TaoLv commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549250395
Cannot reproduce the compile errors. Attached the cmake/make log. Could you
please help to check? @cjolivier01
[cmake_20191104_14351
thomelane opened a new pull request #16719: Updated landing page logos (adding
Dely)
URL: https://github.com/apache/incubator-mxnet/pull/16719
## Description ##
Updated logos PNG on landing page of website
Dely (https://www.dely.jp/) are happy for their logo to be added to this.
haojin2 commented on a change in pull request #16717: [Numpy] add op nan_to_num
URL: https://github.com/apache/incubator-mxnet/pull/16717#discussion_r341919822
##
File path: tests/python/unittest/test_numpy_op.py
##
@@ -4164,6 +4164,108 @@ def hybrid_forward(self, F, x, *ar
haojin2 commented on a change in pull request #16717: [Numpy] add op nan_to_num
URL: https://github.com/apache/incubator-mxnet/pull/16717#discussion_r341919822
##
File path: tests/python/unittest/test_numpy_op.py
##
@@ -4164,6 +4164,108 @@ def hybrid_forward(self, F, x, *ar
haojin2 commented on a change in pull request #16717: [Numpy] add op nan_to_num
URL: https://github.com/apache/incubator-mxnet/pull/16717#discussion_r341919740
##
File path: tests/python/unittest/test_numpy_op.py
##
@@ -4164,6 +4164,108 @@ def hybrid_forward(self, F, x, *ar
haojin2 commented on a change in pull request #16717: [Numpy] add op nan_to_num
URL: https://github.com/apache/incubator-mxnet/pull/16717#discussion_r341919625
##
File path: src/operator/tensor/elemwise_unary_op.h
##
@@ -660,6 +661,131 @@ void AroundOpForward(const nnvm::No
haojin2 commented on a change in pull request #16717: [Numpy] add op nan_to_num
URL: https://github.com/apache/incubator-mxnet/pull/16717#discussion_r341919553
##
File path: src/operator/tensor/elemwise_unary_op.h
##
@@ -660,6 +661,131 @@ void AroundOpForward(const nnvm::No
haojin2 commented on a change in pull request #16717: [Numpy] add op nan_to_num
URL: https://github.com/apache/incubator-mxnet/pull/16717#discussion_r341919452
##
File path: src/operator/tensor/elemwise_unary_op.h
##
@@ -660,6 +661,131 @@ void AroundOpForward(const nnvm::No
hubutui commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549241156
@cjolivier01 According to
https://github.com/apache/incubator-mxnet/blob/master/docs/python_docs/python/tutorials/performance/backend/m
szha commented on issue #16685: Memory planner doesn't respect 'output
independence'. More optimizations possible.
URL:
https://github.com/apache/incubator-mxnet/issues/16685#issuecomment-549240646
yes this needs fixing. are you working on it @DickJC123?
-
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 d16cfee Bump the publis
szha commented on issue #16591: Module.predict() produces only one output
meanwhile Module.forward() and then Module.get_outputs() creates multiple (as
it should)
URL:
https://github.com/apache/incubator-mxnet/issues/16591#issuecomment-549236786
The community is moving away from Module an
szha closed issue #16256: [Bug] Failed to evaluate gradient on samples with
train_mode=False
URL: https://github.com/apache/incubator-mxnet/issues/16256
This is an automated message from the Apache Git Service.
To respond to
szha closed pull request #15905: [Numpy] Basic indexing in symbolic interface
URL: https://github.com/apache/incubator-mxnet/pull/15905
This is an automated message from the Apache Git Service.
To respond to the message, plea
szha commented on issue #15905: [Numpy] Basic indexing in symbolic interface
URL: https://github.com/apache/incubator-mxnet/pull/15905#issuecomment-549233442
closing as inactive duplicate of #16621
This is an automated messag
szha commented on a change in pull request #16716: [Numpy][WIP] Fix
collect_params().zero_grad() in gluon numpy interface
URL: https://github.com/apache/incubator-mxnet/pull/16716#discussion_r341911132
##
File path: python/mxnet/gluon/parameter.py
##
@@ -904,7 +904,11 @@ d
sxjscience commented on a change in pull request #16716: [Numpy][WIP] Fix
collect_params().zero_grad() in gluon numpy interface
URL: https://github.com/apache/incubator-mxnet/pull/16716#discussion_r341909957
##
File path: python/mxnet/gluon/parameter.py
##
@@ -904,7 +904,1
szha commented on a change in pull request #16716: [Numpy][WIP] Fix
collect_params().zero_grad() in gluon numpy interface
URL: https://github.com/apache/incubator-mxnet/pull/16716#discussion_r341909578
##
File path: python/mxnet/gluon/parameter.py
##
@@ -904,7 +904,11 @@ d
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 e139442 support mixed-precision true_divide (#16711)
add 94aab39 [Quantization] Enhance gluon qua
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 e139442 support mixed-precision true_divide (#16711)
add 94aab39 [Quantization] Enhance gluon qua
pengzhao-intel commented on issue #16710: [MKLDNN] support mkldnn gelu
URL: https://github.com/apache/incubator-mxnet/pull/16710#issuecomment-549228581
@TaoLv please take a review again
This is an automated message from the Ap
pengzhao-intel merged pull request #16695: [Quantization] Enhance gluon
quantization API
URL: https://github.com/apache/incubator-mxnet/pull/16695
This is an automated message from the Apache Git Service.
To respond to the m
TaoLv commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549227999
Regrading omp in mkldnn, we removed iomp5 becaused of license issue. We're
going to support omp selection through make/cmake command line
TaoLv commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549227825
Thank you for the investigation, @cjolivier01. We will fix that.
This is
cjolivier01 commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549227708
I am guessing if it’s not failing for you it’s because it can’t find MKL
anyway. on my system, I suppose it can find it.
-
cjolivier01 edited a comment on issue #11417: libomp.so dependency (need REAL
fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549227415
> Could you please also share your cmake command line?
I can get it to build by taking out the USEMKL if found and th
cjolivier01 commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549227363
From CMakeCache:
//CXX compiler libraries for OpenMP parallelization
OpenMP_CXX_LIB_NAMES:STRING=gomp;pthread
---
cjolivier01 commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549227415
> Could you please also share your cmake command line?
I can get it to build by taking out the USEMKL if found and then expli
cjolivier01 commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549227157
btw there's a problem with the mkldnn build. It pulls in libgomp always:
```bash
[coolivie@alien-51:~/src/mxnet/bld (master)]
cjolivier01 commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549226887
from mxnet/bld:
```cmake -DUSE_CUDA=OFF ..```
This is an automa
TaoLv commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549226728
Could you please also share your cmake command line?
This is an automate
cjolivier01 commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549224986
I don't get this error when building with Makefile, by the way
Thi
cjolivier01 edited a comment on issue #11417: libomp.so dependency (need REAL
fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549224808
I think removing this in CMakeLists.txt may fix it:
mxnet_option(USE_MKL_IF_AVAILABLE "Use MKL if found" OFF), but I thin
hgt312 commented on a change in pull request #16717: [Numpy] add op nan_to_num
URL: https://github.com/apache/incubator-mxnet/pull/16717#discussion_r341900803
##
File path: python/mxnet/symbol/numpy/_symbol.py
##
@@ -4824,4 +4824,102 @@ def resize(a, new_shape):
return
hgt312 commented on a change in pull request #16717: [Numpy] add op nan_to_num
URL: https://github.com/apache/incubator-mxnet/pull/16717#discussion_r341904218
##
File path: src/operator/tensor/elemwise_unary_op.h
##
@@ -660,6 +661,141 @@ void AroundOpForward(const nnvm::Nod
hgt312 commented on a change in pull request #16717: [Numpy] add op nan_to_num
URL: https://github.com/apache/incubator-mxnet/pull/16717#discussion_r341899364
##
File path: python/mxnet/ndarray/numpy/_op.py
##
@@ -5168,3 +5169,102 @@ def resize(a, new_shape):
[0
hgt312 commented on a change in pull request #16717: [Numpy] add op nan_to_num
URL: https://github.com/apache/incubator-mxnet/pull/16717#discussion_r341904046
##
File path: src/operator/tensor/elemwise_unary_op.h
##
@@ -660,6 +661,141 @@ void AroundOpForward(const nnvm::Nod
cjolivier01 commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549224808
I think removing this in CMakeLists.txt may fix it:
mxnet_option(USE_MKL_IF_AVAILABLE "Use MKL if found" OFF)
--
cjolivier01 commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549224751
```cpp
/home/coolivie/src/mxnet/src/c_api/../operator/tensor/././../mkl_functions-inl.h:50:5:
error: ‘vsLog2’ was not declared
ChaiBapchya opened a new issue #16718: Cleaner API for utilizing all GPUs if
available
URL: https://github.com/apache/incubator-mxnet/issues/16718
## Description
Can we have a cleaner way of utilizing GPUs?
## Current scenario
According to this [1]
```
from mxnet impo
pengzhao-intel commented on issue #16695: [Quantization] Enhance gluon
quantization API
URL: https://github.com/apache/incubator-mxnet/pull/16695#issuecomment-549220956
@xinyu-intel @ZhennanQin please take a review
This is an
Alicia1529 opened a new pull request #16717: [Numpy] add op nan_to_num
URL: https://github.com/apache/incubator-mxnet/pull/16717
## Description ##
add numpy operator nan_to_num
This is an automated message from
xidulu commented on issue #16713: [Numpy] Improve error message in random
distribution
URL:
https://github.com/apache/incubator-mxnet/issues/16713#issuecomment-549208776
@sxjscience sure, I'll look into this probelm after I submit my Bernoulli
implementation.
---
sxjscience commented on issue #16708: Training an FPN model using
grad_req="add" causes rapid divergence, while manually implemented gradient
accumulation works fine
URL:
https://github.com/apache/incubator-mxnet/issues/16708#issuecomment-549207390
Thanks for reporting this! It looks tha
sxjscience commented on issue #16713: [Numpy] Improve error message in random
distribution
URL:
https://github.com/apache/incubator-mxnet/issues/16713#issuecomment-549207186
@xidulu Would you submit a PR to revise it?
This i
sxjscience opened a new pull request #16716: [Numpy][WIP] Fix
collect_params().zero_grad() in gluon numpy interface
URL: https://github.com/apache/incubator-mxnet/pull/16716
## Description ##
## Checklist ##
### Essentials ###
Please feel free to remove inapplicable items f
access2rohit opened a new pull request #16715: Lamb optimizer update
URL: https://github.com/apache/incubator-mxnet/pull/16715
## Description ##
adding to new operators:
lamb_update
multi_mp_lamb_update
## Checklist ##
### Essentials ###
Please feel free to remove inapp
TaoLv commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549200598
@cjolivier01 Could you please be more specific about the `failing mkldnn
build`? We removed iomp5 dependency in the commit but mkldnn sho
hubutui commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549199632
@cjolivier01 Here is my output:
```
ldd /usr/lib/libmxnet.so|grep omp
libomp.so => /usr/lib/libomp.so (0x7fb5be6e)
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 74e94eb Bump the publis
ChaiBapchya commented on issue #16712: update default windows Visual Studio
from 2015 to 2017
URL: https://github.com/apache/incubator-mxnet/pull/16712#issuecomment-549175041
Ooops (I forgot i had manually upgraded the opencv version on windows
instance)
Putting a PR on the concerned re
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 f7fb2b9 Bump the publis
cjolivier01 commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549160451
I am unable to get mxnet to build with cmake with CUDA disabled, general
cmake build seems to be broken. Failing mkldnn build.
access2rohit opened a new pull request #16714: [WIP] enabling large tensor
support for binary broadcast operators
URL: https://github.com/apache/incubator-mxnet/pull/16714
## Description ##
enabling int64 for binary broadcast ops:
TOPIOp
_contrib_tvm_vadd
numpy binary logic ops
xidulu opened a new issue #16713: [Numpy
URL: https://github.com/apache/incubator-mxnet/issues/16713
## Description
(A clear and concise description of what the bug is.)
### Error Message
(Paste the complete error message, including stack trace.)
## To Reproduce
(If yo
reminisce commented on a change in pull request #16699: [DO NOT MERGE] [DO NOT
REVIEW] [WIP]Mixed binary ops
URL: https://github.com/apache/incubator-mxnet/pull/16699#discussion_r341859430
##
File path: src/common/utils.h
##
@@ -870,12 +870,12 @@ inline int more_precise_ty
cjolivier01 commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549151224
recent changes involving mkl have made changes to omp linkages:
https://github.com/apache/incubator-mxnet/commit/aa1074dc1704d3732a
cjolivier01 commented on issue #11417: libomp.so dependency (need REAL fix)
URL:
https://github.com/apache/incubator-mxnet/issues/11417#issuecomment-549148838
the openmp_found would be a separate issue. what is the exact issue you’re
seeing? does ldd show more than one omp library in your
xinyu-intel commented on a change in pull request #16710: [MKLDNN] support
mkldnn gelu
URL: https://github.com/apache/incubator-mxnet/pull/16710#discussion_r341855594
##
File path: tests/python/mkl/test_subgraph.py
##
@@ -329,6 +329,8 @@ def conv_act(no_bias, data_shape, a
TaoLv commented on a change in pull request #16710: [MKLDNN] support mkldnn gelu
URL: https://github.com/apache/incubator-mxnet/pull/16710#discussion_r341854377
##
File path: tests/python/mkl/test_subgraph.py
##
@@ -329,6 +329,8 @@ def conv_act(no_bias, data_shape, alg):
TaoLv commented on issue #15294: mkldnn is not properly installed
URL:
https://github.com/apache/incubator-mxnet/issues/15294#issuecomment-549141465
@hubutui Do you think it's still a problem if we statically link `mkldnn.a`
into mxnet? I know there was an effort in mxnet community for tha
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 a747ea9 Bump the publis
marcoabreu commented on issue #16712: update default windows Visual Studio from
2015 to 2017
URL: https://github.com/apache/incubator-mxnet/pull/16712#issuecomment-549126390
How did you verify exactly? CI is failing 🙈
This is
haojin2 commented on a change in pull request #16699: [DO NOT MERGE] [DO NOT
REVIEW] [WIP]Mixed binary ops
URL: https://github.com/apache/incubator-mxnet/pull/16699#discussion_r341845058
##
File path: src/common/utils.h
##
@@ -870,12 +870,12 @@ inline int more_precise_type
64 matches
Mail list logo