[incubator-mxnet] branch master updated (3ef00b8 -> a77f774)

This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from 3ef00b8 Refactoring of Pooled Storage Manager classes (#18582) add a77f774 Remove NNPACK

[GitHub] [incubator-mxnet] szha edited a comment on issue #18682: Re-enable PR website Preview

szha edited a comment on issue #18682: URL: https://github.com/apache/incubator-mxnet/issues/18682#issuecomment-659853558 @ys2843 can this be applied onto s3, or otherwise can we get rid of the redirection? This is an

[GitHub] [incubator-mxnet] szha commented on issue #18682: Re-enable PR website Preview

szha commented on issue #18682: URL: https://github.com/apache/incubator-mxnet/issues/18682#issuecomment-659853558 @ys2843 can this be applied onto s3, or otherwise can we get ride of the redirection? This is an automated

[incubator-mxnet] branch master updated: Remove NNPACK integration (#18722)

This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git The following commit(s) were added to refs/heads/master by this push: new a77f774 Remove NNPACK integration

[GitHub] [incubator-mxnet] szha merged pull request #18722: Remove NNPACK integration

szha merged pull request #18722: URL: https://github.com/apache/incubator-mxnet/pull/18722 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

[GitHub] [incubator-mxnet] szha commented on pull request #18368: [WIP] enable large tensor in np

szha commented on pull request #18368: URL: https://github.com/apache/incubator-mxnet/pull/18368#issuecomment-659852424 @sandeep-krishnamurthy the CI is currently stuck on build issue that @TaoLv pointed out. @access2rohit 's PR on the build fix is needed to address the issue but it seems

[GitHub] [incubator-mxnet] ciyongch commented on pull request #18608: Cherry-pick #18310 #18355

ciyongch commented on pull request #18608: URL: https://github.com/apache/incubator-mxnet/pull/18608#issuecomment-659849744 Hi @ChaiBapchya , I took a look at the failure of [unix-gpu] which showed TVM compilation error and GPU OOM runtime error, seems it's more like the CI side issue.

[GitHub] [incubator-mxnet] szha commented on issue #18743: AMP: an illegal memory access was encountered

szha commented on issue #18743: URL: https://github.com/apache/incubator-mxnet/issues/18743#issuecomment-659839040 how do you reproduce the error? This is an automated message from the Apache Git Service. To respond to the

[GitHub] [incubator-mxnet] chengyuz opened a new issue #18743: AMP: an illegal memory access was encountered

chengyuz opened a new issue #18743: URL: https://github.com/apache/incubator-mxnet/issues/18743 ## Description i followed this link(https://mxnet.apache.org/api/python/docs/tutorials/performance/backend/amp.html) to enable amp in my project, but with error:

[GitHub] [incubator-mxnet] wkcn commented on pull request #18688: Fix the flaky test 'test_npx_batch_norm'

wkcn commented on pull request #18688: URL: https://github.com/apache/incubator-mxnet/pull/18688#issuecomment-659811667 @DickJC123 Thank you! The problem size could be reverted back after merging the PR #18694 : )

[GitHub] [incubator-mxnet] ptrendx commented on pull request #18735: Match the CUDA arch to the GPU used for CI testing on Windows

ptrendx commented on pull request #18735: URL: https://github.com/apache/incubator-mxnet/pull/18735#issuecomment-659807152 That is true - I was working on an instance that I thought matches the windows-gpu CI, where I saw that issue and thought the reason of this strange gap at the

[GitHub] [incubator-mxnet] ptrendx edited a comment on pull request #18735: Match the CUDA arch to the GPU used for CI testing on Windows

ptrendx edited a comment on pull request #18735: URL: https://github.com/apache/incubator-mxnet/pull/18735#issuecomment-659807152 That is true - I was working on a g4 instance that I thought matches the windows-gpu CI, where I saw that issue and thought the reason of this strange gap at

[GitHub] [incubator-mxnet] ptrendx closed pull request #18735: Match the CUDA arch to the GPU used for CI testing on Windows

ptrendx closed pull request #18735: URL: https://github.com/apache/incubator-mxnet/pull/18735 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18742: [v1.x] Cherrypick Migrate from private to public jetson toolchain files (#18677)

mxnet-bot commented on pull request #18742: URL: https://github.com/apache/incubator-mxnet/pull/18742#issuecomment-659776652 Hey @ChaiBapchya , 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

[GitHub] [incubator-mxnet] ChaiBapchya opened a new pull request #18742: [v1.x] Cherrypick Migrate from private to public jetson toolchain files (#18677)

ChaiBapchya opened a new pull request #18742: URL: https://github.com/apache/incubator-mxnet/pull/18742 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use

[GitHub] [incubator-mxnet] ChaiBapchya commented on pull request #18735: Match the CUDA arch to the GPU used for CI testing on Windows

ChaiBapchya commented on pull request #18735: URL: https://github.com/apache/incubator-mxnet/pull/18735#issuecomment-659774362 Thus we can close this PR since cuda arch for Tesla M60 is 5.2 https://developer.nvidia.com/cuda-gpus

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18741: Disable sparse op test

mxnet-bot commented on pull request #18741: URL: https://github.com/apache/incubator-mxnet/pull/18741#issuecomment-659773696 Hey @eric-haibin-lin , 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

[GitHub] [incubator-mxnet] eric-haibin-lin opened a new pull request #18741: Disable sparse op test

eric-haibin-lin opened a new pull request #18741: URL: https://github.com/apache/incubator-mxnet/pull/18741 ## Description ## Disabling this test for now to unblock other PRs, while I'm looking into it. https://github.com/apache/incubator-mxnet/issues/18740 ## Checklist ##

[GitHub] [incubator-mxnet] DickJC123 commented on a change in pull request #18694: Unittest tolerance handling improvements

DickJC123 commented on a change in pull request #18694: URL: https://github.com/apache/incubator-mxnet/pull/18694#discussion_r456164385 ## File path: Makefile ## @@ -431,7 +431,7 @@ endif # be JIT-compiled by the updated driver from the included PTX. ifeq ($(USE_CUDA), 1)

[GitHub] [incubator-mxnet] DickJC123 commented on pull request #18688: Fix the flaky test 'test_npx_batch_norm'

DickJC123 commented on pull request #18688: URL: https://github.com/apache/incubator-mxnet/pull/18688#issuecomment-659766979 My PR https://github.com/apache/incubator-mxnet/pull/18694 is now pushed and I am working toward getting a clean CI prior to merging. The sample- vs.

[incubator-mxnet-site] branch asf-site updated: Publish triggered by CI

This is an automated email from the ASF dual-hosted git repository. aaronmarkham pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git The following commit(s) were added to refs/heads/asf-site by this push: new f36fa90 Publish

[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.

This is an automated email from the ASF dual-hosted git repository. aaronmarkham pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git The following commit(s) were added to refs/heads/asf-site by this push: new c073125 Bump the

[GitHub] [incubator-mxnet] Kh4L commented on issue #18729: TensorRT tutorial is missing in master

Kh4L commented on issue #18729: URL: https://github.com/apache/incubator-mxnet/issues/18729#issuecomment-659752290 @ptrendx @aaronmarkham It has been removed in this PR https://github.com/apache/incubator-mxnet/pull/18579 cc @eric-haibin-lin

[GitHub] [incubator-mxnet] leezu commented on pull request #18722: Remove NNPACK integration

leezu commented on pull request #18722: URL: https://github.com/apache/incubator-mxnet/pull/18722#issuecomment-659750261 @mxnet-bot run ci [centos-cpu] This is an automated message from the Apache Git Service. To respond to

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18722: Remove NNPACK integration

mxnet-bot commented on pull request #18722: URL: https://github.com/apache/incubator-mxnet/pull/18722#issuecomment-659750295 Jenkins CI successfully triggered : [centos-cpu] This is an automated message from the Apache Git

[incubator-mxnet] branch master updated (2abf0b8 -> 3ef00b8)

This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from 2abf0b8 Initialize docker cache in build.py for docker-compose containers (#18724) add 3ef00b8

[incubator-mxnet] branch master updated (2abf0b8 -> 3ef00b8)

This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from 2abf0b8 Initialize docker cache in build.py for docker-compose containers (#18724) add 3ef00b8

[GitHub] [incubator-mxnet] szha merged pull request #18582: Refactoring of Pooled Storage Manager classes

szha merged pull request #18582: URL: https://github.com/apache/incubator-mxnet/pull/18582 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

[GitHub] [incubator-mxnet] szha commented on pull request #18582: Refactoring of Pooled Storage Manager classes

szha commented on pull request #18582: URL: https://github.com/apache/incubator-mxnet/pull/18582#issuecomment-659742603 @andrei5055 thanks for the refactor This is an automated message from the Apache Git Service. To respond

[GitHub] [incubator-mxnet] ChaiBapchya commented on pull request #18735: Match the CUDA arch to the GPU used for CI testing on Windows

ChaiBapchya commented on pull request #18735: URL: https://github.com/apache/incubator-mxnet/pull/18735#issuecomment-659742257 @leezu @marcoabreu Looked into the mxnet-ci isengard a/c, to find: windows-gpu instances are all G3 all G3s are Tesla M60. [discussed offline with

[GitHub] [incubator-mxnet] leezu opened a new issue #18740: test_sparse_operator.py::test_elemwise_binary_ops

leezu opened a new issue #18740: URL: https://github.com/apache/incubator-mxnet/issues/18740 ## Description Tests crashes affecting multiple PRs: https://github.com/apache/incubator-mxnet/pull/18711 https://github.com/apache/incubator-mxnet/pull/18694

[GitHub] [incubator-mxnet] leezu commented on pull request #18722: Remove NNPACK integration

leezu commented on pull request #18722: URL: https://github.com/apache/incubator-mxnet/pull/18722#issuecomment-659741070 @mxnet-bot run ci [centos-cpu] This is an automated message from the Apache Git Service. To respond to

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18722: Remove NNPACK integration

mxnet-bot commented on pull request #18722: URL: https://github.com/apache/incubator-mxnet/pull/18722#issuecomment-659741094 Jenkins CI successfully triggered : [centos-cpu] This is an automated message from the Apache Git

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18733: Move gluon.metric api docs

mxnet-bot commented on pull request #18733: URL: https://github.com/apache/incubator-mxnet/pull/18733#issuecomment-659739371 Jenkins CI successfully triggered : [centos-cpu] This is an automated message from the Apache Git

[GitHub] [incubator-mxnet] leezu commented on pull request #18733: Move gluon.metric api docs

leezu commented on pull request #18733: URL: https://github.com/apache/incubator-mxnet/pull/18733#issuecomment-659739351 @mxnet-bot run ci [centos-cpu] This is an automated message from the Apache Git Service. To respond to

[GitHub] [incubator-mxnet] wkcn commented on pull request #18707: [MXNET-1453] Support the intput whose dimension is greater than 6 for Transpose and Rollaxis

wkcn commented on pull request #18707: URL: https://github.com/apache/incubator-mxnet/pull/18707#issuecomment-659732747 @szha In numpy, the output of np.transpose is not contiguous, but the copied ndarray is contiguous. We compute each position of tranpose's output individually, but

[GitHub] [incubator-mxnet] leezu commented on a change in pull request #18690: [WIP] optimize graph in presence of dynamic shape ops

leezu commented on a change in pull request #18690: URL: https://github.com/apache/incubator-mxnet/pull/18690#discussion_r456130118 ## File path: python/mxnet/symbol/symbol.py ## @@ -1470,6 +1470,10 @@ def optimize_for(self, backend, args=None, aux=None, ctx=None,

[GitHub] [incubator-mxnet] leezu commented on pull request #18733: Move gluon.metric api docs

leezu commented on pull request #18733: URL: https://github.com/apache/incubator-mxnet/pull/18733#issuecomment-659677878 @mxnet-bot run ci [centos-cpu] This is an automated message from the Apache Git Service. To respond to

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18733: Move gluon.metric api docs

mxnet-bot commented on pull request #18733: URL: https://github.com/apache/incubator-mxnet/pull/18733#issuecomment-659677902 Jenkins CI successfully triggered : [centos-cpu] This is an automated message from the Apache Git

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18735: Match the CUDA arch to the GPU used for CI testing on Windows

mxnet-bot commented on pull request #18735: URL: https://github.com/apache/incubator-mxnet/pull/18735#issuecomment-659677769 Jenkins CI successfully triggered : [unix-cpu, windows-gpu] This is an automated message from the

[GitHub] [incubator-mxnet] leezu commented on pull request #18735: Match the CUDA arch to the GPU used for CI testing on Windows

leezu commented on pull request #18735: URL: https://github.com/apache/incubator-mxnet/pull/18735#issuecomment-659677721 @mxnet-bot run ci [unix-cpu, windows-gpu] This is an automated message from the Apache Git Service. To

[GitHub] [incubator-mxnet] leezu opened a new issue #18739: Compiling with ps-lite enabled leaves 3rdparty/ps-lite/src/meta.pb.cc 3rdparty/ps-lite/src/meta.pb.h in tree

leezu opened a new issue #18739: URL: https://github.com/apache/incubator-mxnet/issues/18739 ## Description Compilation will generate the `3rdparty/ps-lite/src/meta.pb.cc` `3rdparty/ps-lite/src/meta.pb.h` files. This causes issues when compiling using different Docker containers or

[GitHub] [incubator-mxnet] ys2843 commented on pull request #18738: set website default version - test redirect

ys2843 commented on pull request #18738: URL: https://github.com/apache/incubator-mxnet/pull/18738#issuecomment-659646149 @mxnet-label-bot add [Website] This is an automated message from the Apache Git Service. To respond to

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18738: set website default version - test redirect

mxnet-bot commented on pull request #18738: URL: https://github.com/apache/incubator-mxnet/pull/18738#issuecomment-659635114 Hey @ys2843 , 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

[GitHub] [incubator-mxnet] ys2843 opened a new pull request #18738: set website default version - test redirect

ys2843 opened a new pull request #18738: URL: https://github.com/apache/incubator-mxnet/pull/18738 ## Description ## #18429 , This PR contains the following redirect rules: When user lands on main page "/" 1. If coming from domains except `mxnet.apache.org` or

[GitHub] [incubator-mxnet] DickJC123 opened a new issue #18737: test_operator_gpu.py::test_batch_norm_with_type inadvertently retests cases, misses others

DickJC123 opened a new issue #18737: URL: https://github.com/apache/incubator-mxnet/issues/18737 ## Description Notice the test case duplication in the code:

[GitHub] [incubator-mxnet] DickJC123 commented on issue #18736: unittest/test_numpy_interoperability.py inadvertently puts additional unittests on a fixed seed

DickJC123 commented on issue #18736: URL: https://github.com/apache/incubator-mxnet/issues/18736#issuecomment-659621773 While technically a different issue, test_numpy_interoperability.py has a spurious print statement:

[GitHub] [incubator-mxnet] sandeep-krishnamurthy commented on pull request #18368: [WIP] enable large tensor in np

sandeep-krishnamurthy commented on pull request #18368: URL: https://github.com/apache/incubator-mxnet/pull/18368#issuecomment-659620714 @access2rohit - Please help review. Thanks. This is an automated message from the

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18735: Match the CUDA arch to the GPU used for CI testing on Windows

mxnet-bot commented on pull request #18735: URL: https://github.com/apache/incubator-mxnet/pull/18735#issuecomment-659600625 Jenkins CI successfully triggered : [edge] This is an automated message from the Apache Git

[GitHub] [incubator-mxnet] ptrendx commented on pull request #18735: Match the CUDA arch to the GPU used for CI testing on Windows

ptrendx commented on pull request #18735: URL: https://github.com/apache/incubator-mxnet/pull/18735#issuecomment-659600564 @mxnet-bot run ci [edge] This is an automated message from the Apache Git Service. To respond to the

[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.

This is an automated email from the ASF dual-hosted git repository. aaronmarkham pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git The following commit(s) were added to refs/heads/asf-site by this push: new 02f1618 Bump the

[incubator-mxnet-site] branch asf-site updated: Publish triggered by CI

This is an automated email from the ASF dual-hosted git repository. aaronmarkham pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git The following commit(s) were added to refs/heads/asf-site by this push: new 710fec8 Publish

[GitHub] [incubator-mxnet] DickJC123 opened a new issue #18736: unittest/test_numpy_interoperability.py inadvertently puts additional unittests on a fixed seed

DickJC123 opened a new issue #18736: URL: https://github.com/apache/incubator-mxnet/issues/18736 ## Description The following code in test_numpy_interoperability.py sets a numpy seed unilaterally, thereby making deterministic the random numbers used in subsequent unittests:

[GitHub] [incubator-mxnet] ChaiBapchya commented on pull request #18608: Cherry-pick #18310 #18355

ChaiBapchya commented on pull request #18608: URL: https://github.com/apache/incubator-mxnet/pull/18608#issuecomment-659591292 @leezu @TaoLv @ciyongch gentle ping.. for help.. unix-gpu: Any idea why specifically python3 GPU tests are failing while trying to add mkl headers? edge:

[GitHub] [incubator-mxnet] leezu commented on issue #18734: Multi-gpu Mxnet training in sagemaker gives cuda error when dataloader is using multiprocessing

leezu commented on issue #18734: URL: https://github.com/apache/incubator-mxnet/issues/18734#issuecomment-659584345 Try not using cuda before enabling the multiprocessor as a workaround. There are various bugs in the multiprocessing implementation in MXNet

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18735: Match the CUDA arch to the GPU used for CI testing on Windows

mxnet-bot commented on pull request #18735: URL: https://github.com/apache/incubator-mxnet/pull/18735#issuecomment-659583990 Hey @ptrendx , 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

[GitHub] [incubator-mxnet] ptrendx opened a new pull request #18735: Match the CUDA arch to the GPU used for CI testing on Windows

ptrendx opened a new pull request #18735: URL: https://github.com/apache/incubator-mxnet/pull/18735 ## Description ## Current build script used by CI on Windows sets the CUDA arch to 5.2, which results in the need to do ptx -> sass compilation at the beginning of the tests (depending

[GitHub] [incubator-mxnet] guptaanshul201989 opened a new issue #18734: Multi-gpu Mxnet training in sagemaker gives cuda error when dataloader is using multiprocessing

guptaanshul201989 opened a new issue #18734: URL: https://github.com/apache/incubator-mxnet/issues/18734 I am trying to train a transformer seq-to-seq model on Sagemaker ( The script I am using works fine when I run it on an EC2 multi gpu instance ). When I start a training job on

[GitHub] [incubator-mxnet] leezu commented on a change in pull request #18721: Remove Makefile

leezu commented on a change in pull request #18721: URL: https://github.com/apache/incubator-mxnet/pull/18721#discussion_r455972369 ## File path: Makefile ## @@ -1,768 +0,0 @@ -# Licensed to the Apache Software Foundation (ASF) under one -# or more contributor license

[GitHub] [incubator-mxnet] leezu commented on issue #18725: metric.py change between branches - missing docs in master

leezu commented on issue #18725: URL: https://github.com/apache/incubator-mxnet/issues/18725#issuecomment-659573454 The module should be documented in the same way as any other module on the website. The issue here is that the website build is not tested in PRs and thus it's no surprise

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18733: Move gluon.metric api docs

mxnet-bot commented on pull request #18733: URL: https://github.com/apache/incubator-mxnet/pull/18733#issuecomment-659573656 Hey @leezu , 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

[GitHub] [incubator-mxnet] leezu opened a new pull request #18733: Move gluon.metric api docs

leezu opened a new pull request #18733: URL: https://github.com/apache/incubator-mxnet/pull/18733 Fixes https://github.com/apache/incubator-mxnet/issues/18725 This is an automated message from the Apache Git Service. To

[GitHub] [incubator-mxnet] leezu opened a new issue #18732: Website build is untested

leezu opened a new issue #18732: URL: https://github.com/apache/incubator-mxnet/issues/18732 Website is built from master branch; but there is no PR check ensuring that a PR does not break the website build. Thus website breaks frequently -

[incubator-mxnet] branch master updated (37bdf0b -> 2abf0b8)

This is an automated email from the ASF dual-hosted git repository. lausen pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from 37bdf0b [MXNET-1453] Support the intput whose dimension is greater than 6 for Transpose and Rollaxis

[GitHub] [incubator-mxnet] leezu merged pull request #18724: Initialize docker cache in build.py for docker-compose containers

leezu merged pull request #18724: URL: https://github.com/apache/incubator-mxnet/pull/18724 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

[incubator-mxnet] branch master updated (8198442 -> 37bdf0b)

This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from 8198442 [numpy] symbolic advanced indexing (#18319) add 37bdf0b [MXNET-1453] Support the intput

[incubator-mxnet] branch master updated: [MXNET-1453] Support the intput whose dimension is greater than 6 for Transpose and Rollaxis (#18707)

This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git The following commit(s) were added to refs/heads/master by this push: new 37bdf0b [MXNET-1453] Support the

[GitHub] [incubator-mxnet] szha closed issue #18695: Transpose only supports up to 6 dimensions

szha closed issue #18695: URL: https://github.com/apache/incubator-mxnet/issues/18695 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

[GitHub] [incubator-mxnet] szha merged pull request #18707: [MXNET-1453] Support the intput whose dimension is greater than 6 for Transpose and Rollaxis

szha merged pull request #18707: URL: https://github.com/apache/incubator-mxnet/pull/18707 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

[GitHub] [incubator-mxnet] szha commented on pull request #18707: [MXNET-1453] Support the intput whose dimension is greater than 6 for Transpose and Rollaxis

szha commented on pull request #18707: URL: https://github.com/apache/incubator-mxnet/pull/18707#issuecomment-659556617 the comparison with numpy is unfair in that numpy just sets strides and doesn't have to be c contiguous.

[GitHub] [incubator-mxnet] DickJC123 commented on pull request #18711: Fix (log_)softmax backward on empty ndarray

DickJC123 commented on pull request #18711: URL: https://github.com/apache/incubator-mxnet/pull/18711#issuecomment-659516305 My PR is having the same 'worker crash' problem this PR is having. Without understanding the root cause, I tried marking the failing test in

[GitHub] [incubator-mxnet] DickJC123 edited a comment on pull request #18711: Fix (log_)softmax backward on empty ndarray

DickJC123 edited a comment on pull request #18711: URL: https://github.com/apache/incubator-mxnet/pull/18711#issuecomment-659516305 My PR is having the same 'worker crash' problem this PR is having. Without understanding the root cause, I tried marking the failing test in

[GitHub] [incubator-mxnet] ptrendx commented on issue #18729: TensorRT tutorial is missing in master

ptrendx commented on issue #18729: URL: https://github.com/apache/incubator-mxnet/issues/18729#issuecomment-659506296 @Kh4L Could you check if the content from 1.x is ok / modify the example to work with the current master? @aaronmarkham You gave link to 1.6 version, which seems to

[GitHub] [incubator-mxnet] szha commented on issue #18543: Gradient checkpointing in the Gluon interface

szha commented on issue #18543: URL: https://github.com/apache/incubator-mxnet/issues/18543#issuecomment-659493480 @eric-haibin-lin @ArmageddonKnight could you update on this? This is an automated message from the Apache

[GitHub] [incubator-mxnet] Masquerade0097 edited a comment on issue #18667: MXNet serialization format depends on endianness

Masquerade0097 edited a comment on issue #18667: URL: https://github.com/apache/incubator-mxnet/issues/18667#issuecomment-655500718 @leezu Thanks for your reply. > For now, you can't use the pretrained models directly on s390x architecture but you'd need some workaround where you

[GitHub] [incubator-mxnet] chinakook commented on issue #18543: Gradient checkpointing in the Gluon interface

chinakook commented on issue #18543: URL: https://github.com/apache/incubator-mxnet/issues/18543#issuecomment-659456175 It's an urgent job because the MXNet 2.0 delete the symbolic executor API. This is an automated message

[GitHub] [incubator-mxnet] woreom closed issue #18536: MXNET build not working on armv7 (raspberry pi 4)

woreom closed issue #18536: URL: https://github.com/apache/incubator-mxnet/issues/18536 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

[GitHub] [incubator-mxnet] woreom commented on issue #18536: MXNET build not working on armv7 (raspberry pi 4)

woreom commented on issue #18536: URL: https://github.com/apache/incubator-mxnet/issues/18536#issuecomment-659391586 @Qengineering Thank you for your response, It seems that the CC compiler is the issue however I used GCC-4.9 to overcome it, I wanted to use OpenCV so I had to compile

[GitHub] [incubator-mxnet] lilipj commented on issue #17395: Keras model to ONNX to mxnet error

lilipj commented on issue #17395: URL: https://github.com/apache/incubator-mxnet/issues/17395#issuecomment-659387413 I've used the onnxmltool.convert_keras function, it works with target_opset=8 (9 fails) This is an

[incubator-mxnet-site] branch asf-site updated: Publish triggered by CI

This is an automated email from the ASF dual-hosted git repository. aaronmarkham pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git The following commit(s) were added to refs/heads/asf-site by this push: new 1b26cac Publish

[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.

This is an automated email from the ASF dual-hosted git repository. aaronmarkham pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git The following commit(s) were added to refs/heads/asf-site by this push: new 444aad1 Bump the

[GitHub] [incubator-mxnet] Qengineering edited a comment on issue #18628: Raspberry pi 4 compile failure

Qengineering edited a comment on issue #18628: URL: https://github.com/apache/incubator-mxnet/issues/18628#issuecomment-659372967 I've built the latest version of MXNet (2.0.0) on a Raspberry Pi with 32 bit and 64 bit OS successfully. The instructions can be found here

[GitHub] [incubator-mxnet] Qengineering commented on issue #18628: Raspberry pi 4 compile failure

Qengineering commented on issue #18628: URL: https://github.com/apache/incubator-mxnet/issues/18628#issuecomment-659372967 I've built the latest version of MXNet (2.0.0) on a Raspberry Pi with 32 bit and 64 bit OS successfully. The instruction can be found here

[GitHub] [incubator-mxnet] Qengineering edited a comment on issue #18628: Raspberry pi 4 compile failure

Qengineering edited a comment on issue #18628: URL: https://github.com/apache/incubator-mxnet/issues/18628#issuecomment-659372967 I've built the latest version of MXNet (2.0.0) on a Raspberry Pi with 32 bit and 64 bit OS successfully. The instruction can be found here

[GitHub] [incubator-mxnet] Qengineering edited a comment on issue #18536: MXNET build not working on armv7 (raspberry pi 4)

Qengineering edited a comment on issue #18536: URL: https://github.com/apache/incubator-mxnet/issues/18536#issuecomment-659370080 I've had the same problem with the memory exhausting. It has to do with GNU compiler. The compiler is very memory hungry, especially when long chains off

[GitHub] [incubator-mxnet] Qengineering edited a comment on issue #18536: MXNET build not working on armv7 (raspberry pi 4)

Qengineering edited a comment on issue #18536: URL: https://github.com/apache/incubator-mxnet/issues/18536#issuecomment-659370080 I've had the same problem with the memory exhausting. It has to do with GNU compiler. The compiler is very memory hungry, especially when long chains off

[GitHub] [incubator-mxnet] Qengineering commented on issue #18536: MXNET build not working on armv7 (raspberry pi 4)

Qengineering commented on issue #18536: URL: https://github.com/apache/incubator-mxnet/issues/18536#issuecomment-659370080 I've had the same problem with the memory exhausting. It has to do with GNU compiler. The compiler is very memory hungry, especially when long chains off template

[GitHub] [incubator-mxnet] wkcn edited a comment on pull request #18707: [MXNET-1453] Support the intput whose dimension is greater than 6 for Transpose and Rollaxis

wkcn edited a comment on pull request #18707: URL: https://github.com/apache/incubator-mxnet/pull/18707#issuecomment-659304799 Performance Benchmark: transpose operator on CPU, **axes is generated randomly** ndim | max use (kb) | avg time (ms) ---|---|-- 1|12582.9121|1.0786

[GitHub] [incubator-mxnet] wkcn edited a comment on pull request #18707: [MXNET-1453] Support the intput whose dimension is greater than 6 for Transpose and Rollaxis

wkcn edited a comment on pull request #18707: URL: https://github.com/apache/incubator-mxnet/pull/18707#issuecomment-659304799 Performance Benchmark: transpose operator on CPU, **axes is generated randomly** ndim | max use (kb) | avg time (ms) ---|---|-- 1|12582.9121|1.0786

[GitHub] [incubator-mxnet] wkcn edited a comment on pull request #18707: [MXNET-1453] Support the intput whose dimension is greater than 6 for Transpose and Rollaxis

wkcn edited a comment on pull request #18707: URL: https://github.com/apache/incubator-mxnet/pull/18707#issuecomment-659304799 Performance Benchmark: transpose operator on CPU, **axes is generated randomly** ndim | max use (kb) | avg time (ms) ---|---|-- 1|12582.9121|1.0786

[GitHub] [incubator-mxnet] wkcn edited a comment on pull request #18707: [MXNET-1453] Support the intput whose dimension is greater than 6 for Transpose and Rollaxis

wkcn edited a comment on pull request #18707: URL: https://github.com/apache/incubator-mxnet/pull/18707#issuecomment-659304799 Performance Benchmark: transpose operator on CPU, axes is generated randomly ndim | max use (kb) | avg time (ms) ---|---|-- 1|12582.9121|1.0786

[GitHub] [incubator-mxnet] wkcn edited a comment on pull request #18707: [MXNET-1453] Support the intput whose dimension is greater than 6 for Transpose and Rollaxis

wkcn edited a comment on pull request #18707: URL: https://github.com/apache/incubator-mxnet/pull/18707#issuecomment-659304799 Performance Benchmark: transpose operator on CPU, axes is generated randomly ndim | max use (kb) | avg time (ms) ---|---|-- 1|12582.9121|1.0786

[GitHub] [incubator-mxnet] wkcn commented on pull request #18707: [MXNET-1453] Support the intput whose dimension is greater than 6 for Transpose and Rollaxis

wkcn commented on pull request #18707: URL: https://github.com/apache/incubator-mxnet/pull/18707#issuecomment-659304799 Performance Benchmark: transpose operator on CPU, axes is generated randomly ndim | max use (kb) | avg time (ms) ---|---|-- 1|12582.9121|1.0786

[GitHub] [incubator-mxnet] wkcn commented on pull request #18707: [MXNET-1453] Support the intput whose dimension is greater than 6 for Transpose and Rollaxis

wkcn commented on pull request #18707: URL: https://github.com/apache/incubator-mxnet/pull/18707#issuecomment-659286009 @szha Thank you for the suggestion! When axes.ndim() <= 6, it will call the original function `TransposeImpl`. `TransposeImpl` does not allocate any workspace, and

svn commit: r40510 - in /dev/incubator/mxnet/1.7.0.rc1: ./ apache-mxnet-src-1.7.0.rc1-incubating.tar.gz apache-mxnet-src-1.7.0.rc1-incubating.tar.gz.asc apache-mxnet-src-1.7.0.rc1-incubating.tar.gz.sh

Author: taolv Date: Thu Jul 16 08:15:47 2020 New Revision: 40510 Log: Add mxnet-1.7.0.rc1 Added: dev/incubator/mxnet/1.7.0.rc1/ dev/incubator/mxnet/1.7.0.rc1/apache-mxnet-src-1.7.0.rc1-incubating.tar.gz (with props)

[GitHub] [incubator-mxnet] szha commented on pull request #18706: [1.x] Revert Fix memory leaks in Gluon in 1.x branch

szha commented on pull request #18706: URL: https://github.com/apache/incubator-mxnet/pull/18706#issuecomment-659214645 Oh right, I just realized that _BlockScope is gone. This is an automated message from the Apache Git

[GitHub] [incubator-mxnet] leezu commented on issue #18210: Github Actions: OS X jobs canceled

leezu commented on issue #18210: URL: https://github.com/apache/incubator-mxnet/issues/18210#issuecomment-659202975 Github needs to fix it This is an automated message from the Apache Git Service. To respond to the

[incubator-mxnet] branch master updated (6901325 -> 8198442)

This is an automated email from the ASF dual-hosted git repository. lausen pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from 6901325 Add the newest mxnet discuss version. Add d2l.ai (#18663) add 8198442 [numpy] symbolic

[GitHub] [incubator-mxnet] leezu commented on pull request #18706: [1.x] Revert Fix memory leaks in Gluon in 1.x branch

leezu commented on pull request #18706: URL: https://github.com/apache/incubator-mxnet/pull/18706#issuecomment-659202436 The issue does not apply to master. This is an automated message from the Apache Git Service. To

[incubator-mxnet] branch master updated (6901325 -> 8198442)

This is an automated email from the ASF dual-hosted git repository. lausen pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from 6901325 Add the newest mxnet discuss version. Add d2l.ai (#18663) add 8198442 [numpy] symbolic

  1   2   >