[Quant] Add fused LinearTanh module for onednn backend #88923
Closed
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]
Xia-Weiwen
requested review from
albanD,
jbschlosser,
jerryzh168 and
z-a-f
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/88923
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit ab6b724: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Nov 12, 2022
jerryzh168
reviewed
Nov 12, 2022
Contributor
jerryzh168
left a comment
jerryzh168
left a comment
There was a problem hiding this comment.
looks good overall, but probably need some update after the previous PRs are updated I think
jgong5
approved these changes
Nov 14, 2022
Collaborator
jgong5
left a comment
jgong5
left a comment
There was a problem hiding this comment.
Not sure if we need to rename the files to something like linear_activation from linear_relu since we added linear+tanh into it.
Collaborator
Author
Good idea. Thanks. I will do that after clarifications from Jerry about PT 2.0. |
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule [ghstack-poisoned]
Collaborator
Author
|
Hi @jerryzh168 I have done changes per your comments. Could you take a look again? Thanks! |
Collaborator
Author
|
Hi @jerryzh168. Is it ok to land this? Thanks |
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
Collaborator
Author
|
Hi @jerryzh168 Do you have more comments on this PR? Thanks! |
jerryzh168
reviewed
Dec 14, 2022
torch/nn/intrinsic/__init__.py
Outdated
Comment on lines
14
to
15
| from torch.ao.nn.intrinsic import LinearLeakyReLU | ||
| from torch.ao.nn.intrinsic import LinearTanh |
Contributor
There was a problem hiding this comment.
please revert these, we don't want to add new things to torch/nn/intrinsic folder, this is a folder that we plan to deprecate in the future
Collaborator
Author
There was a problem hiding this comment.
Ok they are reverted.
jerryzh168
requested changes
Dec 14, 2022
Contributor
jerryzh168
left a comment
jerryzh168
left a comment
There was a problem hiding this comment.
please revert changes to torch/nn/intrinsic folder
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
Collaborator
Author
Ok they are reverted. |
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
jerryzh168
reviewed
Dec 15, 2022
test/allowlist_for_publicAPI.json
Outdated
| "torch.nn.intrinsic.quantized.modules.bn_relu": "torch.ao.nn.intrinsic.quantized.modules.bn_relu", | ||
| "torch.nn.intrinsic.quantized.modules.conv_relu": "torch.ao.nn.intrinsic.quantized.modules.conv_relu", | ||
| "torch.nn.intrinsic.quantized.modules.linear_relu": "torch.ao.nn.intrinsic.quantized.modules.linear_relu", | ||
| "torch.nn.intrinsic.quantized.modules.linear_activation": "torch.ao.nn.intrinsic.quantized.modules.linear_activation", |
Contributor
There was a problem hiding this comment.
is this change still valid? we don't have torch.nn.intrinsic.quantized.modules.linear_activation now
Collaborator
Author
There was a problem hiding this comment.
It's still needed whereas the file name is changed:
"torch.nn.intrinsic.quantized.modules.linear_relu": "torch.ao.nn.intrinsic.quantized.modules.linear_activation",
Collaborator
Author
There was a problem hiding this comment.
I reverted the file name change due to CI checks failure.
jerryzh168
approved these changes
Dec 15, 2022
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
Collaborator
Author
I reverted the change due to CI check failure. |
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
**Summary** This PR adds fused `QLinearTanh` module for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown. **Test plan** python test_quantization.py TestStaticQuantizedModule cc jgong5 mingfeima XiaobingSuper sanchitintel ashokei jingxu10 [ghstack-poisoned]
Collaborator
Author
|
@pytorchbot merge |
Collaborator
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
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
This PR adds fused
QLinearTanhmodule for onednn backend, which will be used for int8 inference with onednn backend. Cannot call this module with other quantization backends otherwise an error is thrown.Test plan
python test_quantization.py TestStaticQuantizedModule
cc @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10