Skip to content

[quant] Add set_quantizer_ for QTensor #21852

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
jerryzh168 wants to merge 4 commits into from
Closed

[quant] Add set_quantizer_ for QTensor #21852

jerryzh168 wants to merge 4 commits into from

Conversation

@jerryzh168
Copy link
Contributor

@jerryzh168 jerryzh168 commented Jun 17, 2019 •
edited
Loading

Stack:
    :black_circle:  #21852 [quant] Add set_quantizer_ for QTensor  💛

To enable change of q_scale and q_zero_point in copy_

Differential Revision: D15793427

@jerryzh168
V3: Initial commit
a033b86 
Differential Revision: D15793427
Differential Version: 85027694
@pytorchbot pytorchbot added oncall: jit Add this issue/PR to JIT oncall triage queue module: internals Related to internal abstractions in c10 and ATen module: operators module: pybind Related to our Python bindings / interactions with other Python libraries oncall: quantization Quantization support in PyTorch labels Jun 17, 2019
@jerryzh168
V4: (no description)
1719069 
Differential Revision: D15793427
Differential Version: 85027709
@jerryzh168 jerryzh168 requested a review from gchanan June 17, 2019 15:35
@jerryzh168 jerryzh168 requested review from li-roy and suo June 17, 2019 15:36
dzhulgakov
dzhulgakov reviewed Jun 17, 2019
View reviewed changes
@jerryzh168
V5: comments for ConstQuantizerPtr
830bef1 
Differential Revision: D15793427
Differential Version: 85133145
@jerryzh168
V6: (no description)
7af0f55 
Differential Revision: D15793427
Differential Version: 85139902
@facebook-github-bot
Copy link
Contributor

facebook-github-bot commented Jun 19, 2019

This pull request has been merged in fa5263a.

zdevito pushed a commit to zdevito/ATen that referenced this pull request Jun 19, 2019
@jerryzh168 @facebook-github-bot
Add set_quantizer_ for QTensor (#21852)
0740add 
Summary:
Pull Request resolved: pytorch/pytorch#21852

To enable change of q_scale and q_zero_point in `copy_`

Differential Revision: D15793427

fbshipit-source-id: a7040b5b956d161fd6af6176287f4a4aa877c9be
@ailzhang ailzhang mentioned this pull request Jun 19, 2019
Merged
@ezyang
Copy link
Contributor

ezyang commented Jun 19, 2019

cc @ailzhang

This appears to have broken XLA build.

Jun 19 04:07:20 clang-7 -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/var/lib/jenkins/workspace/xla -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-genfiles -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/protobuf_archive/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/eigen_archive -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_absl -I/var/lib/jenkins/workspace -I/var/lib/jenkins/workspace/torch/csrc -I/var/lib/jenkins/workspace/torch/lib/tmp_install/include -I/opt/conda/lib/python3.6/site-packages/torch/include -I/opt/conda/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -I/opt/conda/lib/python3.6/site-packages/torch/include/TH -I/opt/conda/lib/python3.6/site-packages/torch/include/THC -I/opt/conda/include/python3.6m -c torch_xla/csrc/pooling.cpp -o build/temp.linux-x86_64-3.6/torch_xla/csrc/pooling.o -std=c++14 -Wno-sign-compare -Wno-deprecated-declarations -Wno-return-type -Wno-macro-redefined -Wno-return-std-move -O0 -g -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=_XLAC -D_GLIBCXX_USE_CXX11_ABI=0
Jun 19 04:07:25 clang-7 -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/var/lib/jenkins/workspace/xla -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-genfiles -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/protobuf_archive/src -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/eigen_archive -I/var/lib/jenkins/workspace/xla/third_party/tensorflow/bazel-tensorflow/external/com_google_absl -I/var/lib/jenkins/workspace -I/var/lib/jenkins/workspace/torch/csrc -I/var/lib/jenkins/workspace/torch/lib/tmp_install/include -I/opt/conda/lib/python3.6/site-packages/torch/include -I/opt/conda/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -I/opt/conda/lib/python3.6/site-packages/torch/include/TH -I/opt/conda/lib/python3.6/site-packages/torch/include/THC -I/opt/conda/include/python3.6m -c torch_xla/csrc/tensor_impl.cpp -o build/temp.linux-x86_64-3.6/torch_xla/csrc/tensor_impl.o -std=c++14 -Wno-sign-compare -Wno-deprecated-declarations -Wno-return-type -Wno-macro-redefined -Wno-return-std-move -O0 -g -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=_XLAC -D_GLIBCXX_USE_CXX11_ABI=0
Jun 19 04:07:25 torch_xla/csrc/aten_xla_type_base.cpp:7398:65: error: unknown type name 'ConstQuantizerPtr'; did you mean 'at::ConstQuantizerPtr'?
Jun 19 04:07:25 at::Tensor & AtenXlaTypeBase::set_quantizer_(at::Tensor & self, ConstQuantizerPtr quantizer) const {
Jun 19 04:07:25                                                                 ^~~~~~~~~~~~~~~~~
Jun 19 04:07:25                                                                 at::ConstQuantizerPtr
Jun 19 04:07:25 /opt/conda/lib/python3.6/site-packages/torch/include/ATen/core/TensorMethods.h:21:7: note: 'at::ConstQuantizerPtr' declared here
Jun 19 04:07:25 using ConstQuantizerPtr = const c10::intrusive_ptr<Quantizer>&;
Jun 19 04:07:25       ^
Jun 19 04:07:28 2 errors generated.

https://circleci.com/gh/pytorch/pytorch/2028263?utm_campaign=vcs-integration-link&utm_medium=referral&utm_source=github-build-link/console

@suo
Copy link
Member

suo commented Jun 19, 2019

cc @ailzhang, can you advise on how to unbreak

@ailzhang
Copy link
Contributor

ailzhang commented Jun 19, 2019

@suo yea the breakage caused by this PR is already fixed. Currently test on master is also just fixed in pytorch/xla#761. So we should be good now.

@ezyang ezyang deleted the export-D15793427 branch July 19, 2019 15:52
@jerryzh168 jerryzh168 mentioned this pull request Nov 1, 2019
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@dzhulgakov dzhulgakov dzhulgakov approved these changes

@gchanan gchanan Awaiting requested review from gchanan

@suo suo Awaiting requested review from suo

@li-roy li-roy Awaiting requested review from li-roy

Assignees

No one assigned

Labels

Merged module: internals Related to internal abstractions in c10 and ATen module: pybind Related to our Python bindings / interactions with other Python libraries oncall: jit Add this issue/PR to JIT oncall triage queue oncall: quantization Quantization support in PyTorch

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

10 participants

@jerryzh168 @facebook-github-bot @ezyang @suo @ailzhang @dzhulgakov @li-roy @pytorchbot @mruberry