[OPENCL] optimize pooling with fine-grained #553
Merged
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
lnmdlong
force-pushed
the
feature_opencl_pooling_opt
branch
from
856284b
to
cbc364e
Compare
Codecov Report
@@ Coverage Diff @@
## master #553 +/- ##
=======================================
Coverage 24.56% 24.56%
=======================================
Files 287 287
Lines 9047 9047
=======================================
Hits 2222 2222
Misses 6825 6825 Continue to review full report at Codecov.
|
neiltian-tencent
approved these changes
Nov 14, 2020
neiltian-tencent
pushed a commit
that referenced
this pull request
Nov 14, 2020
seanxcwang
added a commit
that referenced
this pull request
Nov 14, 2020
* [OPTIMIZER][OPT] fuse conv sigmoid mul * [ARM][OPT] 1.conv add swish post act 2.modify deconv condition 3.opt sigmoid 4.unfold binary op * [OPENCL][OPT] support fuse sigmoid mul * [Metal] support sigmoid_mul activation [Test] test different activation types in conv and deconv unit test * [OPTIMIZE][OPT] refactor for fuse conv sigmoid mul * [ARM][OPT] fix arm conv1x1 swish fuse error * [OPENCL] optimize pooling with fine-grained parallelsim (#553) * [iOS] fix tnn xcodeproj Co-authored-by: seanxcwang <seanxcwang@tencent.com> Co-authored-by: devandong <devandong@tencent.com> Co-authored-by: lnmdlong <lnmdlong@hotmail.com>
darrenyao87
added a commit
that referenced
this pull request
Nov 16, 2020
lnmdlong
pushed a commit
that referenced
this pull request
Sep 2, 2021
* master: (55 commits) [OPT] optimize memory useage of instance [ATLAS][OPT] reduce model string copy fix spell error (#564) [OPENCL][FIX] fix share opencl context (#563) add missing parameter in CreateContext (#551) [ADD]support ShareCommandQueue between instances (#556) Hotfix opencl select (#555) Eff opt (#554) [OPENCL] optimize pooling with fine-grained parallelsim (#553) create issue templates (#549) [OPENCL] fix fp16 select with short condition (#544) [Metal] enable CPU N8UC4 Mat in metal ConvertFromMat (#543) Add fuse SpaceToDepth and DepthToSpace (#542) [TOOL][ADD] add output name param support (#537) [ATLAS][BUG] allocate context for each weight memory avoid device mismatch [ATLAS][BUG] fix weights buffer free error [CONVERTER][BUG] fix the bug of fuse conv (#529) [ATLAS][BUG] fix compile error due to api change [UPD] update iOS&macos building scripts, check building errors (#531) [Metal] fix reshape out-of-bound access bug (#496) ...
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.
使用local memory优化OpenCL Pooing
速度测试结果：https://docs.qq.com/sheet/DVXFndE5uek10bkVx