[quant][core][gpu] Implemented max pooling 2D using cudnn #74673
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
[ghstack-poisoned]
Contributor
🔗 Helpful links
💊 CI failures summary and remediationsAs of commit d22eda6 (more details on the Dr. CI page): 💚 💚 Looks good so far! There are no failures yet. 💚 💚 This comment was automatically generated by Dr. CI (expand for details).Please report bugs/suggestions to the (internal) Dr. CI Users group. |
[ghstack-poisoned]
dzdang
added
release notes: quantization
Summary: Quantized 2D max pooling was implemented using cudnn. A corresponding test case was also added. The operator requires cudnn version 8.3.3 or higher. v7 APIs (https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnPoolingForward) are currently used as there are currently no v8 APIs for pooling. Note the current implementation does not support dilated pooling as it is not supported in cudnn. Test plan: In pytorch main dir, execute ``` python test/test_quantization.py TestQuantizedOps.test_max_pool2d_cudnn ``` [ghstack-poisoned]
Contributor
Author
|
@dzdang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Summary: Quantized 2D max pooling was implemented using cudnn. A corresponding test case was also added. The operator requires cudnn version 8.3.3 or higher. v7 APIs (https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnPoolingForward) are currently used as there are currently no v8 APIs for pooling. Note the current implementation does not support dilated pooling as it is not supported in cudnn. Test plan: In pytorch main dir, execute ``` python test/test_quantization.py TestQuantizedOps.test_max_pool2d_cudnn ``` Differential Revision: [D35135200](https://our.internmc.facebook.com/intern/diff/D35135200) [ghstack-poisoned]
dzdang
added a commit
that referenced
this pull request
Mar 25, 2022
Summary: Quantized 2D max pooling was implemented using cudnn. A corresponding test case was also added. The operator requires cudnn version 8.3.3 or higher. v7 APIs (https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnPoolingForward) are currently used as there are currently no v8 APIs for pooling. Note the current implementation does not support dilated pooling as it is not supported in cudnn. Test plan: In pytorch main dir, execute ``` python test/test_quantization.py TestQuantizedOps.test_max_pool2d_cudnn ``` ghstack-source-id: ea8ef66 Pull Request resolved: #74673
Contributor
Author
|
@dzdang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Summary: Quantized 2D max pooling was implemented using cudnn. A corresponding test case was also added. The operator requires cudnn version 8.3.3 or higher. v7 APIs (https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnPoolingForward) are currently used as there are currently no v8 APIs for pooling. Note the current implementation does not support dilated pooling as it is not supported in cudnn. Test plan: In pytorch main dir, execute ``` python test/test_quantization.py TestQuantizedOps.test_max_pool2d_cudnn ``` Differential Revision: [D35135200](https://our.internmc.facebook.com/intern/diff/D35135200) [ghstack-poisoned]
Contributor
Author
|
@dzdang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Summary: Quantized 2D max pooling was implemented using cudnn. A corresponding test case was also added. The operator requires cudnn version 8.3.3 or higher. v7 APIs (https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnPoolingForward) are currently used as there are currently no v8 APIs for pooling. Note the current implementation does not support dilated pooling as it is not supported in cudnn. Test plan: In pytorch main dir, execute ``` python test/test_quantization.py TestQuantizedOps.test_max_pool2d_cudnn ``` Differential Revision: [D35135200](https://our.internmc.facebook.com/intern/diff/D35135200) [ghstack-poisoned]
dzdang
added a commit
that referenced
this pull request
Mar 25, 2022
Summary: Quantized 2D max pooling was implemented using cudnn. A corresponding test case was also added. The operator requires cudnn version 8.3.3 or higher. v7 APIs (https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnPoolingForward) are currently used as there are currently no v8 APIs for pooling. Note the current implementation does not support dilated pooling as it is not supported in cudnn. Test plan: In pytorch main dir, execute ``` python test/test_quantization.py TestQuantizedOps.test_max_pool2d_cudnn ``` ghstack-source-id: 4c9ae1a Pull Request resolved: #74673
Contributor
Author
|
@dzdang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Summary: Quantized 2D max pooling was implemented using cudnn. A corresponding test case was also added. The operator requires cudnn version 8.3.3 or higher. v7 APIs (https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnPoolingForward) are currently used as there are currently no v8 APIs for pooling. Note the current implementation does not support dilated pooling as it is not supported in cudnn. Test plan: In pytorch main dir, execute ``` python test/test_quantization.py TestQuantizedOps.test_max_pool2d_cudnn ``` Differential Revision: [D35135200](https://our.internmc.facebook.com/intern/diff/D35135200) [ghstack-poisoned]
dzdang
added a commit
that referenced
this pull request
Mar 25, 2022
Summary: Quantized 2D max pooling was implemented using cudnn. A corresponding test case was also added. The operator requires cudnn version 8.3.3 or higher. v7 APIs (https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnPoolingForward) are currently used as there are currently no v8 APIs for pooling. Note the current implementation does not support dilated pooling as it is not supported in cudnn. Test plan: In pytorch main dir, execute ``` python test/test_quantization.py TestQuantizedOps.test_max_pool2d_cudnn ``` ghstack-source-id: 1bcda3f Pull Request resolved: #74673
Contributor
Author
|
@dzdang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
Contributor
|
are we planning to register this under native_functions.yaml instead of a new quantized::maxpool function? |
Contributor
Author
I think that's what we're trending towards with the other max pool PRs. I'll make that change |
jerryzh168
reviewed
Apr 6, 2022
jerryzh168
reviewed
Apr 6, 2022
Summary: Quantized 2D max pooling was implemented using cudnn. A corresponding test case was also added. The operator requires cudnn version 8.3.3 or higher. v7 APIs (https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnPoolingForward) are currently used as there are currently no v8 APIs for pooling. Note the current implementation does not support dilated pooling as it is not supported in cudnn. Test plan: In pytorch main dir, execute ``` python test/test_quantization.py TestQuantizedOps.test_max_pool2d_cudnn ``` Differential Revision: [D35135200](https://our.internmc.facebook.com/intern/diff/D35135200) [ghstack-poisoned]
dzdang
added a commit
that referenced
this pull request
Apr 6, 2022
Summary: Quantized 2D max pooling was implemented using cudnn. A corresponding test case was also added. The operator requires cudnn version 8.3.3 or higher. v7 APIs (https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnPoolingForward) are currently used as there are currently no v8 APIs for pooling. Note the current implementation does not support dilated pooling as it is not supported in cudnn. Test plan: In pytorch main dir, execute ``` python test/test_quantization.py TestQuantizedOps.test_max_pool2d_cudnn ``` ghstack-source-id: d96f94a Pull Request resolved: #74673
Contributor
Author
|
@dzdang has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator. |
facebook-github-bot
pushed a commit
that referenced
this pull request
Apr 7, 2022
Summary: Pull Request resolved: #74673 Quantized 2D max pooling was implemented using cudnn. A corresponding test case was also added. The operator requires cudnn version 8.3.3 or higher. v7 APIs (https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnPoolingForward) are currently used as there are currently no v8 APIs for pooling. Note the current implementation does not support dilated pooling as it is not supported in cudnn. Test Plan: In pytorch main dir, execute ``` python test/test_quantization.py TestQuantizedOps.test_max_pool2d_cudnn ``` In pytorch main dir, execute ``` python test/test_quantization.py TestQuantizedOps.test_max_pool2d_cudnn ``` Differential Revision: D35135200 D35135200 Reviewed By: jerryzh168 Pulled By: dzdang fbshipit-source-id: 199991031e0419e13578a1d4abbe17dd2ed98f66
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.
Stack from ghstack (oldest at bottom):
Summary:
Quantized 2D max pooling was implemented using cudnn. A corresponding
test case was also added. The operator requires cudnn version 8.3.3
or higher. v7 APIs (https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnPoolingForward)
are currently used as there are currently no v8
APIs for pooling. Note the current implementation does not support dilated pooling
as it is not supported in cudnn.
Test plan:
In pytorch main dir, execute
Differential Revision: D35135200