You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2020/04/11 07:01:08 UTC
[GitHub] [incubator-mxnet] zixuanweeei opened a new pull request #18028:
[MKLDNN] Support quantized rnn towards v1.6.x
zixuanweeei opened a new pull request #18028: [MKLDNN] Support quantized rnn towards v1.6.x
URL: https://github.com/apache/incubator-mxnet/pull/18028
## Description ##
**Mirror PR of #18001, towards v1.6.x branch**. In this PR, we add support of quantization flow of the rnn operator. Currently, only the LSTM mode supports INT8 inference.
## Checklist ##
### Essentials ###
Please feel free to remove inapplicable items for your PR.
- [x] Changes are complete (i.e. I finished coding on this PR)
- [x] All changes have test coverage:
- Unit tests are added for small changes to verify correctness (e.g. adding a new operator)
- Nightly tests are added for complicated/long-running ones (e.g. changing distributed kvstore)
- Build tests will be added for build configuration changes (e.g. adding a new build option with NCCL)
- [x] Code is well-documented:
- For user-facing API changes, API doc string has been updated.
- For new C++ functions in header files, their functionalities and arguments are documented.
- For new examples, README.md is added to explain the what the example does, the source of the dataset, expected performance on test set and reference to the original paper if applicable
- Check the API doc at https://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html
- [x] To the best of my knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change
### Changes ###
- [x] Add _contrib_quantized_rnn op.
- [x] Add asymmetric quantization - _contrib_quantized_asym op, to quantize FP32 data to U8 data using scale and shift.
- [x] Add MXNET_USE_WEIGHT_CACHE to control rnn init behavior.
- [x] Support data layout in NDArrayIter. Specifically, NDArrayIter supports only `NCHW` layout by default, and there is no way to support other layouts, like sequential `TNC` layout. This PR makes some changes to NDArrayIter to leverage the feature (assuming that N represents the batch).
- [x] Move MKLDNNRnnMemMgr to individual layer.
##Comments##
* Performance results will be added to #18001.
@ciyongch @TaoLv @pengzhao-intel
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
[GitHub] [incubator-mxnet] zixuanweeei commented on issue #18028: [MKLDNN]
Support quantized rnn towards v1.6.x
Posted by GitBox <gi...@apache.org>.
zixuanweeei commented on issue #18028: [MKLDNN] Support quantized rnn towards v1.6.x
URL: https://github.com/apache/incubator-mxnet/pull/18028#issuecomment-612350028
@mxnet-label-bot add [mkldnn]
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
[GitHub] [incubator-mxnet] zixuanweeei commented on issue #18028: [MKLDNN]
Support quantized rnn towards v1.6.x
Posted by GitBox <gi...@apache.org>.
zixuanweeei commented on issue #18028: [MKLDNN] Support quantized rnn towards v1.6.x
URL: https://github.com/apache/incubator-mxnet/pull/18028#issuecomment-612756454
> LGTM for the current implementation, but we still need another DNNL patch (or DNNL version upgrade) to mitigate the overhead of `_contrib_quantize_asym` operator.
Sure, let's wait for the patch.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
[GitHub] [incubator-mxnet] mxnet-bot commented on issue #18028: [MKLDNN]
Support quantized rnn towards v1.6.x
Posted by GitBox <gi...@apache.org>.
mxnet-bot commented on issue #18028: [MKLDNN] Support quantized rnn towards v1.6.x
URL: https://github.com/apache/incubator-mxnet/pull/18028#issuecomment-612349829
Hey @zixuanweeei , Thanks for submitting the PR
All tests are already queued to run once. If tests fail, you can trigger one or more tests again with the following commands:
- To trigger all jobs: @mxnet-bot run ci [all]
- To trigger specific jobs: @mxnet-bot run ci [job1, job2]
***
**CI supported jobs**: [unix-gpu, website, centos-gpu, edge, windows-gpu, sanity, centos-cpu, clang, miscellaneous, unix-cpu, windows-cpu]
***
_Note_:
Only following 3 categories can trigger CI :PR Author, MXNet Committer, Jenkins Admin.
All CI tests must pass before the PR can be merged.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
[GitHub] [incubator-mxnet] pengzhao-intel commented on issue #18028:
[MKLDNN] Support quantized rnn towards v1.6.x
Posted by GitBox <gi...@apache.org>.
pengzhao-intel commented on issue #18028: [MKLDNN] Support quantized rnn towards v1.6.x
URL: https://github.com/apache/incubator-mxnet/pull/18028#issuecomment-612756522
> LGTM for the current implementation, but we still need another DNNL patch (or DNNL version upgrade) to mitigate the overhead of `_contrib_quantize_asym` operator.
Thanks, we can upgrade DNNL a little later and I will merge this PR for early testing.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
[GitHub] [incubator-mxnet] zixuanweeei commented on issue #18028: [MKLDNN]
Support quantized rnn towards v1.6.x
Posted by GitBox <gi...@apache.org>.
zixuanweeei commented on issue #18028: [MKLDNN] Support quantized rnn towards v1.6.x
URL: https://github.com/apache/incubator-mxnet/pull/18028#issuecomment-612751927
CI checks passed. Please take a review. @ciyongch @TaoLv @pengzhao-intel
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
[GitHub] [incubator-mxnet] zixuanweeei commented on issue #18028: [MKLDNN]
Support quantized rnn towards v1.6.x
Posted by GitBox <gi...@apache.org>.
zixuanweeei commented on issue #18028: [MKLDNN] Support quantized rnn towards v1.6.x
URL: https://github.com/apache/incubator-mxnet/pull/18028#issuecomment-612742479
Accuracy and performance: https://github.com/apache/incubator-mxnet/pull/18001#issuecomment-612741705
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
[GitHub] [incubator-mxnet] ciyongch commented on issue #18028: [MKLDNN]
Support quantized rnn towards v1.6.x
Posted by GitBox <gi...@apache.org>.
ciyongch commented on issue #18028: [MKLDNN] Support quantized rnn towards v1.6.x
URL: https://github.com/apache/incubator-mxnet/pull/18028#issuecomment-612754814
LGTM for the current implementation, but we still need another DNNL patch (or DNNL version upgrade) to mitigate the overhead of `_contrib_quantize_asym` operator.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services
[GitHub] [incubator-mxnet] pengzhao-intel merged pull request #18028:
[MKLDNN] Support quantized rnn towards v1.6.x
Posted by GitBox <gi...@apache.org>.
pengzhao-intel merged pull request #18028: [MKLDNN] Support quantized rnn towards v1.6.x
URL: https://github.com/apache/incubator-mxnet/pull/18028
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services