可以用类别维度的熵表示不确定性吗

Lier007

pypi\examples\single_sentence_classification\test.py文件缺少labels变量定义

1

应该添加labels = labels = ['T', 'F']

baiziyuandyufei

How to calculate Bert FLOPs

4

Hi, I have a very rookie question. How can I calculate the FLOPs of BERT model? I tried to use thop, ``` macs, params = profile(model, inputs=(input, ), custom_ops={YourModule: count_your_model})...

ZLKong

Miss attention FLOPS?

4

Hi, I found in MultiHeadedAttention, thop only count the FLOPS of linear layer, missing the attention operation.

YeDeming

有没有批量预测的代码啊？

2

我好想没有找到批量预测的代码块，请问在哪里啊

yufengjiao

What is the Weibo dataset?

1

Would you clarify what the Weibo dataset (one of the benchmarked task in the paper) is or provide a copy in this repo?

ZeweiChu

Vocabulary file line 344 has bad format token

1

这344行是个空行，有啥问题吗

s1162276945

Source code

2

Could you can provide source code early, we want to try and follow your work, Thanks

xhyandwyy

作者大大您好，首先非常感谢您的开源精神和杰出的工作，我有一点疑惑向来咨询一下大大。就是我采用speed=0.5，batch_size=64，预训练加载的是我在其他数据集上已经重新委托过的BERT模型，但是运行fastbert后最终的结果准确度却下降了20%，如下图所示   我有些疑惑，按道理来说不应该出现这样的问题啊，还是说作者大大的模型只对readme中提到的那几个模型有用，在其他数据集上微调的BERT模型不可以使用本文的程序吗？请作者大大解答一下，非常感谢

MiloGame