[Speed]implement cudnn sequence softmax cudnn #8978
Merged
Conversation
This file contains hidden or 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
dzhwinter
changed the title
Member
|
Do we have a benchmark data using this new operator? |
QiJune
reviewed
Mar 12, 2018
| ctx.template device_context<platform::CUDADeviceContext>(); | ||
| use_cudnn &= dev_ctx.cudnn_handle() != nullptr; | ||
| } | ||
| #endif |
Member
There was a problem hiding this comment.
The logic of getting use_cudnn here is a little complex. Could you refine it and make it more readable?
Contributor
Author
|
@QiJune @jacquesqiao has updated. |
QiJune
reviewed
Mar 13, 2018
| using Tensor = framework::Tensor; | ||
|
|
||
| template <typename DeviceContext, typename T> | ||
| class SoftmaxCUDNNKernel : public framework::OpKernel<T> { |
Member
There was a problem hiding this comment.
Since CUDNNKernel is only supported on GPU,
template <typename T>
class SoftmaxCUDNNKernel : public framework::OpKernel<T> {
....
context.cuda_device_context()
...
}
| library_ = framework::LibraryType::kCUDNN; | ||
| } else { | ||
| library_ = framework::LibraryType::kPlain; | ||
| } |
Member
There was a problem hiding this comment.
framework::LibraryType library_ = framework::LibraryType::kPlain;
if (use_cudnn && runtime_cudnn_support) {
library_ = framework::LibraryType::kCUDNN;
}
| using Tensor = framework::Tensor; | ||
| using LoDTensor = framework::LoDTensor; | ||
|
|
||
| template <typename DeviceContext, typename T> |
Member
There was a problem hiding this comment.
No need to take DeviceContext as a template parameter.
QiJune
reviewed
Mar 13, 2018
| library_ = framework::LibraryType::kCUDNN; | ||
| } else { | ||
| library_ = framework::LibraryType::kPlain; | ||
| } |
Member
There was a problem hiding this comment.
framework::LibraryType library_ = framework::LibraryType::kPlain;
if (use_cudnn && runtime_cudnn_support) {
library_ = framework::LibraryType::kCUDNN;
}
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
fix #8594.
This PR contains four operators. sequence_softmax, sequence_softmax_grad, softmax
, softmax_grad.
Take sequence_softmax op as example, compare with the previous implement, the sequence_softmax operator time cost is lower than mul operator now.
The time cost from 1.94211 -> 0.981581 in every minibach.
before optimize
after optimize