Optimizing the performance of fused_layer_norm and top_p_sampling operators #65711
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
|
你的PR提交成功，感谢你对开源项目的贡献! |
yuanlehome
changed the title
YanhuiDua
reviewed
Jul 4, 2024
| float random_ratio = exponential_transform(curand_uniform(&state), 1.0f); | ||
| float random_ratio = | ||
| exponential_transform(curand_uniform(states + bid), 1.0f); | ||
| #endif |
Contributor
There was a problem hiding this comment.
这里也可以用GPU(rand_uniform)包起来
YanhuiDua
reviewed
Jul 4, 2024
| for (int i = idx; i < bs; i += gridDim.x * blockDim.x) { | ||
| if (need_batch_random) { | ||
| #ifdef PADDLE_WITH_HIP | ||
| hiprand_init(seed, i, offset, &state[i]); |
carryyu
approved these changes
Jul 5, 2024
YanhuiDua
reviewed
Jul 31, 2024
| PD_THROW( | ||
| "the input data shape has error in the topp_beam_topk kernel."); | ||
| } | ||
| if (mode == "truncate") { |
Contributor
There was a problem hiding this comment.
需要对应修改search.py的python接口的默认mode 为“truncate”
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Outdated suggestions cannot be applied.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.
PR Category
Inference
PR Types
Performance
Description
pcard-71500
Optimizing the performance of fused_layer_norm and top_p_sampling operators.