[GitHub] [incubator-mxnet] Talgin commented on issue #15553: Installing mxnet on PowerPC

Talgin commented on issue #15553: URL: https://github.com/apache/incubator-mxnet/issues/15553#issuecomment-625645122 @pasa13142, we had the same problem, training was slower on power9 with Tesla V100 than on my workstation. I do not remember if they found a reason, but our server

[GitHub] [incubator-mxnet] vexilligera commented on issue #18184: Flaky Incorrect Results test_numpy_op.py:: test_np_linalg_slogdet

vexilligera commented on issue #18184: URL: https://github.com/apache/incubator-mxnet/issues/18184#issuecomment-625638615 > @vexilligera just making sure that you're running it on windows. I observed that the error only happens on windows. Of course, on G3 with Windows Server 2016.

[GitHub] [incubator-mxnet] ciyongch commented on pull request #18261: [v1.7.x] Backport fixing batch_norm and layer_norm for large tensors (#17805)

ciyongch commented on pull request #18261: URL: https://github.com/apache/incubator-mxnet/pull/18261#issuecomment-625636061 Seems it failed to fetch the external packages for TVM, please re-trigger the CI. This is an

[GitHub] [incubator-mxnet] TaoLv commented on issue #18255: MKL_USE_STATIC_LIBS broken

TaoLv commented on issue #18255: URL: https://github.com/apache/incubator-mxnet/issues/18255#issuecomment-625625333 I cannot reproduce the issue. Pulled the latest master branch and built it with the command line as below (I dont have lapack and ninja in my system): ``` cmake

[GitHub] [incubator-mxnet] szha commented on issue #18184: Flaky Incorrect Results test_numpy_op.py:: test_np_linalg_slogdet

szha commented on issue #18184: URL: https://github.com/apache/incubator-mxnet/issues/18184#issuecomment-625624201 @vexilligera just making sure that you're running it on windows. I observed that the error only happens on windows.

[GitHub] [incubator-mxnet] vexilligera commented on issue #18184: Flaky Incorrect Results test_numpy_op.py:: test_np_linalg_slogdet

vexilligera commented on issue #18184: URL: https://github.com/apache/incubator-mxnet/issues/18184#issuecomment-625612824 > Did you try on G3 instances? The other question is why the `rtol=atol=1e-1` and not lower I later reran the test on a G3 instance for 100 times with tolerance

[GitHub] [incubator-mxnet] ciyongch commented on issue #18246: Night test fail with "test_large_vector.test_nn" on v1.7.x branch

ciyongch commented on issue #18246: URL: https://github.com/apache/incubator-mxnet/issues/18246#issuecomment-625602823 Hi @ChaiBapchya , thanks for your kindly help to identify and point out the fix:) This is an automated

[GitHub] [incubator-mxnet] ChaiBapchya commented on pull request #18261: [v1.7.x] Backport fixing batch_norm and layer_norm for large tensors (#17805)

ChaiBapchya commented on pull request #18261: URL: https://github.com/apache/incubator-mxnet/pull/18261#issuecomment-625597268 Fixes https://github.com/apache/incubator-mxnet/issues/18246 This is an automated message from

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18261: [v1.7.x] Backport fixing batch_norm and layer_norm for large tensors (#17805)

mxnet-bot commented on pull request #18261: URL: https://github.com/apache/incubator-mxnet/pull/18261#issuecomment-625596968 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 #18261: [v1.7.x] Backport fixing batch_norm and layer_norm for large tensors (#17805)

ChaiBapchya opened a new pull request #18261: URL: https://github.com/apache/incubator-mxnet/pull/18261 Backport to fix large tensor nightly test issue Co-authored-by: Rohit Kumar Srivastava Verified the changes locally by setting up the Infra

[GitHub] [incubator-mxnet] jaisanant0 commented on issue #15275: How to run mxnet(C++) in single-thread mode?

jaisanant0 commented on issue #15275: URL: https://github.com/apache/incubator-mxnet/issues/15275#issuecomment-62559 @cyrusbehr as you mentioned i have moved the variables mxnetConfig1 mxnetConfig2 into class member but the number of threads spawned is qual to number of

[GitHub] [incubator-mxnet] ciyongch commented on pull request #17959: [MKLDNN] Add LSTMP to v1.6.x

ciyongch commented on pull request #17959: URL: https://github.com/apache/incubator-mxnet/pull/17959#issuecomment-625589023 Thanks for your understand and help making it clear @TaoLv @ChaiBapchya . This is an automated

[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 0ecd800 Bump the

[incubator-mxnet] branch master updated: fixed overwrite of args/aux variables (#18232)

This is an automated email from the ASF dual-hosted git repository. marcoabreu 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 68cb955 fixed overwrite of

[incubator-mxnet] branch master updated: fixed overwrite of args/aux variables (#18232)

This is an automated email from the ASF dual-hosted git repository. marcoabreu 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 68cb955 fixed overwrite of

[incubator-mxnet] branch master updated: Change include to relative in nvvm_to_onnx.cc (#18249)

This is an automated email from the ASF dual-hosted git repository. marcoabreu 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 21b187b Change include to

[incubator-mxnet] branch master updated (353c243 -> 7f24823)

This is an automated email from the ASF dual-hosted git repository. marcoabreu pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from 353c243 Remove duplicate large_vector_test (#18259) add 7f24823 Fixing #16655 (#18257) No new

[incubator-mxnet] branch master updated (353c243 -> 7f24823)

This is an automated email from the ASF dual-hosted git repository. marcoabreu pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from 353c243 Remove duplicate large_vector_test (#18259) add 7f24823 Fixing #16655 (#18257) No new

[incubator-mxnet] branch master updated (a0e6735 -> 353c243)

This is an automated email from the ASF dual-hosted git repository. marcoabreu pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from a0e6735 Gluon data 2.0: c++ dataloader and built-in image/bbox transforms (#17841) add 353c243

[incubator-mxnet] branch master updated (a0e6735 -> 353c243)

This is an automated email from the ASF dual-hosted git repository. marcoabreu pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from a0e6735 Gluon data 2.0: c++ dataloader and built-in image/bbox transforms (#17841) add 353c243

[incubator-mxnet] branch master updated (353c243 -> 7f24823)

This is an automated email from the ASF dual-hosted git repository. marcoabreu pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from 353c243 Remove duplicate large_vector_test (#18259) add 7f24823 Fixing #16655 (#18257) No new

[incubator-mxnet] branch master updated (a0e6735 -> 353c243)

This is an automated email from the ASF dual-hosted git repository. marcoabreu pushed a change to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git. from a0e6735 Gluon data 2.0: c++ dataloader and built-in image/bbox transforms (#17841) add 353c243

[GitHub] [incubator-mxnet] access2rohit commented on issue #17331: [mxnet 2.0] [item 2.4] Turning on large tensor support by default

access2rohit commented on issue #17331: URL: https://github.com/apache/incubator-mxnet/issues/17331#issuecomment-625542495 @apeforest THe profiling done by @JonTanS was done long back using mxnet-1.6in november. These results are using current master branch of MXNet, bert scripts have

[GitHub] [incubator-mxnet] ChaiBapchya commented on issue #18246: Night test fail with "test_large_vector.test_nn" on v1.7.x branch

ChaiBapchya commented on issue #18246: URL: https://github.com/apache/incubator-mxnet/issues/18246#issuecomment-625539228 layer norm and batchnorm have been fixed in master by https://github.com/apache/incubator-mxnet/pull/17805 But they aren't cherry-picked in 1.7.x as of today.

[GitHub] [incubator-mxnet] sxjscience commented on issue #18117: [Gluon][Numpy] Bug of save parameters with numpy front-end

sxjscience commented on issue #18117: URL: https://github.com/apache/incubator-mxnet/issues/18117#issuecomment-625506608 I think we should not deprecate it. Get Outlook for iOS From: Leonard Lausen Sent: Thursday, May

[GitHub] [incubator-mxnet] leezu commented on issue #18117: [Gluon][Numpy] Bug of save parameters with numpy front-end

leezu commented on issue #18117: URL: https://github.com/apache/incubator-mxnet/issues/18117#issuecomment-625502229 @sxjscience being able to save a single parameter seems quite sensible. Why should we deprecate it? We may change the serialization format though

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18260: Fix interleave matmul doc

mxnet-bot commented on pull request #18260: URL: https://github.com/apache/incubator-mxnet/pull/18260#issuecomment-625476566 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 #18260: Fix interleave matmul doc

eric-haibin-lin opened a new pull request #18260: URL: https://github.com/apache/incubator-mxnet/pull/18260 ## Description ## Fix interleave matmul doc ## Checklist ## ### Essentials ### Please feel free to remove inapplicable items for your PR. - [ ] The PR title starts

[GitHub] [incubator-mxnet] ys2843 commented on pull request #18238: fix when clicking version dropdown it jumps to top of the page

ys2843 commented on pull request #18238: URL: https://github.com/apache/incubator-mxnet/pull/18238#issuecomment-625458290 @ChaiBapchya The change should apply to master now, could you try disable cache and reload the website? Thanks.

[GitHub] [incubator-mxnet] ChaiBapchya commented on pull request #18238: fix when clicking version dropdown it jumps to top of the page

ChaiBapchya commented on pull request #18238: URL: https://github.com/apache/incubator-mxnet/pull/18238#issuecomment-625444923 Works good when the global version of website is 1.6.0 But when global version is master And MXNet version selection dropdown in Get Started page is chosen

[GitHub] [incubator-mxnet] ChaiBapchya commented on pull request #18259: Remove duplicate large_vector_test in nightly introduced in #17388

ChaiBapchya commented on pull request #18259: URL: https://github.com/apache/incubator-mxnet/pull/18259#issuecomment-625436222 @access2rohit @apeforest This is an automated message from the Apache Git Service. To respond to

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18259: Remove duplicate large_vector_test in nightly introduced in #17388

mxnet-bot commented on pull request #18259: URL: https://github.com/apache/incubator-mxnet/pull/18259#issuecomment-625435955 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 #18259: Remove duplicate large_vector_test in nightly introduced in #17388

ChaiBapchya opened a new pull request #18259: URL: https://github.com/apache/incubator-mxnet/pull/18259 ## Description ## Remove duplicate test This is an automated message from the Apache Git Service. To respond to the

[GitHub] [incubator-mxnet] kmfrick commented on issue #18057: Edge device documentation

kmfrick commented on issue #18057: URL: https://github.com/apache/incubator-mxnet/issues/18057#issuecomment-625431210 Yeah I'm building 1.6 now, I noticed half-way through the build that 1.4 was not the most recent release. I'll make a PR against that document when I have some time.

[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 abb937b Bump the

[GitHub] [incubator-mxnet] leezu commented on issue #15553: Installing mxnet on PowerPC

leezu commented on issue #15553: URL: https://github.com/apache/incubator-mxnet/issues/15553#issuecomment-625429275 You need to investigate and find some minimal example that shows the problem. For example, by using the benchmark scripts

[GitHub] [incubator-mxnet] leezu commented on a change in pull request #18083: Changes to mxnet.metric

leezu commented on a change in pull request #18083: URL: https://github.com/apache/incubator-mxnet/pull/18083#discussion_r421715173 ## File path: tests/python/tensorrt/test_cvnets.py ## @@ -29,7 +28,12 @@ def get_classif_model(model_name, use_tensorrt, ctx=mx.gpu(0),

[GitHub] [incubator-mxnet] leezu commented on a change in pull request #18083: Changes to mxnet.metric

leezu commented on a change in pull request #18083: URL: https://github.com/apache/incubator-mxnet/pull/18083#discussion_r421715173 ## File path: tests/python/tensorrt/test_cvnets.py ## @@ -29,7 +28,12 @@ def get_classif_model(model_name, use_tensorrt, ctx=mx.gpu(0),

[GitHub] [incubator-mxnet] leezu commented on issue #18057: Edge device documentation

leezu commented on issue #18057: URL: https://github.com/apache/incubator-mxnet/issues/18057#issuecomment-625427543 @kmfrick you can consider making a PR to update https://github.com/apache/incubator-mxnet/blob/master/docs/static_site/src/pages/get_started/jetson_setup.md If you

[GitHub] [incubator-mxnet] leezu commented on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

leezu commented on issue #13773: URL: https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625425470 You shouldn't downgrade the GCC version. This is a compiler bug. You need to update the version to get a gcc version that has the fix for your bug... If no version

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #17852: [mxnet2.0] gluon 2.0 data: deprecate dataset transform= argument

mxnet-bot commented on pull request #17852: URL: https://github.com/apache/incubator-mxnet/pull/17852#issuecomment-625398499 Jenkins CI successfully triggered : [edge, centos-cpu, centos-gpu, unix-gpu, windows-gpu, unix-cpu, windows-cpu, clang, sanity, website, miscellaneous]

[GitHub] [incubator-mxnet] zhreshold commented on pull request #17852: [mxnet2.0] gluon 2.0 data: deprecate dataset transform= argument

zhreshold commented on pull request #17852: URL: https://github.com/apache/incubator-mxnet/pull/17852#issuecomment-625398374 @mxnet-bot run ci [all] This is an automated message from the Apache Git Service. To respond to the

[GitHub] [incubator-mxnet] access2rohit commented on issue #18258: Opperf fails when running for all operators (both CPU and GPU ctx)

access2rohit commented on issue #18258: URL: https://github.com/apache/incubator-mxnet/issues/18258#issuecomment-625395133 @ChaiBapchya ^ This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] [incubator-mxnet] access2rohit opened a new issue #18258: Opperf fails when running for all operators (both CPU and GPU ctx)

access2rohit opened a new issue #18258: URL: https://github.com/apache/incubator-mxnet/issues/18258 ## Description I ran the following command to obtain individual operator performance w/ and w/o Large Tensor Support on both CPU and GPU ``` #CPU python

[GitHub] [incubator-mxnet] JinboCi opened a new pull request #18257: Fixing #16655

JinboCi opened a new pull request #18257: URL: https://github.com/apache/incubator-mxnet/pull/18257 ## Description ## (Brief description on what this PR is about) This pull request is opened to fix the issue [#16655](https://github.com/apache/incubator-mxnet/issues/16655). The

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18257: Fixing #16655

mxnet-bot commented on pull request #18257: URL: https://github.com/apache/incubator-mxnet/pull/18257#issuecomment-625383183 Hey @JinboCi , 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] ChaiBapchya commented on a change in pull request #18186: Update unix gpu toolchain

ChaiBapchya commented on a change in pull request #18186: URL: https://github.com/apache/incubator-mxnet/pull/18186#discussion_r421651875 ## File path: tests/python/unittest/test_numpy_op.py ## @@ -2220,6 +,7 @@ def hybrid_forward(self, F, x): assert

[GitHub] [incubator-mxnet] ChaiBapchya commented on pull request #17959: [MKLDNN] Add LSTMP to v1.6.x

ChaiBapchya commented on pull request #17959: URL: https://github.com/apache/incubator-mxnet/pull/17959#issuecomment-625373529 @TaoLv Makes sense. Thanks for the clarification. Let's circle back to this point if rc1 is required.

[GitHub] [incubator-mxnet] flygragon commented on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

flygragon commented on issue #13773: URL: https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625364871 > is 7.3.8 your default gcc version? Or you can try downgrade your gcc and g++. > > And check if there is enough swap memory. My gcc version is 7.5.0, RAM

[GitHub] [incubator-mxnet] kohillyang edited a comment on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

kohillyang edited a comment on issue #13773: URL: https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625356733 is 7.3.8 your default gcc version? Or you can try downgrade your gcc and g++. And check if there is enough swap memory.

[GitHub] [incubator-mxnet] leezu commented on issue #18255: MKL_USE_STATIC_LIBS broken

leezu commented on issue #18255: URL: https://github.com/apache/incubator-mxnet/issues/18255#issuecomment-625357850 mkl is used if available by default. This is an automated message from the Apache Git Service. To respond

[GitHub] [incubator-mxnet] kohillyang commented on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

kohillyang commented on issue #13773: URL: https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625356733 is 7.3.8 your default gcc version? Or you can try downgrade your gcc and g++. This is an automated

[GitHub] [incubator-mxnet] flygragon commented on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

flygragon commented on issue #13773: URL: https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625351814 > > Howerver, today I encountered the same issue, and always with the same file, after a carton of the machine. I was working with Ubuntu 18.04 with 8G mem. Could some

[GitHub] [incubator-mxnet] kohillyang commented on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

kohillyang commented on issue #13773: URL: https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625348429 > Howerver, today I encountered the same issue, and always with the same file, after a carton of the machine. I was working with Ubuntu 18.04 with 8G mem. Could some

[GitHub] [incubator-mxnet] kohillyang edited a comment on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

kohillyang edited a comment on issue #13773: URL: https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625348429 > Howerver, today I encountered the same issue, and always with the same file, after a carton of the machine. I was working with Ubuntu 18.04 with 8G mem. Could

[GitHub] [incubator-mxnet] flygragon edited a comment on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

flygragon edited a comment on issue #13773: URL: https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625344577 Howerver, today I encountered the same issue, and always with the same file, after a carton of the machine. I was working with Ubuntu 18.04 with 8G mem. Could

[GitHub] [incubator-mxnet] flygragon commented on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

flygragon commented on issue #13773: URL: https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625344577 Howerver, today I encountered the same issue, and always with the same file, after a carton of the machine. I was working with Ubuntu 18.04 with 8G mem. Could some one

[GitHub] [incubator-mxnet] TaoLv commented on pull request #17959: [MKLDNN] Add LSTMP to v1.6.x

TaoLv commented on pull request #17959: URL: https://github.com/apache/incubator-mxnet/pull/17959#issuecomment-625330433 @ChaiBapchya @ciyongch As far as I know, the feature lives on 1.6.x branch but is not released in any release yet. Given the code of 1.7.0 has been frozen for a while

[GitHub] [incubator-mxnet] mikeobr commented on issue #16173: Saving and loading cudNN autotune and graph optimization

mikeobr commented on issue #16173: URL: https://github.com/apache/incubator-mxnet/issues/16173#issuecomment-625302116 @QueensGambit I'm a bit of a noob level understanding around the internals, but if that allows us to load and start doing inference without taking the tuning hit, then it

[GitHub] [incubator-mxnet] TaoLv commented on issue #18255: MKL_USE_STATIC_LIBS broken

TaoLv commented on issue #18255: URL: https://github.com/apache/incubator-mxnet/issues/18255#issuecomment-625295428 Hi @leezu, I would expect the flag `-DMKL_USE_STATIC_LIBS` should take effect only when `-DUSE_BLAS` is set to `mkl`. That means, with your command line, mkl libraries

[GitHub] [incubator-mxnet] kmfrick edited a comment on issue #18057: Edge device documentation

kmfrick edited a comment on issue #18057: URL: https://github.com/apache/incubator-mxnet/issues/18057#issuecomment-625291919 Piggybacking on this: the pre-built library linked [here, section "Install MXNet for Python"](https://mxnet.apache.org/get_started/jetson_setup) is outdated as

[GitHub] [incubator-mxnet] kmfrick edited a comment on issue #18057: Edge device documentation

kmfrick edited a comment on issue #18057: URL: https://github.com/apache/incubator-mxnet/issues/18057#issuecomment-625291919 Piggybacking on this: the pre-built library linked [here, section "Install MXNet for Python"](https://mxnet.apache.org/get_started/jetson_setup) is outdated as

[GitHub] [incubator-mxnet] kmfrick commented on issue #18057: Edge device documentation

kmfrick commented on issue #18057: URL: https://github.com/apache/incubator-mxnet/issues/18057#issuecomment-625291919 Piggybacking on this: the pre-built library linked [here, section "Install MXNet for Python"](https://mxnet.apache.org/get_started/jetson_setup) is outdated as JetPack

[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 edf7f6a Bump the

[GitHub] [incubator-mxnet] ciyongch commented on pull request #17959: [MKLDNN] Add LSTMP to v1.6.x

ciyongch commented on pull request #17959: URL: https://github.com/apache/incubator-mxnet/pull/17959#issuecomment-625112554 >Backport to 1.6 should be included by default in 1.7 right? Hi @ChaiBapchya , not sure if it's a common practice to do so. As there're several new features

[GitHub] [incubator-mxnet] pasa13142 commented on issue #15553: Installing mxnet on PowerPC

pasa13142 commented on issue #15553: URL: https://github.com/apache/incubator-mxnet/issues/15553#issuecomment-625089795 @leezu @Talgin no idea guys how can i fix this ? :( This is an automated message from the Apache Git

[GitHub] [incubator-mxnet] pasa13142 edited a comment on issue #15553: Installing mxnet on PowerPC

pasa13142 edited a comment on issue #15553: URL: https://github.com/apache/incubator-mxnet/issues/15553#issuecomment-625089795 @leezu @Talgin no idea how can i fix this ? :( This is an automated message from the Apache Git

[GitHub] [incubator-mxnet] ChaiBapchya commented on pull request #17959: [MKLDNN] Add LSTMP to v1.6.x

ChaiBapchya commented on pull request #17959: URL: https://github.com/apache/incubator-mxnet/pull/17959#issuecomment-625084536 @ciyongch While the customer request is resolved with 1.6.x backport, it made sense to include it in upcoming minor release. Backport to 1.6 should be included by

[GitHub] [incubator-mxnet] leezu commented on issue #18184: Flaky Incorrect Results test_numpy_op.py:: test_np_linalg_slogdet

leezu commented on issue #18184: URL: https://github.com/apache/incubator-mxnet/issues/18184#issuecomment-625081994 Did you try on G3 instances? The other question is why the `rtol=atol=1e-1` and not lower This is an

[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 93fe491 Bump the

[GitHub] [incubator-mxnet] yzhliu commented on issue #16357: Numpy Operator Inconsistency

yzhliu commented on issue #16357: URL: https://github.com/apache/incubator-mxnet/issues/16357#issuecomment-625062177 @Yiyan66 is looking into it. This is an automated message from the Apache Git Service. To respond to the

[GitHub] [incubator-mxnet] ciyongch commented on pull request #17959: [MKLDNN] Add LSTMP to v1.6.x

ciyongch commented on pull request #17959: URL: https://github.com/apache/incubator-mxnet/pull/17959#issuecomment-625061521 I suggest to leave this new feature to the next release version (which requires upgrade oneDNN (ex: DNNL)), as the current version 1.7.0 is code freeze and suppose

[GitHub] [incubator-mxnet] sxjscience edited a comment on issue #18256: put more effort over 'easy to use'

sxjscience edited a comment on issue #18256: URL: https://github.com/apache/incubator-mxnet/issues/18256#issuecomment-625050992 @leezu Previously, we have one benchmark by @Jerryzcn : https://github.com/apache/incubator-mxnet/issues/17335 . According to my recent coding experience with

[GitHub] [incubator-mxnet] sxjscience edited a comment on issue #18256: put more effort over 'easy to use'

sxjscience edited a comment on issue #18256: URL: https://github.com/apache/incubator-mxnet/issues/18256#issuecomment-625050992 @leezu Previously, we have one benchmark by @Jerryzcn : https://github.com/apache/incubator-mxnet/issues/17335 . According to my recent coding experience with

[GitHub] [incubator-mxnet] sxjscience commented on issue #18256: put more effort over 'easy to use'

sxjscience commented on issue #18256: URL: https://github.com/apache/incubator-mxnet/issues/18256#issuecomment-625050992 @leezu Previously, we have one benchmark by @Jerryzcn : https://github.com/apache/incubator-mxnet/issues/17335 . According to my recent coding experience with the new

[GitHub] [incubator-mxnet] vexilligera edited a comment on issue #18184: Flaky Incorrect Results test_numpy_op.py:: test_np_linalg_slogdet

vexilligera edited a comment on issue #18184: URL: https://github.com/apache/incubator-mxnet/issues/18184#issuecomment-625047127 Hi, I tried to reproduce the error by running `pytest tests/python/gpu/test_operator_gpu.py::test_np_linalg_slogdet` for 20 times and didn't run into the

[GitHub] [incubator-mxnet] vexilligera commented on issue #18184: Flaky Incorrect Results test_numpy_op.py:: test_np_linalg_slogdet

vexilligera commented on issue #18184: URL: https://github.com/apache/incubator-mxnet/issues/18184#issuecomment-625047127 Hi, I tried to reproduce the error by running `pytest tests/python/gpu/test_operator_gpu.py::test_np_linalg_slogdet` for 20 times both on CPU and GPU, and didn't

[GitHub] [incubator-mxnet] leezu commented on issue #18256: put more effort over 'easy to use'

leezu commented on issue #18256: URL: https://github.com/apache/incubator-mxnet/issues/18256#issuecomment-625046431 I agree that many use-cases can be covered in a simpler fashion without hybridization. One question here is if "unhybridized mxnet" may be slower than "hybridized mxnet"