haojin2 opened a new pull request #17099: [Numpy] Implement atleast_1d/2d/3d
URL: https://github.com/apache/incubator-mxnet/pull/17099
## Description ##
Porting the changes in #16914
## Checklist ##
### Essentials ###
Please feel free to remove inapplicable items for your
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 7c19a9d Bump the
leezu opened a new pull request #17098: Disable OpenMP offloading support for
3rdparty/openmp
URL: https://github.com/apache/incubator-mxnet/pull/17098
## Description ##
OpenMP offloading was introduced some time during the past two years and is
enabled by default. With upgrading
leezu closed pull request #17095: [DONOTMERGE] Test
https://github.com/dmlc/dmlc-core/pull/585
URL: https://github.com/apache/incubator-mxnet/pull/17095
This is an automated message from the Apache Git Service.
To respond
haojin2 commented on a change in pull request #17087: OP All/Any [Numpy]
URL: https://github.com/apache/incubator-mxnet/pull/17087#discussion_r358618728
##
File path: python/mxnet/_numpy_op_doc.py
##
@@ -20,6 +20,107 @@
"""Doc placeholder for numpy ops with prefix _np."""
haojin2 commented on a change in pull request #17087: OP All/Any [Numpy]
URL: https://github.com/apache/incubator-mxnet/pull/17087#discussion_r358618581
##
File path: python/mxnet/_numpy_op_doc.py
##
@@ -20,6 +20,107 @@
"""Doc placeholder for numpy ops with prefix _np."""
iblis17 commented on issue #17092: ERROR: LoadError: LoadError: IOError: open:
permission denied (EACCES) (Julia, macOS)
URL:
https://github.com/apache/incubator-mxnet/issues/17092#issuecomment-566397610
What is the Julia `versioninfo()` on the second machine?
reminisce opened a new issue #17097: [mxnet 2.0][item 10.1] MXNet Imperative Op
Invocation Overhead
URL: https://github.com/apache/incubator-mxnet/issues/17097
MXNet imperative operator invocation overhead is as large as 30-60us, which
is significant compared to the official NumPy
ZhennanQin commented on issue #17094: Add silent option to quantization script
URL: https://github.com/apache/incubator-mxnet/pull/17094#issuecomment-566394537
@zhreshold Code is refactored to be quiet by default. Only when valid logger
is passed, then log will be printed. So just make
ZhennanQin commented on a change in pull request #17094: Add silent option to
quantization script
URL: https://github.com/apache/incubator-mxnet/pull/17094#discussion_r358608515
##
File path: src/operator/quantization/quantize_graph_pass.cc
##
@@ -275,12 +275,15 @@ Graph
ZhennanQin commented on a change in pull request #17094: Add silent option to
quantization script
URL: https://github.com/apache/incubator-mxnet/pull/17094#discussion_r358608515
##
File path: src/operator/quantization/quantize_graph_pass.cc
##
@@ -275,12 +275,15 @@ Graph
arcadiaphy commented on a change in pull request #16502: Add im2col and col2im
operator
URL: https://github.com/apache/incubator-mxnet/pull/16502#discussion_r358605760
##
File path: src/operator/nn/im2col-inl.h
##
@@ -0,0 +1,247 @@
+/*
+ * Licensed to the Apache Software
reminisce opened a new issue #17096: [mxnet 2.0][item 1.4] NumPy Extension
Operator Reorganization and Renaming
URL: https://github.com/apache/incubator-mxnet/issues/17096
We introduced modules `np` and `npx` for registering NumPy and the other
operators, respectively in MXNet. We need to
zhreshold commented on a change in pull request #17094: Add silent option to
quantization script
URL: https://github.com/apache/incubator-mxnet/pull/17094#discussion_r358602453
##
File path: src/operator/quantization/quantize_graph_pass.cc
##
@@ -275,12 +275,15 @@ Graph
zhreshold commented on a change in pull request #17094: Add silent option to
quantization script
URL: https://github.com/apache/incubator-mxnet/pull/17094#discussion_r358602638
##
File path: src/operator/quantization/quantize_graph_pass.cc
##
@@ -516,15 +520,20 @@ static
zhreshold commented on a change in pull request #17094: Add silent option to
quantization script
URL: https://github.com/apache/incubator-mxnet/pull/17094#discussion_r358602495
##
File path: src/operator/quantization/quantize_graph_pass.cc
##
@@ -275,12 +275,15 @@ Graph
zhreshold commented on a change in pull request #17094: Add silent option to
quantization script
URL: https://github.com/apache/incubator-mxnet/pull/17094#discussion_r358602579
##
File path: src/operator/quantization/quantize_graph_pass.cc
##
@@ -516,15 +520,20 @@ static
leezu edited a comment on issue #17063: Fix armv7 build by updating the base
container and pin it
URL: https://github.com/apache/incubator-mxnet/pull/17063#issuecomment-566381845
Regarding Hotfix, `armv7` is not sufficient, we need to update all arches.
It's straightforward to update, ie.
leezu commented on issue #17063: Fix armv7 build by updating the base container
and pin it
URL: https://github.com/apache/incubator-mxnet/pull/17063#issuecomment-566381845
Regarding Hotfix, `armv7` is not sufficient, we need to update all arches.
It's straightforward to update, ie.
leezu commented on issue #17063: Fix armv7 build by updating the base container
and pin it
URL: https://github.com/apache/incubator-mxnet/pull/17063#issuecomment-566380962
Maybe consensus is the wrong word. Essentially I'm not clear what's the
advantage of using `mxnetci/dockcross-X`
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 814be59 Add #include needed for waitpid (#17078)
add f86a8d1 [API] unified API for custom kvstores
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 814be59 Add #include needed for waitpid (#17078)
add f86a8d1 [API] unified API for custom kvstores
eric-haibin-lin merged pull request #17010: [API] unified API for custom
kvstores
URL: https://github.com/apache/incubator-mxnet/pull/17010
This is an automated message from the Apache Git Service.
To respond to the
larroy edited a comment on issue #17063: Fix armv7 build by updating the base
container and pin it
URL: https://github.com/apache/incubator-mxnet/pull/17063#issuecomment-566377729
@leezu how do you suggest to proceed to reach consensus? Also wouldn't it be
good to have a hotfix first for
larroy commented on issue #17063: Fix armv7 build by updating the base
container and pin it
URL: https://github.com/apache/incubator-mxnet/pull/17063#issuecomment-566377729
@leezu how do you suggest to proceed to reach consensus?
leezu opened a new pull request #17095: [DONOTMERGE] Test
https://github.com/dmlc/dmlc-core/pull/585
URL: https://github.com/apache/incubator-mxnet/pull/17095
## Description ##
https://github.com/dmlc/dmlc-core/pull/585
Test with MXNet CI suggested by @larroy
ZhennanQin opened a new pull request #17094: Add silent option to quantization
script
URL: https://github.com/apache/incubator-mxnet/pull/17094
## Description ##
As title.
@pengzhao-intel @zhreshold @hetong007
## Checklist ##
### Essentials ###
Please feel free to
gengyanlei commented on issue #17085: Suggest to add RandomApply in gluon's
transforms
URL:
https://github.com/apache/incubator-mxnet/issues/17085#issuecomment-566346371
比如color_transform，它会让图像值分布变化，但是我想让网络也学习到原始图像值分布，那么就需要一个概率值，对这个tranform发生概率进行设置！
For example, color_transform will
gengyanlei commented on issue #17085: Suggest to add RandomApply in gluon's
transforms
URL:
https://github.com/apache/incubator-mxnet/issues/17085#issuecomment-566345518
https://pytorch.org/docs/1.1.0/torchvision/transforms.html#transforms-on-pil-image
Apply randomly a list of
wkcn edited a comment on issue #17085: Suggest to add RandomApply in gluon's
transforms
URL:
https://github.com/apache/incubator-mxnet/issues/17085#issuecomment-566337339
There are multiple transform methods(lists), and choose one randomly.
wkcn commented on issue #17085: Suggest to add RandomApply in gluon's transforms
URL:
https://github.com/apache/incubator-mxnet/issues/17085#issuecomment-566337339
There are multiple transform methods, and choose one randomly.
This is an automated email from the ASF dual-hosted git repository.
zhasheng pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 18c9a69 Install current OpenCV (#17090)
add 814be59 Add #include needed for waitpid (#17078)
No
szha merged pull request #17078: Add #include needed for waitpid
URL: https://github.com/apache/incubator-mxnet/pull/17078
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.
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 30584a9 Bump the
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 4912465 Update pypi_publish.py to disable nighlty build upload to
Pypi (#17082)
add 18c9a69 Install
leezu closed issue #17061: Built from source uses old version of opencv
URL: https://github.com/apache/incubator-mxnet/issues/17061
This is an automated message from the Apache Git Service.
To respond to the message, please
leezu merged pull request #17090: Install current OpenCV
URL: https://github.com/apache/incubator-mxnet/pull/17090
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
djaym7 commented on issue #16341: [New Op] Add deformable conv v2
URL: https://github.com/apache/incubator-mxnet/pull/16341#issuecomment-566308144
djaym7 edited a comment on issue #16341: [New Op] Add deformable conv v2
URL: https://github.com/apache/incubator-mxnet/pull/16341#issuecomment-565660972
Is anyone working on CuDNN support for DCN v1 and v2 ? @zhreshold
This
This is an automated email from the ASF dual-hosted git repository.
zhasheng pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from db943e9 [BugFix] fix filter channel calculation in
ModulatedDeformableConvV2 (#17070)
add 4912465
This is an automated email from the ASF dual-hosted git repository.
zhasheng pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from db943e9 [BugFix] fix filter channel calculation in
ModulatedDeformableConvV2 (#17070)
add 4912465
szha merged pull request #17082: Update pypi_publish.py to disable nighlty
build upload to Pypi
URL: https://github.com/apache/incubator-mxnet/pull/17082
This is an automated message from the Apache Git Service.
To respond
szha commented on issue #17085: Suggest to add RandomApply in gluon's transforms
URL:
https://github.com/apache/incubator-mxnet/issues/17085#issuecomment-566274002
what's the definition of RandomApply?
This is an automated
larroy commented on a change in pull request #17063: Fix armv7 build by
updating the base container and pin it
URL: https://github.com/apache/incubator-mxnet/pull/17063#discussion_r358387653
##
File path: ci/docker/Dockerfile.build.armv7
##
@@ -18,7 +18,10 @@
#
#
swe-ml opened a new issue #17093: as_in_context support for Symbol objects
URL: https://github.com/apache/incubator-mxnet/issues/17093
## Description
Symbol objects currently have multiple unsupported operations raising
NotImplementedforSymbol errors. I would love to see as_in_context
TobyRoseman commented on issue #12266: Python 3.7 Support
URL:
https://github.com/apache/incubator-mxnet/issues/12266#issuecomment-566209007
No problem @yuriploc. We're tracking that issues
[here](https://github.com/apple/turicreate/issues/788) and getting very close
to releasing a fix.
apeforest commented on a change in pull request #17010: [API] unified API for
custom kvstores
URL: https://github.com/apache/incubator-mxnet/pull/17010#discussion_r358422469
##
File path: python/mxnet/kvstore/base.py
##
@@ -0,0 +1,452 @@
+# Licensed to the Apache Software
apeforest commented on a change in pull request #17010: [API] unified API for
custom kvstores
URL: https://github.com/apache/incubator-mxnet/pull/17010#discussion_r358422469
##
File path: python/mxnet/kvstore/base.py
##
@@ -0,0 +1,452 @@
+# Licensed to the Apache Software
This is an automated email from the ASF dual-hosted git repository.
zhreshold pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 52c9a45 [MKLDNN] enable MaxPooling with full pooling convention
(#16860)
add db943e9 [BugFix]
This is an automated email from the ASF dual-hosted git repository.
zhreshold pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 52c9a45 [MKLDNN] enable MaxPooling with full pooling convention
(#16860)
add db943e9 [BugFix]
zhreshold merged pull request #17070: [BugFix] fix filter channel calculation
in ModulatedDeformableConvV2
URL: https://github.com/apache/incubator-mxnet/pull/17070
This is an automated message from the Apache Git Service.
yuriploc commented on issue #12266: Python 3.7 Support
URL:
https://github.com/apache/incubator-mxnet/issues/12266#issuecomment-566200706
My apologies, @TobyRoseman. I'm eager for a new version of `turicreate` to
support 3.7 and did not read the whole list correctly.
apeforest commented on a change in pull request #17010: [API] unified API for
custom kvstores
URL: https://github.com/apache/incubator-mxnet/pull/17010#discussion_r358408648
##
File path: python/mxnet/model.py
##
@@ -122,10 +130,10 @@ def _initialize_kvstore(kvstore,
apeforest commented on a change in pull request #17010: [API] unified API for
custom kvstores
URL: https://github.com/apache/incubator-mxnet/pull/17010#discussion_r358407839
##
File path: python/mxnet/kvstore/kvstore.py
##
@@ -327,39 +344,37 @@ def pushpull(self, key,
apeforest commented on a change in pull request #17010: [API] unified API for
custom kvstores
URL: https://github.com/apache/incubator-mxnet/pull/17010#discussion_r358406887
##
File path: python/mxnet/kvstore/base.py
##
@@ -0,0 +1,452 @@
+# Licensed to the Apache Software
truedichotomy opened a new issue #17092: ERROR: LoadError: LoadError: IOError:
open: permission denied (EACCES) (Julia, macOS)
URL: https://github.com/apache/incubator-mxnet/issues/17092
## Description
I'm getting LoadError in Julia trying to precompile MXNet on macOS 10.15.2.
apeforest commented on a change in pull request #17010: [API] unified API for
custom kvstores
URL: https://github.com/apache/incubator-mxnet/pull/17010#discussion_r358403167
##
File path: python/mxnet/gluon/trainer.py
##
@@ -354,14 +366,30 @@ def allreduce_grads(self):
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 b3d44ea Bump the
larroy removed a comment on issue #17063: Fix armv7 build by updating the base
container and pin it
URL: https://github.com/apache/incubator-mxnet/pull/17063#issuecomment-566178791
@mxnet-label-bot add pr-awaiting-review
larroy commented on issue #17063: Fix armv7 build by updating the base
container and pin it
URL: https://github.com/apache/incubator-mxnet/pull/17063#issuecomment-566178923
@mxnet-label-bot add [pr-awaiting-review]
This is
larroy commented on issue #17063: Fix armv7 build by updating the base
container and pin it
URL: https://github.com/apache/incubator-mxnet/pull/17063#issuecomment-566178791
@mxnet-label-bot add pr-awaiting-review
This is an
larroy commented on a change in pull request #17063: Fix armv7 build by
updating the base container and pin it
URL: https://github.com/apache/incubator-mxnet/pull/17063#discussion_r358387653
##
File path: ci/docker/Dockerfile.build.armv7
##
@@ -18,7 +18,10 @@
#
#
TobyRoseman commented on issue #12266: Python 3.7 Support
URL:
https://github.com/apache/incubator-mxnet/issues/12266#issuecomment-566176300
@yuriploc - The `mxnet-1.5.1.post0-py2.py3-none-manylinux1_x86_64.whl` file
from the link you gave is for Linux.
yuriploc edited a comment on issue #12266: Python 3.7 Support
URL:
https://github.com/apache/incubator-mxnet/issues/12266#issuecomment-566167110
@TobyRoseman This should be reopen, as it looks like MXNet still doesn't
support Python 3.7 on Linux: https://pypi.org/project/mxnet/#files
yuriploc commented on issue #12266: Python 3.7 Support
URL:
https://github.com/apache/incubator-mxnet/issues/12266#issuecomment-566167110
@TobyRoseman This should be reopen, as it looks like MXNet still doesn't
support Python 3.7: https://pypi.org/project/mxnet/#files
marcoabreu commented on a change in pull request #17063: Fix armv7 build by
updating the base container and pin it
URL: https://github.com/apache/incubator-mxnet/pull/17063#discussion_r358307698
##
File path: ci/docker/Dockerfile.build.armv7
##
@@ -18,7 +18,10 @@
#
#
leezu commented on issue #17086: [MKLDNN] RNN Op gradient computation is broken
URL:
https://github.com/apache/incubator-mxnet/issues/17086#issuecomment-566102608
Yes, `MXNET_USE_MKLDNN_RNN=0` on 52c9a45 is a workaround for the issue.
zixuanweeei edited a comment on issue #17086: [MKLDNN] RNN Op gradient
computation is broken
URL:
https://github.com/apache/incubator-mxnet/issues/17086#issuecomment-566089968
> @zixuanweeei I previously confirmed the issue still happens with
zixuanweeei commented on issue #17086: [MKLDNN] RNN Op gradient computation is
broken
URL:
https://github.com/apache/incubator-mxnet/issues/17086#issuecomment-566089968
> @zixuanweeei I previously confirmed the issue still happens with
leezu commented on issue #17089: can't install mxnet through pip install command
URL:
https://github.com/apache/incubator-mxnet/issues/17089#issuecomment-566089144
In particular, you can download the file here:
leezu commented on issue #17089: can't install mxnet through pip install command
URL:
https://github.com/apache/incubator-mxnet/issues/17089#issuecomment-566088717
There seem to be frequent problems with the tsinghua mirror. Another user
reported similar problems earlier this month.
leezu closed issue #17089: can't install mxnet through pip install command
URL: https://github.com/apache/incubator-mxnet/issues/17089
This is an automated message from the Apache Git Service.
To respond to the message,
leezu commented on issue #17086: [MKLDNN] RNN Op gradient computation is broken
URL:
https://github.com/apache/incubator-mxnet/issues/17086#issuecomment-566087906
@zixuanweeei I previously confirmed the issue still happens with
897f4fae075c5261958de1a07cded0b5b3058a7a
TaoLv opened a new pull request #17091: broadcast_axis optimization
URL: https://github.com/apache/incubator-mxnet/pull/17091
## Description ##
Improve the performance of single axis broadcasting.
## Checklist ##
### Essentials ###
Please feel free to remove inapplicable
xidulu commented on issue #17088: HuberLoss() TypeError
URL:
https://github.com/apache/incubator-mxnet/issues/17088#issuecomment-566087653
Unfortunately, only L1 and L2 loss are supported by DeepNumpy(MXNet numpy
ndarray) currently. It's actually pretty easy to fix, but I am not sure if
LukasAdamowicz opened a new pull request #17090: Install current OpenCV
URL: https://github.com/apache/incubator-mxnet/pull/17090
Install the latest version of OpenCV instead of OpenCV 3, which causes
problems during installation.
## Description ##
Update MacOSX build from source
LittleFatHero opened a new issue #17089: can't install mxnet through pip
install command
URL: https://github.com/apache/incubator-mxnet/issues/17089
os: ubuntu 18.04
python: 3.6
problem : When I install mxnet-cu101 using the following command:
```
pip install mxnet-cu101
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 71f02e3 Bump the
GPoloVera opened a new issue #17088: HuberLoss() ValueRror
URL: https://github.com/apache/incubator-mxnet/issues/17088
## Description
Using the **HuberLoss()** (with or without parameters) from the module loss
raise a **ValueError** exception with the message. Using it directly, with a
zixuanweeei commented on issue #17086: [MKLDNN] RNN Op gradient computation is
broken
URL:
https://github.com/apache/incubator-mxnet/issues/17086#issuecomment-566022563
@liuzh91 I can reproduce the result. The gradients of bias in LSTM are very
large as you mentioned, and the gradient of
romeolandry commented on issue #16967: Cannot import [yolo3]() model from
onnx/model zoo
URL:
https://github.com/apache/incubator-mxnet/issues/16967#issuecomment-566005773
Thank you for the answer!
I checked the both repository.
what i wont to do is to import a .onnx files and
perdasilva commented on issue #17074: Static Build and CD for
mxnet-cu102/mxnet-cu102mkl
URL: https://github.com/apache/incubator-mxnet/pull/17074#issuecomment-565990739
We also should add the ci/docker/Dockerfile.build.ubuntu_gpu_cu102 - for
testing and building the dynamic libmxnet ^^
zixuanweeei commented on issue #17086: [MKLDNN] RNN Op gradient computation is
broken
URL:
https://github.com/apache/incubator-mxnet/issues/17086#issuecomment-565985553
Thanks for reporting this issue. We will take a look.
Tommliu opened a new pull request #17087: OP All/Any [Numpy]
URL: https://github.com/apache/incubator-mxnet/pull/17087
## Description ##
Numpy operator All/Any
## Checklist ##
### Essentials ###
Please feel free to remove inapplicable items for your PR.
- [ ] The PR title
liuzh91 opened a new issue #17086: [MKLDNN] Gradient computation is broken for
source built MXNET
URL: https://github.com/apache/incubator-mxnet/issues/17086
## Description
Gradient computation on CPU is broken for the source built mxnet.
I was running a language model training
xidulu commented on a change in pull request #17084: [numpy] add op median
URL: https://github.com/apache/incubator-mxnet/pull/17084#discussion_r358095673
##
File path: python/mxnet/numpy/multiarray.py
##
@@ -5340,6 +5340,59 @@ def swapaxes(a, axis1, axis2):
return
xidulu commented on a change in pull request #17084: [numpy] add op median
URL: https://github.com/apache/incubator-mxnet/pull/17084#discussion_r358095673
##
File path: python/mxnet/numpy/multiarray.py
##
@@ -5340,6 +5340,59 @@ def swapaxes(a, axis1, axis2):
return
xidulu commented on a change in pull request #17084: [numpy] add op median
URL: https://github.com/apache/incubator-mxnet/pull/17084#discussion_r358095426
##
File path: src/operator/numpy/np_broadcast_reduce_op_value.cc
##
@@ -320,6 +321,37 @@
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 897f4fa [MKLDNN] mkldnn RNN operator enhancement (#17075)
add 52c9a45 [MKLDNN] enable MaxPooling
pengzhao-intel merged pull request #16860: [MKLDNN] enable MaxPooling with full
pooling convention
URL: https://github.com/apache/incubator-mxnet/pull/16860
This is an automated message from the Apache Git Service.
To
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 843daf5 adding stacktrace in Jenkinsfile_utils.groovy to inspect
Python2 failure cause in CI (#17065)
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 843daf5 adding stacktrace in Jenkinsfile_utils.groovy to inspect
Python2 failure cause in CI (#17065)
pengzhao-intel merged pull request #17075: [MKLDNN] mkldnn RNN operator
enhancement
URL: https://github.com/apache/incubator-mxnet/pull/17075
This is an automated message from the Apache Git Service.
To respond to the
93 matches
Mail list logo
