Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

win系统+RTX2060能顺利训练模型吗 #16

Open
lijielijie opened this issue May 21, 2022 · 9 comments
Open

win系统+RTX2060能顺利训练模型吗 #16

lijielijie opened this issue May 21, 2022 · 9 comments

Comments

@lijielijie
Copy link

lijielijie commented May 21, 2022 •
edited
Loading

使用这个项目，能在win10+RTX2060的环境下训练fastspeech2等模型吗
@lijielijie lijielijie changed the title win系统+NVIDIA RTX2060能顺利训练模型吗 win系统+RTX2060能顺利训练模型吗 May 21, 2022
@jerryuhoo
Copy link
Owner

没试过不太清楚，但是训练速度肯定会比较慢

@lijielijie
Copy link
Author

我执行2.4. 使用ASR获得文本时，提示我 No module named 'paddlespeech'，我直接用pip install paddlespeech命令安装就可以吧？

@jerryuhoo
Copy link
Owner

是的

@lijielijie
Copy link
Author

萌新再问一个入门的问题，单人跟多人具体是啥区别，我是从B站找了一个主播的视频，拿来训练，这是算是单人数据集吗

@jerryuhoo
Copy link
Owner

数据集过小的话建议多人，比如几百句。数据集足够的话可以单人训练，比如有一万句。你这个算是单人训练。

@lijielijie
Copy link
Author

这里好像少了1个参数：--speaker-dict=dump/speaker_id_map.txt
image

@lijielijie
Copy link
Author

lijielijie commented May 24, 2022 •
edited by jerryuhoo
Loading

然后再问一个训练的问题，这五张图主要看哪个指标的loss，这个loss低到多少才算训练初成
image

@jerryuhoo
Copy link
Owner

这里好像少了1个参数：--speaker-dict=dump/speaker_id_map.txt image

现在好像是单人和多人都需要speaker-dict，那么应该不需要分开了，我之后改一下。

@jerryuhoo
Copy link
Owner

然后再问一个训练的问题，这五张图主要看哪个指标的loss，这个loss低到多少才算训练初成 image

主要看eval/loss/你这个训练好像有点问题，检查一下数据吧

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@lijielijie @jerryuhoo