This is an automated email from the ASF dual-hosted git repository.
haoj pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 71b6272 use identity_with_cast (#16913)
add 44cd63e [Numpy] Implement numpy operator 'average' (#16720)
haojin2 merged pull request #16720: [Numpy] Implement numpy operator 'average'
URL: https://github.com/apache/incubator-mxnet/pull/16720
This is an automated message from the Apache Git Service.
To respond to the message, ple
This is an automated email from the ASF dual-hosted git repository.
haoj pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 7736bfd [Numpy] add op full_like, c++ impl, fix zeros_like, ones_like
type inference (#16804)
add 71b62
This is an automated email from the ASF dual-hosted git repository.
haoj pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 7736bfd [Numpy] add op full_like, c++ impl, fix zeros_like, ones_like
type inference (#16804)
add 71b62
haojin2 merged pull request #16913: [numpy] add op linalg solve
URL: https://github.com/apache/incubator-mxnet/pull/16913
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
junrushao1994 commented on issue #17018: Replace mxnet_option macro with
standard CMAKE_DEPENDENT_OPTION
URL: https://github.com/apache/incubator-mxnet/pull/17018#issuecomment-563108671
I think I am a very beginner in using cmake. Looks like
`CMAKE_DEPENDENT_OPTION ` provides a native way
perdasilva commented on issue #16966: [CD] dynamic libmxet pipeline fix + small
fixes
URL: https://github.com/apache/incubator-mxnet/pull/16966#issuecomment-563106814
@DickJC123 I've created an issue to track this problem: #17020 - thanks
again for looking into it
perdasilva commented on issue #17020: [CUDA 9.0] NVRTC Compilation failed
URL:
https://github.com/apache/incubator-mxnet/issues/17020#issuecomment-563106604
@DickJC123 I've created a proper issue so we can track this problem - thanks
for looking into it!
--
perdasilva opened a new issue #17020: [CUDA 9.0] NVRTC Compilation failed
URL: https://github.com/apache/incubator-mxnet/issues/17020
## Description
Since #15167, the CD pipeline for CUDA 9.0 have been failing.
### Error Message
Many examples can be taken from the CD pipelin
This is an automated email from the ASF dual-hosted git repository.
haoj pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 3b8fdac skip quantized conv flaky case (#16866)
add 7736bfd [Numpy] add op full_like, c++ impl, fix zer
This is an automated email from the ASF dual-hosted git repository.
haoj pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 3b8fdac skip quantized conv flaky case (#16866)
add 7736bfd [Numpy] add op full_like, c++ impl, fix zer
haojin2 merged pull request #16804: add numpy op full_like, c++ impl, fix
zeros_like, ones_like type infe…
URL: https://github.com/apache/incubator-mxnet/pull/16804
This is an automated message from the Apache Git Service.
T
leezu opened a new pull request #17019: Fix CUDNN detection for CMake build
URL: https://github.com/apache/incubator-mxnet/pull/17019
## Description ##
- Use FindCUDNN.cmake instead of the previous macro
- Previous macro did not detect CUDNN correctly on my system (Deep Learning
AMI
szha commented on issue #17018: Replace mxnet_option macro with standard
CMAKE_DEPENDENT_OPTION
URL: https://github.com/apache/incubator-mxnet/pull/17018#issuecomment-563094643
cc @junrushao1994 who may want to offer suggestions
-
leezu opened a new pull request #17018: Replace mxnet_option macro with
standard CMAKE_DEPENDENT_OPTION
URL: https://github.com/apache/incubator-mxnet/pull/17018
## Description ##
Replace mxnet_option macro with standard CMAKE_DEPENDENT_OPTION.
Using standard language constructs impro
pengzhao-intel commented on issue #16830: CI error in unix gpu
test_quantization_gpu.test_quantized_conv
URL:
https://github.com/apache/incubator-mxnet/issues/16830#issuecomment-563093760
skip the test now and please merge the code.
haojin2 commented on issue #16830: CI error in unix gpu
test_quantization_gpu.test_quantized_conv
URL:
https://github.com/apache/incubator-mxnet/issues/16830#issuecomment-563092812
Happening again:
http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/mxnet-validation%2Funix-cp
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 c3d424b Bump the publis
rondogency commented on issue #17006: [RFC] Custom Operator Part 2
URL:
https://github.com/apache/incubator-mxnet/issues/17006#issuecomment-563085461
Need to include a fix for the test error
https://github.com/apache/incubator-mxnet/pull/15921#pullrequestreview-328686634
-
janelu9 commented on issue #9359: Does gluon's dnn support data format of
libsvm other than mxnet's?
URL:
https://github.com/apache/incubator-mxnet/issues/9359#issuecomment-563085016
I think I can use embedding
This is an au
rondogency commented on a change in pull request #15921: dynamic custom
operator support
URL: https://github.com/apache/incubator-mxnet/pull/15921#discussion_r355280560
##
File path: tests/python/unittest/test_extensions.py
##
@@ -21,15 +21,16 @@
import platform
import u
Tommliu commented on issue #16989: Op_Diagonal [Numpy]
URL: https://github.com/apache/incubator-mxnet/pull/16989#issuecomment-563080401
@haojin2 Diagonal done.
This is an automated message from the Apache Git Service.
To resp
szha commented on issue #17017: USE_BLAS=apple broken on OSX 10.15
URL:
https://github.com/apache/incubator-mxnet/issues/17017#issuecomment-563078396
It looks broken too:
https://github.com/apache/incubator-mxnet/blob/master/cmake/Modules/FindAccelerate.cmake#L27
-
This is an automated email from the ASF dual-hosted git repository.
patriczhao pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from b009864 large tensor faq doc fix (#16953)
add 3b8fdac skip quantized conv flaky case (#16866)
No
leezu commented on issue #17017: USE_BLAS=apple broken on OSX 10.15
URL:
https://github.com/apache/incubator-mxnet/issues/17017#issuecomment-563076929
How about CMake?
This is an automated message from the Apache Git Service.
This is an automated email from the ASF dual-hosted git repository.
patriczhao pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from b009864 large tensor faq doc fix (#16953)
add 3b8fdac skip quantized conv flaky case (#16866)
No
pengzhao-intel merged pull request #16866: skip quantized conv flaky case
URL: https://github.com/apache/incubator-mxnet/pull/16866
This is an automated message from the Apache Git Service.
To respond to the message, please l
szha opened a new issue #17017: USE_BLAS=apple broken on OSX 10.15
URL: https://github.com/apache/incubator-mxnet/issues/17017
## Description
Starting from 10.15, Accelerate/vecLib frameworks are not shipped by default
([macOS frameworks are now thinned for the x86-64
architecture](http
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 d58f6cb Add micro averaging strategy to pearsonr metric (#16878)
add b009864 large tensor faq doc fix
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 d58f6cb Add micro averaging strategy to pearsonr metric (#16878)
add b009864 large tensor faq doc fix
leezu merged pull request #16953: large tensor faq doc fix
URL: https://github.com/apache/incubator-mxnet/pull/16953
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
wkcn commented on a change in pull request #15921: dynamic custom operator
support
URL: https://github.com/apache/incubator-mxnet/pull/15921#discussion_r355268745
##
File path: tests/python/unittest/test_extensions.py
##
@@ -21,15 +21,16 @@
import platform
import unittes
xidulu opened a new pull request #17016: [Numpy] Fix axis=-1 bug in max
URL: https://github.com/apache/incubator-mxnet/pull/17016
## Description ##
Fix
https://github.com/apache/incubator-mxnet/issues/17011
## Checklist ##
### Essentials ###
Please feel free to remove inap
TaoLv commented on a change in pull request #15921: dynamic custom operator
support
URL: https://github.com/apache/incubator-mxnet/pull/15921#discussion_r355262466
##
File path: tests/python/unittest/test_extensions.py
##
@@ -21,15 +21,16 @@
import platform
import unitte
access2rohit commented on issue #16953: large tensor faq doc fix
URL: https://github.com/apache/incubator-mxnet/pull/16953#issuecomment-563059985
@mxnet-label-bot add [pr-awaiting-merge]
This is an automated message from the A
sxjscience commented on issue #16990: [numpy] add op matmul
URL: https://github.com/apache/incubator-mxnet/pull/16990#issuecomment-563047925
I think you may call the BatchedGEMMStrided Kernel directly:
https://devblogs.nvidia.com/cublas-strided-batched-matrix-multiply/
Get Outlook fo
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 251e6f6 Fix NDArrayIter cant pad when size is large (#17001)
add d58f6cb Add micro averaging strategy
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 251e6f6 Fix NDArrayIter cant pad when size is large (#17001)
add d58f6cb Add micro averaging strategy
leezu merged pull request #16878: add micro to pearsonr
URL: https://github.com/apache/incubator-mxnet/pull/16878
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub an
kice opened a new pull request #17015: optimize onnx ops import
URL: https://github.com/apache/incubator-mxnet/pull/17015
Set padding for convolution if is symmetric
Add 'CRD' support for depthtospace
CRD mode is the default mode for PixelShuffle of pytorch when export with
opset
xidulu edited a comment on issue #17011: [Numpy] additional default parameter
for mxnet.numpy.max: axis=-1
URL:
https://github.com/apache/incubator-mxnet/issues/17011#issuecomment-562963658
Looks like the problem occurs in this line
https://github.com/apache/incubator-mxnet/blob/9b25d
leezu commented on issue #17012: Upgrade 3rdparty/openmp to release_90 version
URL: https://github.com/apache/incubator-mxnet/pull/17012#issuecomment-563034941
Force push to retrigger CI. Failed with Github API error for 5/11 jobs
(other 6 passed):
```
Pull request #17012 opened
JiangZhaoh commented on a change in pull request #16990: [numpy] add op matmul
URL: https://github.com/apache/incubator-mxnet/pull/16990#discussion_r355237851
##
File path: src/operator/numpy/np_matmul_op-inl.h
##
@@ -0,0 +1,356 @@
+/*
+ * Licensed to the Apache Software Fo
JiangZhaoh commented on a change in pull request #16990: [numpy] add op matmul
URL: https://github.com/apache/incubator-mxnet/pull/16990#discussion_r355237789
##
File path: src/operator/numpy/np_matmul_op-inl.h
##
@@ -0,0 +1,356 @@
+/*
+ * Licensed to the Apache Software Fo
JiangZhaoh commented on a change in pull request #16990: [numpy] add op matmul
URL: https://github.com/apache/incubator-mxnet/pull/16990#discussion_r355237789
##
File path: src/operator/numpy/np_matmul_op-inl.h
##
@@ -0,0 +1,356 @@
+/*
+ * Licensed to the Apache Software Fo
wkcn edited a comment on issue #16994: DataLoader: Missing batch_size dim when
num_workers > 0
URL:
https://github.com/apache/incubator-mxnet/issues/16994#issuecomment-563028046
Hi @cuibuaa , please use the latest version of MXNet. It is a bug in the old
version.
https://github.com
wkcn commented on issue #16994: DataLoader: Missing batch_size dim when
num_workers > 0
URL:
https://github.com/apache/incubator-mxnet/issues/16994#issuecomment-563028046
Hi @cuibuaa , please use the latest version of MXNet. It is a bug in the old
version.
---
MediosZ closed issue #16961: Unable to hardcode the OMP_NUM_THREADS into program
URL: https://github.com/apache/incubator-mxnet/issues/16961
This is an automated message from the Apache Git Service.
To respond to the message,
MediosZ commented on issue #16961: Unable to hardcode the OMP_NUM_THREADS into
program
URL:
https://github.com/apache/incubator-mxnet/issues/16961#issuecomment-563027547
Yes, the threads are created by openblas, and I rebuild the openblas setting
the number of threads to one, then problem
kice edited a comment on issue #16590: import_onnx.py parser for onnx opset >=
9 has bug
URL:
https://github.com/apache/incubator-mxnet/issues/16590#issuecomment-563018787
UPDATE:
I think mxnet should look at the initializers first. I checked the doc for
pytorch export onnx model,
leezu commented on issue #17012: Upgrade 3rdparty/openmp to release_90 version
URL: https://github.com/apache/incubator-mxnet/pull/17012#issuecomment-563020783
Please see the last line of the PR description above. Given that the assert
is still there and it is not failing anymore I think we
kice edited a comment on issue #16590: import_onnx.py parser for onnx opset >=
9 has bug
URL:
https://github.com/apache/incubator-mxnet/issues/16590#issuecomment-563018787
In my case, I use pytorch to export onnx with opset less than 8, it still
cannot import the model to mxnet.
`
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 66e55d5 Bump the publis
kice commented on issue #16590: import_onnx.py parser for onnx opset >= 9 has
bug
URL:
https://github.com/apache/incubator-mxnet/issues/16590#issuecomment-563018787
In my case, I use pytorch to export onnx with opset less than 8, it still
cannot import the model to mxnet.
```
>
This is an automated email from the ASF dual-hosted git repository.
roywei 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 251e6f6 Fix NDArrayIter cant pad when
roywei closed issue #16996: mx.io.NDArrayIter cant pad when size is large
URL: https://github.com/apache/incubator-mxnet/issues/16996
This is an automated message from the Apache Git Service.
To respond to the message, please
roywei merged pull request #17001: Fix NDArrayIter cant pad when size is large
URL: https://github.com/apache/incubator-mxnet/pull/17001
This is an automated message from the Apache Git Service.
To respond to the message, ple
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 1d6ba33 Bump the publis
vexilligera opened a new pull request #17014: [NumPy][WIP] Add NumPy support
for norm
URL: https://github.com/apache/incubator-mxnet/pull/17014
## Description ##
implements np.linalg.norm
## Checklist ##
### Essentials ###
Please feel free to remove inapplicable items for yo
xidulu commented on issue #17011: [Numpy] additional default parameter for
mxnet.numpy.max: axis=-1
URL:
https://github.com/apache/incubator-mxnet/issues/17011#issuecomment-562963658
Looks like the problem occurs in this line
https://github.com/apache/incubator-mxnet/blob/9b25db05b4de
cjolivier01 commented on issue #14979: [BUG] Using a package with MKL and GPU
versions, using python to open a new process will cause an error
URL:
https://github.com/apache/incubator-mxnet/issues/14979#issuecomment-562952406
what is the source file and line number of that crash in libmxne
artor1os commented on issue #16720: [Numpy] Implement numpy operator 'average'
URL: https://github.com/apache/incubator-mxnet/pull/16720#issuecomment-562952064
@haojin2 ready to be merged
This is an automated message from 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 6d26f64 Bump the publis
fatherMatrix opened a new 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
## Description
I like MXNet very much and I'd like to use tools about mxnet community in my
daily work. But I find
TaoLv commented on issue #16891: Upgrading MKLDNN to 1.0 causes performance
regression.
URL:
https://github.com/apache/incubator-mxnet/issues/16891#issuecomment-562940173
@ChaiBapchya The file is used to build mxnet-mkl pip package. If you want to
change the configurations, I think you ne
leezu opened a new pull request #17012: Upgrade 3rdparty/openmp to release_90
version
URL: https://github.com/apache/incubator-mxnet/pull/17012
## Description ##
Fixes https://github.com/apache/incubator-mxnet/issues/10856
## Checklist ##
### Essentials ###
Please feel free
leezu edited a comment on issue #14979: [BUG] Using a package with MKL and GPU
versions, using python to open a new process will cause an error
URL:
https://github.com/apache/incubator-mxnet/issues/14979#issuecomment-562926756
There are currently two hypotheses about the root cause of this
leezu commented on issue #14979: [BUG] Using a package with MKL and GPU
versions, using python to open a new process will cause an error
URL:
https://github.com/apache/incubator-mxnet/issues/14979#issuecomment-562926756
There are currently two hypotheses about the root cause of this error
Alicia1529 opened a new issue #17011: [Numpy] additional default parameter for
mxnet.numpy.max: axis=-1
URL: https://github.com/apache/incubator-mxnet/issues/17011
Max operator in official numpy can support argument axis=-1 and return max
value from its last dimension. Wonder if deepnumpy
z01nl1o02 commented on issue #15957: error: call of overloaded is ambiguous
URL:
https://github.com/apache/incubator-mxnet/issues/15957#issuecomment-562922365
add copy constructor may fix it
` NodeEntry(const struct NodeEntry& entry):
node(std::move(entry.node)),
index(ent
70 matches
Mail list logo