Godlike_Yue

您好，我想在pytorch_version版本基础上改用Robert-tiny或者Albert-tiny之类的小型bert模型进行实验，请问改动大吗？

jieba_fast好像存在内存泄漏

2

windows环境，jiebe_fast == 0.53 本地用tracemalloc输出内存泄漏代码 

ImportError: cannot import name 'AutoModel' from 'transformers' (unknown location)

4

pip install ltp==4.1.5.post2 win10 引用ltp 报错 

transformers中使用clue/roberta_chinese_pair_tiny的疑问

1

我通过transformers使用roberta_chinese_pair_tiny，提示以下warning 1. You are using a model of type roberta to instantiate a model of type bert. This is not supported for all configurations of models and can yield errors....

亲测可用，snli数据集上可复现实验结果

3

在snli数据集上可以复现实验结果，ACC可以达到89.2 {'acc': 0.8926099348534202, 'loss': 0.3331541208659901, 'score': 0.8926099348534202, 'updates': 136400}

what is the verison of transformers ?

1

Please provide the version of transformers ?