Deep learning with CPU much lower accuracy than GPU?

11 次查看（过去 30 天）

Sabina Stefan 2020-1-20

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/501108-deep-learning-with-cpu-much-lower-accuracy-than-gpu

评论： Adam 2020-1-21

Hi everyone,

I've been wracking my brain for the past several days trying to understand why training my CNN on a CPU yields much lower accuracy than on a GPU. Then, I realized that implementing the MATLAB example (https://www.mathworks.com/help/vision/ug/semantic-segmentation-using-dilated-convolutions.html) with the MATLAB command also performs much worse when trained on a CPU vs GPU. What's going on here?

Here are some screenshots of the results on my system:

Screen Shot 2020-01-20 at 11.05.39 AM.png

Screen Shot 2020-01-20 at 11.05.21 AM.png

Thanks!

3 个评论
显示 1更早的评论隐藏 1更早的评论

Sabina Stefan 2020-1-20

Yes! I ran the example exactly as it is without changing any parameters (same for my own data).

Adam 2020-1-21

I just ran both on that example and got slightly higher accuracy with GPU than CPU, but given randomness not enough to conclude it is generally better, but certainly enough to conclude that CPU wasn't strictly better than GPU. You on;y posted one set of results above. which were they and what did the other look like?

请先登录，再进行评论。

请先登录，再回答此问题。