larroy commented on issue #16995: CI timeout on unix-cpu Python2 test
URL:
https://github.com/apache/incubator-mxnet/issues/16995#issuecomment-564892769
@marcoabreu you didn't read well, because nobody wrote that it was related
to ccache, that's why it hurts.
leezu commented on issue #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#issuecomment-564891476
@larroy besides `march` there is `mtune`. But on ARM it's `mcpu` instead of
`mtune`. We also need to handle such discrepancy
oljike edited a comment on issue #16969: Gradient accumulation in Module
URL:
https://github.com/apache/incubator-mxnet/issues/16969#issuecomment-563903578
Thanks, I have resolved the problem.
This is an automated message
leezu commented on issue #17031: Use CMake standard library to handle CUDA
URL: https://github.com/apache/incubator-mxnet/pull/17031#issuecomment-564890881
Ok, thank you. I saw Marco mentioned this is done automatically every 7 days
larroy commented on issue #17038: use env var to control stack trace logging
URL: https://github.com/apache/incubator-mxnet/pull/17038#issuecomment-564890635
Why the two different names DMLC_LOG_STACK_TRACE_SIZE and
DMLC_LOG_STACK_TRACE_DEPTH? wouldn't it be better to have a single name?
larroy commented on issue #17031: Use CMake standard library to handle CUDA
URL: https://github.com/apache/incubator-mxnet/pull/17031#issuecomment-564888971
> > > i see
[apache/incubator-mxnet-ci#17](https://github.com/apache/incubator-mxnet-ci/pull/17)
is merged.When can upgrade
larroy commented on a change in pull request #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#discussion_r356994279
##
File path: CMakeLists.txt
##
@@ -28,6 +28,7 @@ option(USE_OPENMP "Build with Openmp support" ON)
leezu commented on issue #17031: Use CMake standard library to handle CUDA
URL: https://github.com/apache/incubator-mxnet/pull/17031#issuecomment-564877114
@marcoabreu can you share the date when mxnet-CI will use the updated
Windows AMI?
djaym7 opened a new issue #17053: Save quantized params in int8
URL: https://github.com/apache/incubator-mxnet/issues/17053
## Description
For the current quantize_net solution, the params are saved in fp32 format
and converted to int8 during first pass and cached in memory. It would
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 55ff222 Bump the
leezu commented on issue #17051: fix parameter names in the estimator api
URL: https://github.com/apache/incubator-mxnet/pull/17051#issuecomment-564865638
@szha @liuzh91 while the API is considered experimental, it's not good to
break it every release. Thus if it's not merged for 1.6, it's
liuzh91 commented on issue #17051: fix parameter names in the estimator api
URL: https://github.com/apache/incubator-mxnet/pull/17051#issuecomment-564864220
@ptrendx @leezu @szha Thanks for the reply. I understand the risk of
changing parameter names may break existing api. But the
szha commented on issue #16995: CI timeout on unix-cpu Python2 test
URL:
https://github.com/apache/incubator-mxnet/issues/16995#issuecomment-564863588
we should probably use parallel testing with naive engine for operator
tests, so as not to have such spillage of failure.
szha commented on issue #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#issuecomment-564863078
Binary release is one simpler use case of cross compilation. We should
probably build it as a preset in cmake which includes fine-grained control at
szha commented on issue #17051: fix parameter names in the estimator api
URL: https://github.com/apache/incubator-mxnet/pull/17051#issuecomment-564860423
@leezu @liuzh91 thanks for making the change. Note that beyond the announced
code freeze time, it's first at the discretion of the
leezu commented on issue #16852: Could not build with CUDA_ARCH_LIST=Common
URL:
https://github.com/apache/incubator-mxnet/issues/16852#issuecomment-564859663
Try building `binutils` locally with `--enable-64-bit-archive`.
leezu edited a comment on issue #16852: Could not build with
CUDA_ARCH_LIST=Common
URL:
https://github.com/apache/incubator-mxnet/issues/16852#issuecomment-564859663
Try building `binutils` locally with `--enable-64-bit-archive`.
leezu commented on issue #14263: MXNet Master build for CUDA with DEBUG=1
failing
URL:
https://github.com/apache/incubator-mxnet/issues/14263#issuecomment-564859500
@access2rohit Regarding `ar` and building the more recent version on CI,
leezu commented on a change in pull request #17031: Use CMake standard library
to handle CUDA
URL: https://github.com/apache/incubator-mxnet/pull/17031#discussion_r356965291
##
File path: 3rdparty/mshadow/CMakeLists.txt
##
@@ -1,4 +1,4 @@
-cmake_minimum_required(VERSION
leezu commented on a change in pull request #17031: Use CMake standard library
to handle CUDA
URL: https://github.com/apache/incubator-mxnet/pull/17031#discussion_r356965291
##
File path: 3rdparty/mshadow/CMakeLists.txt
##
@@ -1,4 +1,4 @@
-cmake_minimum_required(VERSION
leezu commented on issue #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#issuecomment-564856392
We need to set `-march` and `-mtune` reasonably when building releases for
other users (such as nightly build). But there's no problem to do that on
This is an automated email from the ASF dual-hosted git repository.
ptrendx pushed a change to tag 1.6.0.rc0
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
at c675520 (commit)
No new revisions were added by this update.
leezu commented on issue #17051: fix parameter names in the estimator api
URL: https://github.com/apache/incubator-mxnet/pull/17051#issuecomment-564854989
@ptrendx It unifies the naming of the API changes introduced in 1.6. It may
be useful to include for 1.6. If not, then we shouldn't
ptrendx commented on issue #17051: fix parameter names in the estimator api
URL: https://github.com/apache/incubator-mxnet/pull/17051#issuecomment-564854152
Sorry, but no. There has to be a time where no more stuff gets into the
release. I want to do RC tonight and this PR is not even
eric-haibin-lin merged pull request #17052: Backport #17007 to 1.6
URL: https://github.com/apache/incubator-mxnet/pull/17052
This is an automated message from the Apache Git Service.
To respond to the message, please log on
This is an automated email from the ASF dual-hosted git repository.
haibin pushed a commit to branch v1.6.x
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git
The following commit(s) were added to refs/heads/v1.6.x by this push:
new c675520 [BUGFIX] Fix race condition
ptrendx opened a new pull request #17052: Backport #17007 to 1.6
URL: https://github.com/apache/incubator-mxnet/pull/17052
## Description ##
Backport #17007 to 1.6
FYI @eric-haibin-lin
This is an automated message
TaoLv commented on issue #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#issuecomment-564847268
> Why do we need such gate keeper? How about defaulting to march=native
-mtune=native?
I also considered to have native as the default value.
cjolivier01 commented on issue #16995: CI timeout on unix-cpu Python2 test
URL:
https://github.com/apache/incubator-mxnet/issues/16995#issuecomment-564846017
notice the last test is (or is very near to) test_nump_op.test_np_unique
so what’s the next test after that? that could be
leezu commented on a change in pull request #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#discussion_r356946208
##
File path: CMakeLists.txt
##
@@ -28,6 +28,7 @@ option(USE_OPENMP "Build with Openmp support" ON)
TaoLv commented on a change in pull request #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#discussion_r356945554
##
File path: CMakeLists.txt
##
@@ -28,6 +28,7 @@ option(USE_OPENMP "Build with Openmp support" ON)
liuzh91 commented on issue #17051: fix parameter names in the estimator api
URL: https://github.com/apache/incubator-mxnet/pull/17051#issuecomment-564835125
@ptrendx please tag this fix for r1.6.0
This is an automated message
liuzh91 opened a new pull request #17051: fix parameter names in the estimator
api
URL: https://github.com/apache/incubator-mxnet/pull/17051
## Description ##
This fix is to make parameter names of estimator api consistent. Previously,
we have `val_metrics`, `eval_net` and
leezu commented on issue #16995: CI timeout on unix-cpu Python2 test
URL:
https://github.com/apache/incubator-mxnet/issues/16995#issuecomment-564832877
> @leezu could you link to the PRs? Is there something interesting in the
logs?
fatherMatrix closed issue #17013: How can I use mshadow in C++ as a matrix
compute framework like Eigen?
URL: https://github.com/apache/incubator-mxnet/issues/17013
This is an automated message from the Apache Git Service.
fatherMatrix commented on issue #17013: How can I use mshadow in C++ as a
matrix compute framework like Eigen?
URL:
https://github.com/apache/incubator-mxnet/issues/17013#issuecomment-564831933
Thank you very much for your reply. And I sincerely hope that the DMLC can
get better and
leezu commented on issue #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#issuecomment-564830820
> Yes, there is no need to set -msse3 explicitly if -march has a more
advanced arch. But I keep -msse3 here as a gate keeper since we won't want to
leezu commented on a change in pull request #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#discussion_r356940241
##
File path: CMakeLists.txt
##
@@ -28,6 +28,7 @@ option(USE_OPENMP "Build with Openmp support" ON)
TaoLv commented on issue #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#issuecomment-564826824
> Does it also need to interact with USE_SSE? if you set march I don't think
we need to set sse level explicitly.
Yes, there is no need to
This is an automated email from the ASF dual-hosted git repository.
ptrendx pushed a commit to branch v1.6.x
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git
The following commit(s) were added to refs/heads/v1.6.x by this push:
new 5b9f79a Omp fork numthreads fix 1.6
ptrendx merged pull request #17000: Omp fork numthreads fix 1.6
URL: https://github.com/apache/incubator-mxnet/pull/17000
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
leezu commented on issue #17031: Use CMake standard library to handle CUDA
URL: https://github.com/apache/incubator-mxnet/pull/17031#issuecomment-564819494
> > i see
[apache/incubator-mxnet-ci#17](https://github.com/apache/incubator-mxnet-ci/pull/17)
is merged.When can upgrade complete?
TaoLv edited a comment on issue #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#issuecomment-564818798
@larroy Do you have any suggestion about the name? I choose this one because:
1. `-march` is under the section `x86 options` of [gcc
TaoLv commented on issue #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#issuecomment-564818798
@larroy Do you have any suggestion about the name? I choose this one because:
1. `-march` is under the section `x86 options` of [gcc
TaoLv commented on issue #16891: Upgrading MKLDNN to 1.0 causes performance
regression.
URL:
https://github.com/apache/incubator-mxnet/issues/16891#issuecomment-564817678
@vpirogov @samskalicky Although MKL BLAS may also have positive impact to
the case demonstrated above, I think the
chinakook commented on issue #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#issuecomment-564807453
Can we link this dynamically, like MKL?
This is an automated message from the
ZhennanQin commented on issue #17047: Add build flag for x86 arch
URL: https://github.com/apache/incubator-mxnet/pull/17047#issuecomment-564806455
Shall we add this to the MXNet build features list? For now we've had below
CPU features:
```
✔ CPU_SSE
✔ CPU_SSE2
✔ CPU_SSE3
✔
eric-haibin-lin commented on issue #17010: [API] unified API for custom kvstores
URL: https://github.com/apache/incubator-mxnet/pull/17010#issuecomment-564799825
@ChaokunChang FYI
This is an automated message from the Apache
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 e112706 Bump the
This is an automated email from the ASF dual-hosted git repository.
lanking pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 05af5c4 [BUGFIX] Fix race condition in kvstore.pushpull (#17007)
add 634f95e swap wget to curl in
This is an automated email from the ASF dual-hosted git repository.
lanking pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 05af5c4 [BUGFIX] Fix race condition in kvstore.pushpull (#17007)
add 634f95e swap wget to curl in
lanking520 merged pull request #17041: swap wget to curl in Scala scripts
URL: https://github.com/apache/incubator-mxnet/pull/17041
This is an automated message from the Apache Git Service.
To respond to the message, please
TEChopra1000 opened a new pull request #17050: Website edits
URL: https://github.com/apache/incubator-mxnet/pull/17050
* I moved the autogluon images to the python _static folder (where images
are stored) so that they would get picked up during CI.
* I added the gluon data tutorials
This is an automated email from the ASF dual-hosted git repository.
haibin pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 04ebe45 Prevent after-fork number of OMP threads being bigger than 1.
(#16999)
add 05af5c4 [BUGFIX]
This is an automated email from the ASF dual-hosted git repository.
haibin pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 04ebe45 Prevent after-fork number of OMP threads being bigger than 1.
(#16999)
add 05af5c4 [BUGFIX]
eric-haibin-lin merged pull request #17007: [BUGFIX] Fix race condition in
kvstore.pushpull
URL: https://github.com/apache/incubator-mxnet/pull/17007
This is an automated message from the Apache Git Service.
To respond to
djaym7 commented on issue #14875: MXNet to ONNX export bug
URL:
https://github.com/apache/incubator-mxnet/issues/14875#issuecomment-564773614
nope, raised a ticket in Amazon but no one is currently working on this ..
This is
cjolivier01 closed issue #11417: libomp.so dependency (need REAL fix)
URL: https://github.com/apache/incubator-mxnet/issues/11417
This is an automated message from the Apache Git Service.
To respond to the message, please
cjolivier01 commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-564765440
Closing due to it not being an issue. Also, those other tickets have been
fixed.
szha commented on issue #17013: How can I use mshadow in C++ as a matrix
compute framework like Eigen?
URL:
https://github.com/apache/incubator-mxnet/issues/17013#issuecomment-564762189
We have cholesky factorization
marcoabreu commented on issue #16995: CI timeout on unix-cpu Python2 test
URL:
https://github.com/apache/incubator-mxnet/issues/16995#issuecomment-564761718
Reading this thread really hurts. Can we please have less guessing and more
rootcausing?
How should a unit test be related
This is an automated email from the ASF dual-hosted git repository.
ptrendx pushed a commit to branch v1.6.x
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git
The following commit(s) were added to refs/heads/v1.6.x by this push:
new c7dab17 Backport Gluon estimator
ptrendx merged pull request #17048: Backport Gluon estimator changes to 1.6
URL: https://github.com/apache/incubator-mxnet/pull/17048
This is an automated message from the Apache Git Service.
To respond to the message,
szha commented on a change in pull request #17038: use env var to control stack
trace logging
URL: https://github.com/apache/incubator-mxnet/pull/17038#discussion_r356856321
##
File path: ci/docker/runtime_functions.sh
##
@@ -1487,26 +1512,30 @@
szha edited a comment on issue #17038: use env var to control stack trace
logging
URL: https://github.com/apache/incubator-mxnet/pull/17038#issuecomment-564732017
@larroy 10 has been the existing limit. These changes were tested locally by
triggering errors and comparing results.
szha edited a comment on issue #17038: use env var to control stack trace
logging
URL: https://github.com/apache/incubator-mxnet/pull/17038#issuecomment-564732017
@larroy 10 has been the existing limit. These changes were tested locally by
triggering errors and comparing results.
szha commented on issue #17038: use env var to control stack trace logging
URL: https://github.com/apache/incubator-mxnet/pull/17038#issuecomment-564732017
@larroy 10 has been the existing limit. These changes were tested locally by
triggering errors and comparing results.
szha edited a comment on issue #17038: use env var to control stack trace
logging
URL: https://github.com/apache/incubator-mxnet/pull/17038#issuecomment-564732017
@larroy 10 has been the existing limit. These changes were tested locally by
triggering errors and comparing results.
This is an automated email from the ASF dual-hosted git repository.
anirudh2290 pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from c82af38 Add support of plug and play fit_batch and evaluate_batch
(#16982)
add 04ebe45 Prevent
This is an automated email from the ASF dual-hosted git repository.
anirudh2290 pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from c82af38 Add support of plug and play fit_batch and evaluate_batch
(#16982)
add 04ebe45 Prevent
anirudh2290 merged pull request #16999: Prevent after-fork number of OMP
threads being bigger than 1.
URL: https://github.com/apache/incubator-mxnet/pull/16999
This is an automated message from the Apache Git Service.
To
larroy edited a comment on issue #17031: Use CMake standard library to handle
CUDA
URL: https://github.com/apache/incubator-mxnet/pull/17031#issuecomment-564707380
> i see
[apache/incubator-mxnet-ci#17](https://github.com/apache/incubator-mxnet-ci/pull/17)
is merged.When can upgrade
larroy commented on issue #17031: Use CMake standard library to handle CUDA
URL: https://github.com/apache/incubator-mxnet/pull/17031#issuecomment-564707380
> i see
[apache/incubator-mxnet-ci#17](https://github.com/apache/incubator-mxnet-ci/pull/17)
is merged.When can upgrade complete?
larroy commented on a change in pull request #17031: Use CMake standard library
to handle CUDA
URL: https://github.com/apache/incubator-mxnet/pull/17031#discussion_r356803913
##
File path: 3rdparty/mshadow/CMakeLists.txt
##
@@ -1,4 +1,4 @@
-cmake_minimum_required(VERSION
eric-haibin-lin commented on a change in pull request #17007: [BUGFIX] Fix race
condition in kvstore.pushpull
URL: https://github.com/apache/incubator-mxnet/pull/17007#discussion_r356792730
##
File path: src/kvstore/kvstore_dist_server.h
##
@@ -364,21 +364,34 @@ class
eric-haibin-lin commented on a change in pull request #17007: [BUGFIX] Fix race
condition in kvstore.pushpull
URL: https://github.com/apache/incubator-mxnet/pull/17007#discussion_r356790519
##
File path: src/kvstore/kvstore_dist_server.h
##
@@ -364,21 +364,34 @@ class
larroy commented on issue #16995: CI timeout on unix-cpu Python2 test
URL:
https://github.com/apache/incubator-mxnet/issues/16995#issuecomment-564694243
http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/mxnet-validation%2Funix-cpu/detail/PR-16976/2/pipeline/294
larroy commented on issue #17000: Omp fork numthreads fix 1.6
URL: https://github.com/apache/incubator-mxnet/pull/17000#issuecomment-564692285
alright, thanks.
This is an automated message from the Apache Git Service.
To
ptrendx commented on issue #17000: Omp fork numthreads fix 1.6
URL: https://github.com/apache/incubator-mxnet/pull/17000#issuecomment-564691971
Since the #16999 is approved, I will merge this once it passes CI
This is an
larroy commented on issue #16995: CI timeout on unix-cpu Python2 test
URL:
https://github.com/apache/incubator-mxnet/issues/16995#issuecomment-564688591
EFS is disabled now. I think the throughput was set to max but there's too
many small files and the cache was too big. Maybe with a more
larroy commented on issue #16995: CI timeout on unix-cpu Python2 test
URL:
https://github.com/apache/incubator-mxnet/issues/16995#issuecomment-564688767
@leezu could you link to the PRs? Is there something interesting in the logs?
larroy edited a comment on issue #17038: use env var to control stack trace
logging
URL: https://github.com/apache/incubator-mxnet/pull/17038#issuecomment-564687649
> > Why do we need both an env variable and a compile flag. Also did you
test with backtrace_test in test_main.cpp ?
>
larroy commented on issue #17038: use env var to control stack trace logging
URL: https://github.com/apache/incubator-mxnet/pull/17038#issuecomment-564687649
> > Why do we need both an env variable and a compile flag. Also did you
test with backtrace_test in test_main.cpp ?
>
> The
ptrendx opened a new pull request #17049: Fix operators lying about their
number of inputs
URL: https://github.com/apache/incubator-mxnet/pull/17049
## Description ##
This PR fixes a number of backward operators that advertise the wrong number
of inputs (either by having a wrong number
larroy commented on a change in pull request #17038: use env var to control
stack trace logging
URL: https://github.com/apache/incubator-mxnet/pull/17038#discussion_r356779603
##
File path: ci/docker/runtime_functions.sh
##
@@ -1487,26 +1512,30 @@
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 3e96cc3 Bump the
samskalicky edited a comment on issue #16891: Upgrading MKLDNN to 1.0 causes
performance regression.
URL:
https://github.com/apache/incubator-mxnet/issues/16891#issuecomment-564663345
@TaoLv @pengzhao-intel Are there features in MXNet that require MKL as the
BLAS library? I was able to
samskalicky edited a comment on issue #16891: Upgrading MKLDNN to 1.0 causes
performance regression.
URL:
https://github.com/apache/incubator-mxnet/issues/16891#issuecomment-564663345
@TaoLv @pengzhao-intel Are there features in MXNet that require MKL as the
BLAS library? I was able to
samskalicky commented on issue #16891: Upgrading MKLDNN to 1.0 causes
performance regression.
URL:
https://github.com/apache/incubator-mxnet/issues/16891#issuecomment-564663345
@TaoLv @PatricZhao Are there features in MXNet that require MKL as the BLAS
library? I was able to find this
vpirogov commented on issue #16891: Upgrading MKLDNN to 1.0 causes performance
regression.
URL:
https://github.com/apache/incubator-mxnet/issues/16891#issuecomment-564650459
Hi @TaoLv, @samskalicky,
Intel MKL-DNN includes GEMM implementation that is comparable in terms of
ptrendx commented on issue #17000: Omp fork numthreads fix 1.6
URL: https://github.com/apache/incubator-mxnet/pull/17000#issuecomment-564649121
@larroy Could you update this PR with new commits you did in #16999?
This is an
ptrendx opened a new pull request #17048: Backport Gluon estimator changes to
1.6
URL: https://github.com/apache/incubator-mxnet/pull/17048
## Description ##
Backport #16957, #16971, #16982 and #17042 to 1.6 branch.
@leezu, @liuzh91 FYI
TaoLv opened a new pull request #17047: Build arch
URL: https://github.com/apache/incubator-mxnet/pull/17047
## Description ##
Add a new build flag `USE_X86_ARCH` for CPU build, so users can leverage
advanced x86 instructions when build mxnet from source code. Previously, only
sse3 is
leezu edited a comment on issue #17045: Relocation truncation issues
URL:
https://github.com/apache/incubator-mxnet/issues/17045#issuecomment-564558015
Looking at the `cmake -GNinja -DUSE_SIGNAL_HANDLER=ON -DUSE_CUDA=ON
-DUSE_CUDNN=ON -DUSE_TVM_OP=ON -DPython3_EXECUTABLE=/usr/bin/python3
leezu edited a comment on issue #17045: Relocation truncation issues
URL:
https://github.com/apache/incubator-mxnet/issues/17045#issuecomment-564558015
Looking at the `cmake -GNinja -DUSE_SIGNAL_HANDLER=ON -DUSE_CUDA=ON
-DUSE_CUDNN=ON -DUSE_TVM_OP=ON -DPython3_EXECUTABLE=/usr/bin/python3
leezu edited a comment on issue #17045: Relocation truncation issues
URL:
https://github.com/apache/incubator-mxnet/issues/17045#issuecomment-564558015
Looking at the `cmake -GNinja -DUSE_SIGNAL_HANDLER=ON -DUSE_CUDA=ON
-DUSE_CUDNN=ON -DUSE_TVM_OP=ON -DPython3_EXECUTABLE=/usr/bin/python3
leezu commented on issue #17045: Relocation truncation issues
URL:
https://github.com/apache/incubator-mxnet/issues/17045#issuecomment-564558015
Thus I think the build logic is flaky and just works due to some lucky
coincidence in the way things are linked together?
Looking at the
leezu edited a comment on issue #17045: Relocation truncation issues
URL:
https://github.com/apache/incubator-mxnet/issues/17045#issuecomment-564558015
Looking at the `cmake -GNinja -DUSE_SIGNAL_HANDLER=ON -DUSE_CUDA=ON
-DUSE_CUDNN=ON -DUSE_TVM_OP=ON -DPython3_EXECUTABLE=/usr/bin/python3
YCAyca commented on issue #17046: Error at memory location & Access violation
reading location
URL:
https://github.com/apache/incubator-mxnet/issues/17046#issuecomment-564546213
Ah ok I got it now! Sorry here is the simple code with .json and .params
files. Actually I wanted to add the
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 01288fb Bump the
1 - 100 of 111 matches
Mail list logo