对Pytorch 添加多卡训练支持 #55
Open
+1,171
−7
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…orward().shape(0) is batch_size otherwise it's difficult to change code for multipgpu
…e recognizing uppercase letter to uppercase + lowercase
在forward() 里面，LSTM之前的shape跟作者保持一致，不然训练不出来。
|
I'm a little late to the party, but I noticed you all aren't using a notebook review tool and wanted to invite you to review this pull request with GitNotebooks: https://gitnotebooks.com/ypwhs/captcha_break/pull/55 It lets you do things like comment on rendered markdown and code cells, so might be an easy win for your PR reviews. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
修改说明
修改1
module.forward() 返回值维度第一个最好是batch
也就是说
在计算ctc_loss 的时候再改回来。
不然多卡训练会失败，如果要使用DataParallel就必须这样做
修改2. 添加DataDistributed支持
文件ctc_pytorch_multigpu.py。 在这里我修改了