Mu Li

i made a patch in a hurry, will do more tests tomorrow. On Fri, Jun 19, 2015 at 1:17 AM, Zhongwen Xu [email protected] wrote: > @mli https://github.com/mli The bug still...

updated README.md for how to run

@kevinthesun

Hi Lisen, I'm working on the dev branch, which provides better fault tolerance, changes from std::shared_ptr to Message\* in API, more google c++ style and better doxygen documents. i hope...

the second one also makes sense. but you can think the first one is special case of the second one, the former uses a fullyc on both `user` and `pred`...

@piiswrong can you review this notebook. preview at https://github.com/karishmamalkan/mxnet-notebooks/blob/rnn_api_tutorial/python/rnn_api_tutorial/RNN%20API%20tutorial.ipynb

Thanks for pointing it out. That's a missing feature.. We need to put the files in the demo/ folder into the pypi package. But I don't have a plan to...

看上去不错。我没有windows，测试不了。但看包里面似乎没有pytorch？ 而且，包里面有whisper模型吗？ 此外，用docker打包是不是可移植性更高些？

有点怀疑没有pytorch和tensorflow，是怎么做预测的。不知道没有别的同学也能试一下？

> 简单的方式可以用标点符号断, 设一个阈值, duration用字数比例做系数就行了, 我之前做项目尝试过, 对中文来说效果还行. vad是有这种小缺陷. 有兴趣的话欢迎PR这个feature