junrushao1994 opened a new pull request #16351: [IR-PATCH] Support Attrs from
Conv2d
URL: https://github.com/apache/incubator-mxnet/pull/16351
@tqchen @szha @sxjscience @eric-haibin-lin
Ideally, we should stop using `Attrs`. But it may not be a good time now as
the infra hasn't
kshitij12345 commented on issue #15515: [MXNET-978] Higher Order Gradient
Support `arcsin`, `arccos`.
URL: https://github.com/apache/incubator-mxnet/pull/15515#issuecomment-537462280
@apeforest @larroy @sxjscience Gentle Ping for review.:)
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 0523722 Bump the
ChaiBapchya edited a comment on issue #14378: Win10 build error: gtest, no
'object' file generated
URL:
https://github.com/apache/incubator-mxnet/issues/14378#issuecomment-537365815
Is this related to issue found in pipeline for unrelated PR #16328
ChaiBapchya commented on issue #14378: Win10 build error: gtest, no 'object'
file generated
URL:
https://github.com/apache/incubator-mxnet/issues/14378#issuecomment-537365815
Is this related to issue found in pipeline for #16328
```
[ 5%] Building CXX object
mxnet-label-bot commented on issue #16352: [CI] Julia Windows CPU timeout
URL:
https://github.com/apache/incubator-mxnet/issues/16352#issuecomment-537361222
Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some labels so
that the
ChaiBapchya opened a new issue #16352: [CI] Julia Windows CPU timeout
URL: https://github.com/apache/incubator-mxnet/issues/16352
http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/mxnet-validation%2Fwindows-cpu/detail/PR-16336/6/pipeline/176
#16336
Violates
ChaiBapchya commented on issue #16353:
test_operator_gpu.test_kernel_error_checking Fails
URL:
https://github.com/apache/incubator-mxnet/issues/16353#issuecomment-537364775
@mxnet-label-bot add [Test]
This is an automated
mxnet-label-bot commented on issue #16353:
test_operator_gpu.test_kernel_error_checking
URL:
https://github.com/apache/incubator-mxnet/issues/16353#issuecomment-537364632
Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some labels so
that
ChaiBapchya commented on issue #15880: [CI] unix cpu validation Timeout
URL:
https://github.com/apache/incubator-mxnet/issues/15880#issuecomment-537360425
4 hr timeout again! MKL CPU
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 c240ea35 Bump the
rondogency commented on issue #15921: dynamic custom operator support
URL: https://github.com/apache/incubator-mxnet/pull/15921#issuecomment-537376762
@samskalicky @wkcn @szha @junrushao1994 @eric-haibin-lin This PR is ready
for review.
c_api.cc and lib_api.h files are the most
ChaiBapchya commented on issue #16352: [CI] Julia Windows CPU timeout
URL:
https://github.com/apache/incubator-mxnet/issues/16352#issuecomment-537361722
@mxnet-label-bot add [CI, Test, Julia]
This is an automated message
ChaiBapchya opened a new issue #16353:
test_operator_gpu.test_kernel_error_checking
URL: https://github.com/apache/incubator-mxnet/issues/16353
UNIX-GPU
Unrelated PR #16328
Pagey commented on issue #16261: website bug: missing documentation
URL:
https://github.com/apache/incubator-mxnet/issues/16261#issuecomment-537405745
there's also a serious issue with chrome find in the new document pages- it
seems that because the docs are not in the main page frame,
iblis17 commented on issue #16348: Julia: add API docs back
URL: https://github.com/apache/incubator-mxnet/pull/16348#issuecomment-537513502
Just found that this PR isn't merged into master.
Please open another PR for merging it into master.
Always do that first, and I will backport
iblis17 commented on issue #16348: Julia: add API docs back
URL: https://github.com/apache/incubator-mxnet/pull/16348#issuecomment-537511259
I think the CI failures are unrelated.
This is an automated message from the Apache
This is an automated email from the ASF dual-hosted git repository.
iblis pushed a change to branch ib/backport-jl
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 6559b80 julia: fix `mx.forward` kwargs checking (#16138)
add f3b4377 Julia: add API docs back
iblis17 merged pull request #16348: Julia: add API docs back
URL: https://github.com/apache/incubator-mxnet/pull/16348
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
vpirogov commented on issue #16300: intel-mkl package failing to install during
ubuntu builds
URL:
https://github.com/apache/incubator-mxnet/issues/16300#issuecomment-537523013
It turns out that GPG key used to sign packages in apt channel has expired.
The issue can be resolved by
zu3st opened a new pull request #16354: Fix inconsistent interpolation method
values
URL: https://github.com/apache/incubator-mxnet/pull/16354
Documentation of `CreateDetAugmenter`s inter_method doesn't match the one of
cv2.
This has already been fixed in image.py, see the relevant PR
marcoabreu commented on a change in pull request #16341: [WIP][New Op] Add
deformable conv v2
URL: https://github.com/apache/incubator-mxnet/pull/16341#discussion_r330663090
##
File path: tests/python/unittest/test_contrib_operator.py
##
@@ -333,6 +334,46 @@ def
ChaiBapchya edited a comment on issue #15880: [CI] unix cpu validation Timeout
URL:
https://github.com/apache/incubator-mxnet/issues/15880#issuecomment-537595655
http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/mxnet-validation%2Funix-cpu/detail/PR-16336/7/pipeline
MoisesHer opened a new pull request #16355: Pr embedding gradient
URL: https://github.com/apache/incubator-mxnet/pull/16355
## Description ##
This PR includes a specific Embedding-backward operator for GPU.
Two new CUDA kernels have been implemented for improving the performance of
lanking520 commented on issue #16300: intel-mkl package failing to install
during ubuntu builds
URL:
https://github.com/apache/incubator-mxnet/issues/16300#issuecomment-537598437
@TaoLv I saw someone is trying to solve this problem, but still seemed not
working. Is there an ETA for this?
Caenorst commented on issue #16345: Flaky test: test_preloaded_multi_sgd
URL:
https://github.com/apache/incubator-mxnet/issues/16345#issuecomment-537605288
It turned out that values very close to 0. are the most inaccurate so
bumping atol instead of rtol.
ChaiBapchya commented on issue #10227: Flaky test: test_operator.test_layer_norm
URL:
https://github.com/apache/incubator-mxnet/issues/10227#issuecomment-537611017
Fixing the seed isn't a good practice is it? Removing finite difference will
make this test loose right?
sxjscience commented on issue #10227: Flaky test: test_operator.test_layer_norm
URL:
https://github.com/apache/incubator-mxnet/issues/10227#issuecomment-537614136
@ChaiBapchya I mean the test would be as strict as the original. Also,
fixing the seed works in some randomness tests.
ThomasDelteil commented on a change in pull request #16342: adding redirects so
that old website API links surfaced from searches
URL: https://github.com/apache/incubator-mxnet/pull/16342#discussion_r330660544
##
File path: docs/static_site/src/.htaccess
##
@@ -1,4 +1,3
ChaiBapchya removed a comment on issue #15880: [CI] unix cpu validation Timeout
URL:
https://github.com/apache/incubator-mxnet/issues/15880#issuecomment-537595655
http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/mxnet-validation%2Funix-cpu/detail/PR-16336/7/pipeline
ChaiBapchya commented on issue #15880: [CI] unix cpu validation Timeout
URL:
https://github.com/apache/incubator-mxnet/issues/15880#issuecomment-537595655
http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/mxnet-validation%2Funix-cpu/detail/PR-16336/7/pipeline
stu1130 opened a new issue #16357: Numpy Transpose Operator Inconsistency
URL: https://github.com/apache/incubator-mxnet/issues/16357
Numpy
```
>>>import numpy as np
>>> a = np.arange(4).reshape(2, 2)
>>> np.transpose(a, (-1, 0))
array([[0, 2],
[1, 3]])
```
mxnet-label-bot commented on issue #16357: Numpy Transpose Operator
Inconsistency
URL:
https://github.com/apache/incubator-mxnet/issues/16357#issuecomment-537613571
Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some labels so
that the
stu1130 commented on issue #16357: Numpy Transpose Operator Inconsistency
URL:
https://github.com/apache/incubator-mxnet/issues/16357#issuecomment-537613765
@mxnet-label-bot add [Numpy]
This is an automated message from the
ptrendx closed issue #15725: AMP issue in _gen_atomic_symbol
URL: https://github.com/apache/incubator-mxnet/issues/15725
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
ptrendx commented on issue #15725: AMP issue in _gen_atomic_symbol
URL:
https://github.com/apache/incubator-mxnet/issues/15725#issuecomment-537594443
This merged PR should fix the issue you are observing, if you still see the
problem, please reopen.
sxjscience commented on issue #10227: Flaky test: test_operator.test_layer_norm
URL:
https://github.com/apache/incubator-mxnet/issues/10227#issuecomment-537597790
I guess we need to use 1E-3
This is an automated message from
sxjscience commented on issue #10227: Flaky test: test_operator.test_layer_norm
URL:
https://github.com/apache/incubator-mxnet/issues/10227#issuecomment-537601528
@ChaiBapchya Yep, would it be possible to just test a few seeds? Also, we
could somehow remove the finite-difference test.
sxjscience commented on issue #10227: Flaky test: test_operator.test_layer_norm
URL:
https://github.com/apache/incubator-mxnet/issues/10227#issuecomment-537611501
@ChaiBapchya In theory it will not if our manual backward logic is correct.
apeforest closed issue #15575: CMake will modify the content in
3rdparty/dmlc-core
URL: https://github.com/apache/incubator-mxnet/issues/15575
This is an automated message from the Apache Git Service.
To respond to the
apeforest commented on issue #15575: CMake will modify the content in
3rdparty/dmlc-core
URL:
https://github.com/apache/incubator-mxnet/issues/15575#issuecomment-537591661
Thanks for the quick fix.
This is an automated
ChaiBapchya commented on issue #10227: Flaky test: test_operator.test_layer_norm
URL:
https://github.com/apache/incubator-mxnet/issues/10227#issuecomment-537596535
Happened again: #16336 (unrelated PR)
ChaiBapchya commented on issue #10227: Flaky test: test_operator.test_layer_norm
URL:
https://github.com/apache/incubator-mxnet/issues/10227#issuecomment-537600327
Um.. 1E-1, 1E-2, now 1E-3.. it has to stop somewhere.. but anyway.. for
temporary fix I will push that as PR then
Caenorst opened a new pull request #16356: Fix atol for test_preloaded_multi_sgd
URL: https://github.com/apache/incubator-mxnet/pull/16356
## Description ##
Fix atol for test_preloaded_multi_sgd
Following Issue: #16345
## Checklist ##
### Essentials ###
- [x] Changes are
ptrendx commented on a change in pull request #16355: Embedding gradient
operator on GPU
URL: https://github.com/apache/incubator-mxnet/pull/16355#discussion_r330682129
##
File path: src/operator/tensor/indexing_op.cu
##
@@ -545,6 +545,245 @@ void TakeOpForward(const
ChaiBapchya commented on a change in pull request #16356: Fix atol for
test_preloaded_multi_sgd
URL: https://github.com/apache/incubator-mxnet/pull/16356#discussion_r330706625
##
File path: tests/python/gpu/test_operator_gpu.py
##
@@ -396,7 +396,7 @@ def
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 a7d5ac2 Bump the
Caenorst commented on a change in pull request #16356: Fix atol for
test_preloaded_multi_sgd
URL: https://github.com/apache/incubator-mxnet/pull/16356#discussion_r330715406
##
File path: tests/python/gpu/test_operator_gpu.py
##
@@ -396,7 +396,7 @@ def
marcoabreu merged pull request #16336: S3 upload artifacts
URL: https://github.com/apache/incubator-mxnet/pull/16336
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
rondogency commented on issue #16358: CI Flaky Test:
test_operator_gpu.test_np_around
URL:
https://github.com/apache/incubator-mxnet/issues/16358#issuecomment-537665096
@mxnet-label-bot add [Test, Flaky]
This is an
reminisce commented on a change in pull request #15909: [numpy] random.rand
URL: https://github.com/apache/incubator-mxnet/pull/15909#discussion_r330763145
##
File path: python/mxnet/numpy/random.py
##
@@ -20,7 +20,34 @@
from __future__ import absolute_import
from
larroy commented on issue #16326: Mxnet 1.5.0: Crash while training mask-rcnn
with horovod
URL:
https://github.com/apache/incubator-mxnet/issues/16326#issuecomment-537631390
Another one:
```
[1,6]:INFO:root:[Epoch 0 Iteration 100] Set learning rate to 0.004
[1,4]:corrupted
Caenorst commented on issue #16356: Fix atol for test_preloaded_multi_sgd
URL: https://github.com/apache/incubator-mxnet/pull/16356#issuecomment-537631624
> Thanks for the quick fix. @Caenorst
> Can you add "Fixes #PR" that would auto close the issue if/when this PR
gets merged.
>
ChaiBapchya commented on issue #16356: Fix atol for test_preloaded_multi_sgd
URL: https://github.com/apache/incubator-mxnet/pull/16356#issuecomment-537644099
Generally in the description of the PR
This is an automated message
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 e6e2e2e Fix code block formatting in Why MXNet doc page (#16334)
add 480b50c S3 upload artifacts
rondogency opened a new issue #16358: CI Flaky Test:
test_operator_gpu.test_np_around
URL: https://github.com/apache/incubator-mxnet/issues/16358
Centos GPU test_operator_gpu.test_np_around fails on tolerance
should be a flaky test since it is the only occurrence I know
mxnet-label-bot commented on issue #16358: CI Flaky Test:
test_operator_gpu.test_np_around
URL:
https://github.com/apache/incubator-mxnet/issues/16358#issuecomment-537664399
Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some labels so
larroy commented on a change in pull request #15515: [MXNET-978] Higher Order
Gradient Support `arcsin`, `arccos`.
URL: https://github.com/apache/incubator-mxnet/pull/15515#discussion_r330752970
##
File path: src/operator/tensor/elemwise_unary_op_trig.cc
##
@@ -188,7
ptrendx commented on issue #16355: Embedding gradient performance optimization
on GPU
URL: https://github.com/apache/incubator-mxnet/pull/16355#issuecomment-537686935
@sxjscience FYI
This is an automated message from the
samskalicky opened a new issue #16360: [Feature Request] Add ability to treat
warnings as errors
URL: https://github.com/apache/incubator-mxnet/issues/16360
For model development in federated environments, logs may not be shown to
the end users. As a result, some mxnet warning messages
mxnet-label-bot commented on issue #16360: [Feature Request] Add ability to
treat warnings as errors
URL:
https://github.com/apache/incubator-mxnet/issues/16360#issuecomment-537691280
Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some
anirudh2290 merged pull request #16356: Fix atol for test_preloaded_multi_sgd
URL: https://github.com/apache/incubator-mxnet/pull/16356
This is an automated message from the Apache Git Service.
To respond to the message,
sxjscience commented on issue #16357: Numpy Operator Inconsistency
URL:
https://github.com/apache/incubator-mxnet/issues/16357#issuecomment-537699375
duplicate: https://github.com/apache/incubator-mxnet/issues/16048
This is
zachgk opened a new issue #16359: Flaky Scala Nightly Release Profiler
URL: https://github.com/apache/incubator-mxnet/issues/16359
There is a flaky test on the Scala nightly Jenkins pipeline that
occasionally causes it to fail. Sample failure:
```
- Example CI: Test GAN MNIST
mxnet-label-bot commented on issue #16359: Flaky Scala Nightly Release Profiler
URL:
https://github.com/apache/incubator-mxnet/issues/16359#issuecomment-537682561
Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some labels so
that the
ptrendx edited a comment on issue #15589: [Discussion] 1.6.0 Roadmap
URL:
https://github.com/apache/incubator-mxnet/issues/15589#issuecomment-526373840
We have multiple improvements to BERT inference and training speed that we
would like to be part of 1.6 release:
- [x] Softmax
roywei opened a new issue #11774: Gluon BatchNorm beta=False not working
properly
URL: https://github.com/apache/incubator-mxnet/issues/11774
**Description**
I came from #10401
In that issue, there is no reproducible code, so not sure whats the use case
there.
When I was
roywei commented on issue #11774: Gluon BatchNorm beta=False not working
properly
URL:
https://github.com/apache/incubator-mxnet/issues/11774#issuecomment-537688602
I'm re-opening this issue as the fix PR got reverted, and there is a user
reported this does not work in keras-mxnet.
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 480b50c S3 upload artifacts (#16336)
add 8136d49 fix atol for test_preloaded_multi_sgd (#16356)
This is an automated email from the ASF dual-hosted git repository.
apeforest pushed a change to branch benchmark
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
discard 19a5b88 Merge remote-tracking branch 'Kh4L/mrcnn_mask_target_op' into
benchmark
discard e92870a
This is an automated email from the ASF dual-hosted git repository.
apeforest pushed a change to branch benchmark
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from f4a2caa Revert "Merge remote-tracking branch 'Jerryzcn/pr-rcnn' into
benchmark"
add ff52b9e
haojin2 opened a new pull request #16361: boolean_mask_assign operator for
future boolean indexing
URL: https://github.com/apache/incubator-mxnet/pull/16361
## Description ##
A step toward boolean indexing such as:
```Python
a[mask] = val
```
where `mask` is a boolean array
zachgk opened a new pull request #16362: Fix nightly scala pipeline
URL: https://github.com/apache/incubator-mxnet/pull/16362
## Description ##
Fix the Dockerfiles used for the Scala nightly publishing job
@samskalicky @aaronmarkham
## Checklist ##
### Essentials ###
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 d12cb4a Bump the
zixuanweeei commented on issue #16256: [Bug] Failed to evaluate gradient on
samples with train_mode=False
URL:
https://github.com/apache/incubator-mxnet/issues/16256#issuecomment-537745745
@ZhiminPeng Thank you for your trying. It should be noticed that the fix is
just temporal, and it
KellenSunderland commented on issue #15589: [Discussion] 1.6.0 Roadmap
URL:
https://github.com/apache/incubator-mxnet/issues/15589#issuecomment-537797358
TRT is now working from the CPP package. I think to consider it a released
feature we'd want to update documentation and possible
mxnet-label-bot commented on issue #16365: Exception: unknown storage type: -1
URL:
https://github.com/apache/incubator-mxnet/issues/16365#issuecomment-537778507
Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some labels so
that the
jiashu-zhu opened a new issue #16365: Exception: unknown storage type: -1
URL: https://github.com/apache/incubator-mxnet/issues/16365
## Description
I encounter a the exception "Exception: unknown storage type: -1" when I use
my focal loss
## my focal loss
the shape of
KellenSunderland commented on issue #16173: Saving and loading cudNN autotune
and graph optimization
URL:
https://github.com/apache/incubator-mxnet/issues/16173#issuecomment-537796546
I agree this is an issue and really like the approach here @QueensGambit.
TensorRT and Autotuning are
blessalanou commented on issue #16291: Unable to access
https://mxnet.incubator.apache.org/ in China, got 404 response.
URL:
https://github.com/apache/incubator-mxnet/issues/16291#issuecomment-537742505
Thank you guys. https://mxnet.incubator.apache.org works OK.
blessalanou closed issue #16291: Unable to access
https://mxnet.incubator.apache.org/ in China, got 404 response.
URL: https://github.com/apache/incubator-mxnet/issues/16291
This is an automated message from the Apache Git
zixuanweeei commented on issue #16256: [Bug] Failed to evaluate gradient on
samples with train_mode=False
URL:
https://github.com/apache/incubator-mxnet/issues/16256#issuecomment-537748979
@szha Thanks for your reply. I will delve into the concept of `train_mode`.
And thanks for the hint
jiashu-zhu closed issue #16364: ption
URL: https://github.com/apache/incubator-mxnet/issues/16364
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL
mxnet-label-bot commented on issue #16364: ption
URL:
https://github.com/apache/incubator-mxnet/issues/16364#issuecomment-537771701
Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some labels so
that the appropriate MXNet community members
jiashu-zhu opened a new issue #16364: ption
URL: https://github.com/apache/incubator-mxnet/issues/16364
Note: Providing complete information in the most concise form is the best
way to get help. This issue template serves as the checklist for essential
information to most of the technical
samskalicky commented on issue #15921: dynamic custom operator support
URL: https://github.com/apache/incubator-mxnet/pull/15921#issuecomment-537741566
Failing centos-gpu job is failing with Numpy issue #16358 unrelated to this
PR. We've sync'ed with the authors and they confirmed. Assume
samskalicky removed a comment on issue #15921: dynamic custom operator support
URL: https://github.com/apache/incubator-mxnet/pull/15921#issuecomment-537741566
Failing centos-gpu job is failing with Numpy issue #16358 unrelated to this
PR. We've sync'ed with the authors and they confirmed.
apua opened a new pull request #16363: Julia: add API docs back
URL: https://github.com/apache/incubator-mxnet/pull/16363
## Description ##
`ndarray` and `symbolic-node` have been refactored by splitting into a
few files in commits ed8307121 and 36a3cb828, but the corresponding
KellenSunderland commented on issue #14341: Is the mxnet-tensorrt integration
available in C++?
URL:
https://github.com/apache/incubator-mxnet/issues/14341#issuecomment-537763704
The CPP package has been updated to include TensorRT binding ability in
KellenSunderland closed issue #14341: Is the mxnet-tensorrt integration
available in C++?
URL: https://github.com/apache/incubator-mxnet/issues/14341
This is an automated message from the Apache Git Service.
To respond to
90 matches
Mail list logo