[GitHub] [incubator-mxnet] Talgin commented on issue #15553: Installing mxnet on PowerPC

[GitBox]

Talgin commented on issue #15553:
URL: 
https://github.com/apache/incubator-mxnet/issues/15553#issuecomment-625645122


   @pasa13142, we had the same problem, training was slower on power9 with 
Tesla V100  than on my workstation. I do not remember if they found a reason, 
but our server management service said that ppc was not originally made to do 
such things (training) :) Like they are good at calculation, maybe some kind of 
server operations, but they (ibms ppc) do not have much experience with GPUs. 
   I would also be happy if you share the solution, if you found one.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] vexilligera commented on issue #18184: Flaky Incorrect Results test_numpy_op.py:: test_np_linalg_slogdet

[GitBox]

vexilligera commented on issue #18184:
URL: 
https://github.com/apache/incubator-mxnet/issues/18184#issuecomment-625638615


   > @vexilligera just making sure that you're running it on windows. I 
observed that the error only happens on windows.
   
   Of course, on G3 with Windows Server 2016.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ciyongch commented on pull request #18261: [v1.7.x] Backport fixing batch_norm and layer_norm for large tensors (#17805)

[GitBox]

ciyongch commented on pull request #18261:
URL: https://github.com/apache/incubator-mxnet/pull/18261#issuecomment-625636061


   Seems it failed to fetch the external packages for TVM, please re-trigger 
the CI.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] TaoLv commented on issue #18255: MKL_USE_STATIC_LIBS broken

[GitBox]

TaoLv commented on issue #18255:
URL: 
https://github.com/apache/incubator-mxnet/issues/18255#issuecomment-625625333


   I cannot reproduce the issue. Pulled the latest master branch and built it 
with the command line as below (I dont have lapack and ninja in my system):
   ```
   cmake -DUSE_MKLDNN=0 -DMKL_USE_STATIC_LIBS=0 -DUSE_CUDA=0 .. -DUSE_LAPACK=0
   make -j40
   ```
   Try to reproduce:
   ```
   (mxnet) [lvtao@mlt2-clx017 ~]$ python3 -c 'import mxnet as mx; 
print(mx.nd.square(mx.nd.random.uniform(shape=(1024,'
   
   [0.30119628 0.35146472 0.51149577 ... 0.12286104 0.28916994 0.35195467]
   
   ```
   
   openblas is linked. MKL is installed to /opt/intel/mkl, so I assume it can 
be found in cmake.
   ```
   (mxnet) [lvtao@mlt2-clx017 build]$ ldd libmxnet.so
   linux-vdso.so.1 =>  (0x7ffd5e5bf000)
   libopenblas.so.0 => 
/home/lvtao/miniconda3/envs/mxnet/lib/libopenblas.so.0 (0x7f73d603)
   librt.so.1 => /lib64/librt.so.1 (0x7f73d5e28000)
   libopencv_imgproc.so.2.4 => /lib64/libopencv_imgproc.so.2.4 
(0x7f73d59aa000)
   libopencv_highgui.so.2.4 => /lib64/libopencv_highgui.so.2.4 
(0x7f73d5763000)
   libopencv_core.so.2.4 => /lib64/libopencv_core.so.2.4 
(0x7f73d532a000)
   libomp.so => 
/home/lvtao/Workspace/mxnet-temp/build/3rdparty/openmp/runtime/src/libomp.so 
(0x7f73d5046000)
   libpthread.so.0 => /lib64/libpthread.so.0 (0x7f73d4e2a000)
   libdl.so.2 => /lib64/libdl.so.2 (0x7f73d4c26000)
   libgomp.so.1 => /home/lvtao/miniconda3/envs/mxnet/lib/libgomp.so.1 
(0x7f73d483c000)
   libstdc++.so.6 => 
/home/lvtao/miniconda3/envs/mxnet/lib/libstdc++.so.6 (0x7f73dc258000)
   libm.so.6 => /lib64/libm.so.6 (0x7f73d453a000)
   libgcc_s.so.1 => /home/lvtao/miniconda3/envs/mxnet/lib/libgcc_s.so.1 
(0x7f73dc243000)
   libc.so.6 => /lib64/libc.so.6 (0x7f73d416c000)
   /lib64/ld-linux-x86-64.so.2 (0x7f73dc1c7000)
   libgfortran.so.4 => 
/home/lvtao/miniconda3/envs/mxnet/lib/./libgfortran.so.4 (0x7f73d403e000)
   libz.so.1 => /home/lvtao/miniconda3/envs/mxnet/lib/libz.so.1 
(0x7f73dc222000)
   libjpeg.so.62 => /lib64/libjpeg.so.62 (0x7f73d3de9000)
   libpng15.so.15 => /lib64/libpng15.so.15 (0x7f73d3bbe000)
   libtiff.so.5 => /lib64/libtiff.so.5 (0x7f73d394a000)
   libjasper.so.1 => /lib64/libjasper.so.1 (0x7f73d36f)
   libImath.so.6 => /lib64/libImath.so.6 (0x7f73d34de000)
   libIlmImf.so.7 => /lib64/libIlmImf.so.7 (0x7f73d3216000)
   libIex.so.6 => /lib64/libIex.so.6 (0x7f73d2ff7000)
   libHalf.so.6 => /lib64/libHalf.so.6 (0x7f73d2db4000)
   libIlmThread.so.6 => /lib64/libIlmThread.so.6 (0x7f73d2bad000)
   libgtk-x11-2.0.so.0 => /lib64/libgtk-x11-2.0.so.0 
(0x7f73d252b000)
   libgdk-x11-2.0.so.0 => /lib64/libgdk-x11-2.0.so.0 
(0x7f73d226a000)
   libatk-1.0.so.0 => /lib64/libatk-1.0.so.0 (0x7f73d2044000)
   libgio-2.0.so.0 => /lib64/libgio-2.0.so.0 (0x7f73d1ca5000)
   libpangoft2-1.0.so.0 => /lib64/libpangoft2-1.0.so.0 
(0x7f73d1a8f000)
   libpangocairo-1.0.so.0 => /lib64/libpangocairo-1.0.so.0 
(0x7f73d1881000)
   libgdk_pixbuf-2.0.so.0 => /lib64/libgdk_pixbuf-2.0.so.0 
(0x7f73d1659000)
   libcairo.so.2 => /lib64/libcairo.so.2 (0x7f73d1322000)
   libpango-1.0.so.0 => /lib64/libpango-1.0.so.0 (0x7f73d10dc000)
   libfontconfig.so.1 => /lib64/libfontconfig.so.1 (0x7f73d0e9a000)
   libgobject-2.0.so.0 => /lib64/libgobject-2.0.so.0 
(0x7f73d0c49000)
   libglib-2.0.so.0 => /lib64/libglib-2.0.so.0 (0x7f73d0933000)
   libfreetype.so.6 => /lib64/libfreetype.so.6 (0x7f73d0674000)
   libgthread-2.0.so.0 => /lib64/libgthread-2.0.so.0 
(0x7f73d0472000)
   libgstbase-0.10.so.0 => /lib64/libgstbase-0.10.so.0 
(0x7f73d021e000)
   libgstreamer-0.10.so.0 => /lib64/libgstreamer-0.10.so.0 
(0x7f73cff35000)
   libgmodule-2.0.so.0 => /lib64/libgmodule-2.0.so.0 
(0x7f73cfd31000)
   libxml2.so.2 => /lib64/libxml2.so.2 (0x7f73cf9c7000)
   libgstapp-0.10.so.0 => /lib64/libgstapp-0.10.so.0 
(0x7f73cf7bb000)
   libgstvideo-0.10.so.0 => /lib64/libgstvideo-0.10.so.0 
(0x7f73cf59e000)
   libv4l1.so.0 => /lib64/libv4l1.so.0 (0x7f73cf398000)
   libquadmath.so.0 => 
/home/lvtao/miniconda3/envs/mxnet/lib/./libquadmath.so.0 (0x7f73cf35e000)
   libjbig.so.2.0 => /lib64/libjbig.so.2.0 (0x7f73cf152000)
   libIexMath.so.6 => /lib64/libIexMath.so.6 (0x7f73cef4d000)
   libX11.so.6 => /lib64/libX11.so.6 (0x7f73cec0f000)
   libXfixes.so.3 => /lib64/libXfixes.so.3 (0x7f73cea09000)
   libXrender.so.1 => 

[GitHub] [incubator-mxnet] szha commented on issue #18184: Flaky Incorrect Results test_numpy_op.py:: test_np_linalg_slogdet

[GitBox]

szha commented on issue #18184:
URL: 
https://github.com/apache/incubator-mxnet/issues/18184#issuecomment-625624201


   @vexilligera just making sure that you're running it on windows. I observed 
that the error only happens on windows.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] vexilligera commented on issue #18184: Flaky Incorrect Results test_numpy_op.py:: test_np_linalg_slogdet

[GitBox]

vexilligera commented on issue #18184:
URL: 
https://github.com/apache/incubator-mxnet/issues/18184#issuecomment-625612824


   > Did you try on G3 instances? The other question is why the 
`rtol=atol=1e-1` and not lower
   
   I later reran the test on a G3 instance for 100 times with tolerance at 
1e-2, still didn't get the error.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ciyongch commented on issue #18246: Night test fail with "test_large_vector.test_nn" on v1.7.x branch

[GitBox]

ciyongch commented on issue #18246:
URL: 
https://github.com/apache/incubator-mxnet/issues/18246#issuecomment-625602823


   Hi @ChaiBapchya , thanks for your kindly help to identify and point out the 
fix:)



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ChaiBapchya commented on pull request #18261: [v1.7.x] Backport fixing batch_norm and layer_norm for large tensors (#17805)

[GitBox]

ChaiBapchya commented on pull request #18261:
URL: https://github.com/apache/incubator-mxnet/pull/18261#issuecomment-625597268


   Fixes https://github.com/apache/incubator-mxnet/issues/18246



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18261: [v1.7.x] Backport fixing batch_norm and layer_norm for large tensors (#17805)

[GitBox]

mxnet-bot commented on pull request #18261:
URL: https://github.com/apache/incubator-mxnet/pull/18261#issuecomment-625596968


   Hey @ChaiBapchya , Thanks for submitting the PR 
   All tests are already queued to run once. If tests fail, you can trigger one 
or more tests again with the following commands: 
   - To trigger all jobs: @mxnet-bot run ci [all] 
   - To trigger specific jobs: @mxnet-bot run ci [job1, job2] 
   *** 
   **CI supported jobs**: [miscellaneous, centos-cpu, sanity, centos-gpu, 
website, windows-gpu, unix-cpu, clang, unix-gpu, edge, windows-cpu]
   *** 
   _Note_: 
Only following 3 categories can trigger CI :PR Author, MXNet Committer, 
Jenkins Admin. 
   All CI tests must pass before the PR can be merged. 
   



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ChaiBapchya opened a new pull request #18261: [v1.7.x] Backport fixing batch_norm and layer_norm for large tensors (#17805)

[GitBox]

ChaiBapchya opened a new pull request #18261:
URL: https://github.com/apache/incubator-mxnet/pull/18261


   Backport to fix large tensor nightly test issue
   Co-authored-by: Rohit Kumar Srivastava 
   
   Verified the changes locally by
   
   setting up the Infra 
https://cwiki.apache.org/confluence/display/MXNET/Reproducing+test+results
   Build : 
   ```
   sudo ci/build.py --docker-registry mxnetci --platform ubuntu_nightly_cpu 
--docker-build-retries 3 --shm-size 500m /work/runtime_functions.sh 
build_ubuntu_cpu_large_tensor
   ```
   Test
   ```
   sudo ci/build.py --docker-registry mxnetci --platform ubuntu_nightly_cpu 
--docker-build-retries 3 --shm-size 500m /work/runtime_functions.sh 
nightly_test_large_vector
   ```
   ```



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] jaisanant0 commented on issue #15275: How to run mxnet(C++) in single-thread mode?

[GitBox]

jaisanant0 commented on issue #15275:
URL: 
https://github.com/apache/incubator-mxnet/issues/15275#issuecomment-62559


   @cyrusbehr 
   
   as you mentioned i have moved the variables mxnetConfig1 mxnetConfig2  into 
class member but the number of threads spawned is qual to number of cores.
   Any solution



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ciyongch commented on pull request #17959: [MKLDNN] Add LSTMP to v1.6.x

[GitBox]

ciyongch commented on pull request #17959:
URL: https://github.com/apache/incubator-mxnet/pull/17959#issuecomment-625589023


   Thanks for your understand and help making it clear @TaoLv @ChaiBapchya .



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.

[aaronmarkham]

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 0ecd800  Bump the publish timestamp.
0ecd800 is described below

commit 0ecd800d42307918a4225599322097674ae9075d
Author: mxnet-ci 
AuthorDate: Fri May 8 00:47:49 2020 +

Bump the publish timestamp.
---
 date.txt | 1 +
 1 file changed, 1 insertion(+)

diff --git a/date.txt b/date.txt
new file mode 100644
index 000..f92e913
--- /dev/null
+++ b/date.txt
@@ -0,0 +1 @@
+Fri May  8 00:47:49 UTC 2020

[incubator-mxnet] branch master updated: fixed overwrite of args/aux variables (#18232)

[marcoabreu]

This is an automated email from the ASF dual-hosted git repository.

marcoabreu 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 68cb955  fixed overwrite of args/aux variables (#18232)
68cb955 is described below

commit 68cb9555c4b4779aaae90e593b745270cbb59033
Author: Sam Skalicky 
AuthorDate: Thu May 7 16:31:49 2020 -0700

fixed overwrite of args/aux variables (#18232)

* fixed overwrite of args/aux variables

* fixed spacing
---
 python/mxnet/symbol/symbol.py | 16 
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/python/mxnet/symbol/symbol.py b/python/mxnet/symbol/symbol.py
index 14944a5..d4ff795 100644
--- a/python/mxnet/symbol/symbol.py
+++ b/python/mxnet/symbol/symbol.py
@@ -1485,18 +1485,18 @@ class Symbol(SymbolBase):
 assert isinstance(backend, str)
 
 if args is None or len(args) == 0:
-args = []
+args_ = []
 args_handle = c_array(NDArrayHandle, [])
 else:
-args_handle, args = self._get_ndarray_inputs('args', args,
- 
self.list_arguments(), False)
+args_handle, args_ = self._get_ndarray_inputs('args', args,
+  
self.list_arguments(), False)
 
 if aux is None or len(aux) == 0:
-aux = []
+aux_ = []
 aux_handle = c_array(NDArrayHandle, [])
 else:
-aux_handle, aux = self._get_ndarray_inputs('aux_states', aux,
-   
self.list_auxiliary_states(), False)
+aux_handle, aux_ = self._get_ndarray_inputs('aux_states', aux,
+
self.list_auxiliary_states(), False)
 if ctx is None:
 ctx = current_context()
 assert isinstance(ctx, Context)
@@ -1517,9 +1517,9 @@ class Symbol(SymbolBase):
  c_str(backend),
  ctypes.c_int(ctx.device_typeid),
  ctypes.byref(out),
- mx_uint(len(args)),
+ mx_uint(len(args_)),
  args_handle,
- mx_uint(len(aux)),
+ mx_uint(len(aux_)),
  aux_handle,
  mx_uint(len(key_list)),
  c_str_array(key_list),

[incubator-mxnet] branch master updated: fixed overwrite of args/aux variables (#18232)

[marcoabreu]

This is an automated email from the ASF dual-hosted git repository.

marcoabreu 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 68cb955  fixed overwrite of args/aux variables (#18232)
68cb955 is described below

commit 68cb9555c4b4779aaae90e593b745270cbb59033
Author: Sam Skalicky 
AuthorDate: Thu May 7 16:31:49 2020 -0700

fixed overwrite of args/aux variables (#18232)

* fixed overwrite of args/aux variables

* fixed spacing
---
 python/mxnet/symbol/symbol.py | 16 
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/python/mxnet/symbol/symbol.py b/python/mxnet/symbol/symbol.py
index 14944a5..d4ff795 100644
--- a/python/mxnet/symbol/symbol.py
+++ b/python/mxnet/symbol/symbol.py
@@ -1485,18 +1485,18 @@ class Symbol(SymbolBase):
 assert isinstance(backend, str)
 
 if args is None or len(args) == 0:
-args = []
+args_ = []
 args_handle = c_array(NDArrayHandle, [])
 else:
-args_handle, args = self._get_ndarray_inputs('args', args,
- 
self.list_arguments(), False)
+args_handle, args_ = self._get_ndarray_inputs('args', args,
+  
self.list_arguments(), False)
 
 if aux is None or len(aux) == 0:
-aux = []
+aux_ = []
 aux_handle = c_array(NDArrayHandle, [])
 else:
-aux_handle, aux = self._get_ndarray_inputs('aux_states', aux,
-   
self.list_auxiliary_states(), False)
+aux_handle, aux_ = self._get_ndarray_inputs('aux_states', aux,
+
self.list_auxiliary_states(), False)
 if ctx is None:
 ctx = current_context()
 assert isinstance(ctx, Context)
@@ -1517,9 +1517,9 @@ class Symbol(SymbolBase):
  c_str(backend),
  ctypes.c_int(ctx.device_typeid),
  ctypes.byref(out),
- mx_uint(len(args)),
+ mx_uint(len(args_)),
  args_handle,
- mx_uint(len(aux)),
+ mx_uint(len(aux_)),
  aux_handle,
  mx_uint(len(key_list)),
  c_str_array(key_list),

[incubator-mxnet] branch master updated: Change include to relative in nvvm_to_onnx.cc (#18249)

[marcoabreu]

This is an automated email from the ASF dual-hosted git repository.

marcoabreu 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 21b187b  Change include to relative in nvvm_to_onnx.cc (#18249)
21b187b is described below

commit 21b187bc3283df5f6deafca574a11f6c9316491b
Author: Serge Panev 
AuthorDate: Fri May 8 01:31:19 2020 +0200

Change include to relative in nvvm_to_onnx.cc (#18249)

Signed-off-by: Serge Panev 
---
 src/operator/subgraph/tensorrt/nnvm_to_onnx.cc | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/operator/subgraph/tensorrt/nnvm_to_onnx.cc 
b/src/operator/subgraph/tensorrt/nnvm_to_onnx.cc
index 19d0f26..4f80d27 100644
--- a/src/operator/subgraph/tensorrt/nnvm_to_onnx.cc
+++ b/src/operator/subgraph/tensorrt/nnvm_to_onnx.cc
@@ -31,7 +31,6 @@
 #include 
 #include 
 #include 
-#include 
 
 #include "../../../common/utils.h"
 #include "../../../ndarray/ndarray_function.h"
@@ -39,6 +38,7 @@
 #include "../../nn/activation-inl.h"
 #include "../../nn/batch_norm-inl.h"
 #include "../../nn/convolution-inl.h"
+#include "../../nn/deconvolution-inl.h"
 #include "../../nn/fully_connected-inl.h"
 #include "../../nn/pooling-inl.h"
 #include "../../nn/concat-inl.h"

[incubator-mxnet] branch master updated (353c243 -> 7f24823)

[marcoabreu]

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 353c243  Remove duplicate large_vector_test (#18259)
 add 7f24823  Fixing #16655 (#18257)

No new revisions were added by this update.

Summary of changes:
 python/mxnet/numpy/multiarray.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

[incubator-mxnet] branch master updated (353c243 -> 7f24823)

[marcoabreu]

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 353c243  Remove duplicate large_vector_test (#18259)
 add 7f24823  Fixing #16655 (#18257)

No new revisions were added by this update.

Summary of changes:
 python/mxnet/numpy/multiarray.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

[incubator-mxnet] branch master updated (a0e6735 -> 353c243)

[marcoabreu]

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 a0e6735  Gluon data 2.0: c++ dataloader and built-in image/bbox 
transforms (#17841)
 add 353c243  Remove duplicate large_vector_test (#18259)

No new revisions were added by this update.

Summary of changes:
 ci/docker/runtime_functions.sh | 10 --
 1 file changed, 10 deletions(-)

[incubator-mxnet] branch master updated (a0e6735 -> 353c243)

[marcoabreu]

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 a0e6735  Gluon data 2.0: c++ dataloader and built-in image/bbox 
transforms (#17841)
 add 353c243  Remove duplicate large_vector_test (#18259)

No new revisions were added by this update.

Summary of changes:
 ci/docker/runtime_functions.sh | 10 --
 1 file changed, 10 deletions(-)

[incubator-mxnet] branch master updated (353c243 -> 7f24823)

[marcoabreu]

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 353c243  Remove duplicate large_vector_test (#18259)
 add 7f24823  Fixing #16655 (#18257)

No new revisions were added by this update.

Summary of changes:
 python/mxnet/numpy/multiarray.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

[incubator-mxnet] branch master updated (a0e6735 -> 353c243)

[marcoabreu]

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 a0e6735  Gluon data 2.0: c++ dataloader and built-in image/bbox 
transforms (#17841)
 add 353c243  Remove duplicate large_vector_test (#18259)

No new revisions were added by this update.

Summary of changes:
 ci/docker/runtime_functions.sh | 10 --
 1 file changed, 10 deletions(-)

[GitHub] [incubator-mxnet] access2rohit commented on issue #17331: [mxnet 2.0] [item 2.4] Turning on large tensor support by default

[GitBox]

access2rohit commented on issue #17331:
URL: 
https://github.com/apache/incubator-mxnet/issues/17331#issuecomment-625542495


   @apeforest THe profiling done by @JonTanS was done long back using 
mxnet-1.6in november. These results are using current master branch of MXNet, 
bert scripts have changed too. If there are newer setting for running BERT on 
single node they are not available on Gluon NLP site. If @eric-haibin-lin or 
@szhengac can verify whether my BERT is correct or not and also provide proper 
tuning params to run BERT on single node I will re-run benchmarks and update 
the results here.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ChaiBapchya commented on issue #18246: Night test fail with "test_large_vector.test_nn" on v1.7.x branch

[GitBox]

ChaiBapchya commented on issue #18246:
URL: 
https://github.com/apache/incubator-mxnet/issues/18246#issuecomment-625539228


   layer norm and batchnorm have been fixed in master by 
https://github.com/apache/incubator-mxnet/pull/17805
   
   But they aren't cherry-picked in 1.7.x as of today.
   Cherry-picked them now and tested locally
   Verified the large tensor tests pass
   
   



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] sxjscience commented on issue #18117: [Gluon][Numpy] Bug of save parameters with numpy front-end

[GitBox]

sxjscience commented on issue #18117:
URL: 
https://github.com/apache/incubator-mxnet/issues/18117#issuecomment-625506608


   I think we should not deprecate it.
   
   Get Outlook for iOS
   
   From: Leonard Lausen 
   Sent: Thursday, May 7, 2020 2:17:13 PM
   To: apache/incubator-mxnet 
   Cc: Xingjian SHI ; Mention 

   Subject: Re: [apache/incubator-mxnet] [Gluon][Numpy] Bug of save parameters 
with numpy front-end (#18117)
   
   
   @sxjscience being able to save a single 
parameter seems quite sensible. Why should we deprecate it? We may change the 
serialization format though
   
   —
   You are receiving this because you were mentioned.
   Reply to this email directly, view it on 
GitHub,
 or 
unsubscribe.
   



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] leezu commented on issue #18117: [Gluon][Numpy] Bug of save parameters with numpy front-end

[GitBox]

leezu commented on issue #18117:
URL: 
https://github.com/apache/incubator-mxnet/issues/18117#issuecomment-625502229


   @sxjscience being able to save a single parameter seems quite sensible. Why 
should we deprecate it? We may change the serialization format though



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18260: Fix interleave matmul doc

[GitBox]

mxnet-bot commented on pull request #18260:
URL: https://github.com/apache/incubator-mxnet/pull/18260#issuecomment-625476566


   Hey @eric-haibin-lin , Thanks for submitting the PR 
   All tests are already queued to run once. If tests fail, you can trigger one 
or more tests again with the following commands: 
   - To trigger all jobs: @mxnet-bot run ci [all] 
   - To trigger specific jobs: @mxnet-bot run ci [job1, job2] 
   *** 
   **CI supported jobs**: [clang, miscellaneous, centos-cpu, unix-gpu, 
unix-cpu, website, edge, sanity, windows-cpu, windows-gpu, centos-gpu]
   *** 
   _Note_: 
Only following 3 categories can trigger CI :PR Author, MXNet Committer, 
Jenkins Admin. 
   All CI tests must pass before the PR can be merged. 
   



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] eric-haibin-lin opened a new pull request #18260: Fix interleave matmul doc

[GitBox]

eric-haibin-lin opened a new pull request #18260:
URL: https://github.com/apache/incubator-mxnet/pull/18260


   ## Description ##
   Fix interleave matmul doc
   
   ## Checklist ##
   ### Essentials ###
   Please feel free to remove inapplicable items for your PR.
   - [ ] The PR title starts with [MXNET-$JIRA_ID], where $JIRA_ID refers to 
the relevant [JIRA issue](https://issues.apache.org/jira/projects/MXNET/issues) 
created (except PRs with tiny changes)
   - [ ] Changes are complete (i.e. I finished coding on this PR)
   - [ ] All changes have test coverage:
   - Unit tests are added for small changes to verify correctness (e.g. adding 
a new operator)
   - Nightly tests are added for complicated/long-running ones (e.g. changing 
distributed kvstore)
   - Build tests will be added for build configuration changes (e.g. adding a 
new build option with NCCL)
   - [ ] Code is well-documented: 
   - For user-facing API changes, API doc string has been updated. 
   - For new C++ functions in header files, their functionalities and arguments 
are documented. 
   - For new examples, README.md is added to explain the what the example does, 
the source of the dataset, expected performance on test set and reference to 
the original paper if applicable
   - Check the API doc at 
https://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html
   - [ ] To the best of my knowledge, examples are either not affected by this 
change, or have been fixed to be compatible with this change
   
   ### Changes ###
   - [ ] Feature1, tests, (and when applicable, API doc)
   - [ ] Feature2, tests, (and when applicable, API doc)
   
   ## Comments ##
   - If this change is a backward incompatible change, why must this change be 
made.
   - Interesting edge cases to note here
   



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ys2843 commented on pull request #18238: fix when clicking version dropdown it jumps to top of the page

[GitBox]

ys2843 commented on pull request #18238:
URL: https://github.com/apache/incubator-mxnet/pull/18238#issuecomment-625458290


   @ChaiBapchya The change should apply to master now, could you try disable 
cache and reload the website? Thanks.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ChaiBapchya commented on pull request #18238: fix when clicking version dropdown it jumps to top of the page

[GitBox]

ChaiBapchya commented on pull request #18238:
URL: https://github.com/apache/incubator-mxnet/pull/18238#issuecomment-625444923


   Works good when the global version of website is 1.6.0
   But when global version is master
   And MXNet version selection dropdown in Get Started page is chosen then it 
misbehaves [goes to top of the page]



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ChaiBapchya commented on pull request #18259: Remove duplicate large_vector_test in nightly introduced in #17388

[GitBox]

ChaiBapchya commented on pull request #18259:
URL: https://github.com/apache/incubator-mxnet/pull/18259#issuecomment-625436222


   @access2rohit @apeforest 



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18259: Remove duplicate large_vector_test in nightly introduced in #17388

[GitBox]

mxnet-bot commented on pull request #18259:
URL: https://github.com/apache/incubator-mxnet/pull/18259#issuecomment-625435955


   Hey @ChaiBapchya , Thanks for submitting the PR 
   All tests are already queued to run once. If tests fail, you can trigger one 
or more tests again with the following commands: 
   - To trigger all jobs: @mxnet-bot run ci [all] 
   - To trigger specific jobs: @mxnet-bot run ci [job1, job2] 
   *** 
   **CI supported jobs**: [centos-gpu, unix-gpu, clang, windows-gpu, edge, 
miscellaneous, centos-cpu, sanity, website, unix-cpu, windows-cpu]
   *** 
   _Note_: 
Only following 3 categories can trigger CI :PR Author, MXNet Committer, 
Jenkins Admin. 
   All CI tests must pass before the PR can be merged. 
   



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ChaiBapchya opened a new pull request #18259: Remove duplicate large_vector_test in nightly introduced in #17388

[GitBox]

ChaiBapchya opened a new pull request #18259:
URL: https://github.com/apache/incubator-mxnet/pull/18259


   ## Description ##
   Remove duplicate test



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] kmfrick commented on issue #18057: Edge device documentation

[GitBox]

kmfrick commented on issue #18057:
URL: 
https://github.com/apache/incubator-mxnet/issues/18057#issuecomment-625431210


   Yeah I'm building 1.6 now, I noticed half-way through the build that 1.4 was 
not the most recent release. I'll make a PR against that document when I have 
some time. Thank you!



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.

[aaronmarkham]

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 abb937b  Bump the publish timestamp.
abb937b is described below

commit abb937b9df36eb26d7213515d51a377491947d53
Author: mxnet-ci 
AuthorDate: Thu May 7 18:43:31 2020 +

Bump the publish timestamp.
---
 date.txt | 1 +
 1 file changed, 1 insertion(+)

diff --git a/date.txt b/date.txt
new file mode 100644
index 000..90759e5
--- /dev/null
+++ b/date.txt
@@ -0,0 +1 @@
+Thu May  7 18:43:31 UTC 2020

[GitHub] [incubator-mxnet] leezu commented on issue #15553: Installing mxnet on PowerPC

[GitBox]

leezu commented on issue #15553:
URL: 
https://github.com/apache/incubator-mxnet/issues/15553#issuecomment-625429275


   You need to investigate and find some minimal example that shows the 
problem. For example, by using the benchmark scripts 
https://github.com/apache/incubator-mxnet/tree/master/benchmark/opperf



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] leezu commented on a change in pull request #18083: Changes to mxnet.metric

[GitBox]

leezu commented on a change in pull request #18083:
URL: https://github.com/apache/incubator-mxnet/pull/18083#discussion_r421715173



##
File path: tests/python/tensorrt/test_cvnets.py
##
@@ -29,7 +28,12 @@
 def get_classif_model(model_name, use_tensorrt, ctx=mx.gpu(0), batch_size=128):
 mx.contrib.tensorrt.set_use_fp16(False)
 h, w = 32, 32
-net = gluoncv.model_zoo.get_model(model_name, pretrained=True)
+model_url = 
"https://raw.githubusercontent.com/dmlc/web-data/master/gluoncv/models/;

Review comment:
   Please don't hardcode `master` in the URL here. The repository may 
change and will then break the CI. Instead, use the commit ID: 
`https://raw.githubusercontent.com/dmlc/web-data/221ce5b7c6d5b0777a1e3471f7f03ff98da90a0a/gluoncv/models`





This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] leezu commented on a change in pull request #18083: Changes to mxnet.metric

[GitBox]

leezu commented on a change in pull request #18083:
URL: https://github.com/apache/incubator-mxnet/pull/18083#discussion_r421715173



##
File path: tests/python/tensorrt/test_cvnets.py
##
@@ -29,7 +28,12 @@
 def get_classif_model(model_name, use_tensorrt, ctx=mx.gpu(0), batch_size=128):
 mx.contrib.tensorrt.set_use_fp16(False)
 h, w = 32, 32
-net = gluoncv.model_zoo.get_model(model_name, pretrained=True)
+model_url = 
"https://raw.githubusercontent.com/dmlc/web-data/master/gluoncv/models/;

Review comment:
   Please don't hardcode `master` in the URL here. The repository may 
change and will then break the CI. Instead, use the commit ID: 
`https://raw.githubusercontent.com/dmlc/web-data/221ce5b7c6d5b0777a1e3471f7f03ff98da90a0a/gluoncv/models/;





This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] leezu commented on issue #18057: Edge device documentation

[GitBox]

leezu commented on issue #18057:
URL: 
https://github.com/apache/incubator-mxnet/issues/18057#issuecomment-625427543


   @kmfrick you can consider making a PR to update 
https://github.com/apache/incubator-mxnet/blob/master/docs/static_site/src/pages/get_started/jetson_setup.md
   
   If you don't need MXNet 1.4 in particular, I'd rather recommend you to build 
1.6, or even the per-release 1.7. It's unfortunate that above guide "hardcoded" 
the 1.4 version when it was written, instead of referring to the latest stable 
version.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] leezu commented on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

[GitBox]

leezu commented on issue #13773:
URL: 
https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625425470


   You shouldn't downgrade the GCC version. This is a compiler bug. You need to 
update the version to get a gcc version that has the fix for your bug...
   
   If no version contains a fix, then you need to report a bug report at 
https://gcc.gnu.org/bugs/
   
   I'll close this issue as it's unrelated to MXNet, but feel free to discuss 
here further.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #17852: [mxnet2.0] gluon 2.0 data: deprecate dataset transform= argument

[GitBox]

mxnet-bot commented on pull request #17852:
URL: https://github.com/apache/incubator-mxnet/pull/17852#issuecomment-625398499


   Jenkins CI successfully triggered : [edge, centos-cpu, centos-gpu, unix-gpu, 
windows-gpu, unix-cpu, windows-cpu, clang, sanity, website, miscellaneous]



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] zhreshold commented on pull request #17852: [mxnet2.0] gluon 2.0 data: deprecate dataset transform= argument

[GitBox]

zhreshold commented on pull request #17852:
URL: https://github.com/apache/incubator-mxnet/pull/17852#issuecomment-625398374


   @mxnet-bot run ci [all]



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] access2rohit commented on issue #18258: Opperf fails when running for all operators (both CPU and GPU ctx)

[GitBox]

access2rohit commented on issue #18258:
URL: 
https://github.com/apache/incubator-mxnet/issues/18258#issuecomment-625395133


   @ChaiBapchya ^



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] access2rohit opened a new issue #18258: Opperf fails when running for all operators (both CPU and GPU ctx)

[GitBox]

access2rohit opened a new issue #18258:
URL: https://github.com/apache/incubator-mxnet/issues/18258


   ## Description
   I ran the following command to obtain individual operator performance w/ and 
w/o Large Tensor Support on both CPU and GPU
   
   ```
   #CPU
   python incubator-mxnet/benchmark/opperf/opperf.py --output-format json 
--output-file mxnet_operator_benchmark_results.json
   
   #GPU
   python incubator-mxnet/benchmark/opperf/opperf.py --ctx gpu --output-format 
json --output-file mxnet_operator_benchmark_results.json
   ```
   
   ### Error Message
   In both contexts(CPU and GPU) I get the following error log:
   
   ```
   INFO:root:Begin Benchmark - BatchNorm
   INFO:root:Complete Benchmark - BatchNorm
   INFO:root:Begin Benchmark - Correlation
   INFO:root:Complete Benchmark - Correlation
   INFO:root:Begin Benchmark - Dropout
   INFO:root:Complete Benchmark - Dropout
   INFO:root:Begin Benchmark - Embedding
   INFO:root:Complete Benchmark - Embedding
   INFO:root:Begin Benchmark - FullyConnected
   INFO:root:Complete Benchmark - FullyConnected
   Traceback (most recent call last):
 File "benchmark/opperf/opperf.py", line 227, in 
   sys.exit(main())
 File "benchmark/opperf/opperf.py", line 207, in main
   benchmark_results = run_all_mxnet_operator_benchmarks(ctx=ctx, 
dtype=dtype, profiler=profiler, int64_tensor=int64_tensor, warmup=warmup, 
runs=runs)
 File "benchmark/opperf/opperf.py", line 113, in 
run_all_mxnet_operator_benchmarks
   
mxnet_operator_benchmark_results.append(run_nn_basic_operators_benchmarks(ctx=ctx,
 dtype=dtype, profiler=profiler, int64_tensor=int64_tensor, warmup=warmup, 
runs=runs))
 File 
"/home/ubuntu/workspace/incubator-mxnet/benchmark/opperf/nd_operations/nn_basic_operators.py",
 line 143, in run_nn_basic_operators_benchmarks
   mx_nn_basic_op_results = run_op_benchmarks(mx_nn_basic_ops, dtype, ctx, 
profiler, int64_tensor, warmup, runs)
 File 
"/home/ubuntu/workspace/incubator-mxnet/benchmark/opperf/utils/benchmark_utils.py",
 line 210, in run_op_benchmarks
   warmup=warmup, runs=runs)
 File 
"/home/ubuntu/workspace/incubator-mxnet/benchmark/opperf/utils/benchmark_utils.py",
 line 177, in run_performance_test
   benchmark_result = _run_nd_operator_performance_test(op, inputs, 
run_backward, warmup, runs, kwargs_list, profiler)
 File 
"/home/ubuntu/workspace/incubator-mxnet/benchmark/opperf/utils/benchmark_utils.py",
 line 114, in _run_nd_operator_performance_test
   _, _ = benchmark_helper_func(op, warmup, **kwargs_list[0])
 File 
"/home/ubuntu/workspace/incubator-mxnet/benchmark/opperf/utils/profiler_utils.py",
 line 200, in cpp_profile_it
   res = func(*args, **kwargs)
 File 
"/home/ubuntu/workspace/incubator-mxnet/benchmark/opperf/utils/ndarray_utils.py",
 line 60, in nd_forward_backward_and_profile
   nd.waitall()
 File 
"/home/ubuntu/workspace/incubator-mxnet/python/mxnet/ndarray/ndarray.py", line 
211, in waitall
   check_call(_LIB.MXNDArrayWaitAll())
 File "/home/ubuntu/workspace/incubator-mxnet/python/mxnet/base.py", line 
246, in check_call
   raise get_last_ffi_error()
   mxnet.base.MXNetError: Traceback (most recent call last):
 File "../include/mxnet/././tensor_blob.h", line 198
   ```
   
   ## Environment
   
   We recommend using our script for collecting the diagnositc information. Run 
the following command and paste the outputs below:
   ```
   curl --retry 10 -s 
https://raw.githubusercontent.com/dmlc/gluon-nlp/master/tools/diagnose.py | 
python
   
   # paste outputs here
   ```
   
   ```
   ubuntu@ip-172-31-0-156 ~/workspace/incubator-mxnet (master) $ curl --retry 
10 -s https://raw.githubusercontent.com/dmlc/gluon-nlp/master/tools/diagnose.py 
| python
   --Python Info--
   Version  : 3.6.6
   Compiler : GCC 7.3.0
   Build: ('default', 'Oct  9 2018 12:34:16')
   Arch : ('64bit', '')
   Pip Info---
   Version  : 19.3.1
   Directory: /home/ubuntu/anaconda3/lib/python3.6/site-packages/pip
   --MXNet Info---
   Version  : 2.0.0
   Directory: /home/ubuntu/workspace/incubator-mxnet/python/mxnet
   Num GPUs : 8
   Hashtag not found. Not installed from pre-built package.
   --System Info--
   Platform : Linux-5.3.0-1017-aws-x86_64-with-debian-buster-sid
   system   : Linux
   node : ip-172-31-0-156
   release  : 5.3.0-1017-aws
   version  : #18~18.04.1-Ubuntu SMP Wed Apr 8 15:12:16 UTC 2020
   --Hardware Info--
   machine  : x86_64
   processor: x86_64
   Architecture:x86_64
   CPU op-mode(s):  32-bit, 64-bit
   Byte Order:  Little Endian
   CPU(s):  64
   On-line CPU(s) list: 0-63
   Thread(s) per core:  2
   Core(s) per socket:  16
   Socket(s):   2
   NUMA node(s):2
   Vendor ID:   GenuineIntel
   CPU family:  6
   Model:

[GitHub] [incubator-mxnet] JinboCi opened a new pull request #18257: Fixing #16655

[GitBox]

JinboCi opened a new pull request #18257:
URL: https://github.com/apache/incubator-mxnet/pull/18257


   ## Description ##
   (Brief description on what this PR is about)
   This pull request is opened to fix the issue 
[#16655](https://github.com/apache/incubator-mxnet/issues/16655). The previous 
implementation causes the error message 
   
   > TypeError: 'dict_keys' object does not support indexing
   
   in Python 3.6.6.
   
   ## Checklist ##
   ### Essentials ###
   Please feel free to remove inapplicable items for your PR.
   - [ ] The PR title starts with [MXNET-$JIRA_ID], where $JIRA_ID refers to 
the relevant [JIRA issue](https://issues.apache.org/jira/projects/MXNET/issues) 
created (except PRs with tiny changes)
   - [ ] Changes are complete (i.e. I finished coding on this PR)
   - [ ] All changes have test coverage:
   - Unit tests are added for small changes to verify correctness (e.g. adding 
a new operator)
   - Nightly tests are added for complicated/long-running ones (e.g. changing 
distributed kvstore)
   - Build tests will be added for build configuration changes (e.g. adding a 
new build option with NCCL)
   - [ ] Code is well-documented: 
   - For user-facing API changes, API doc string has been updated. 
   - For new C++ functions in header files, their functionalities and arguments 
are documented. 
   - For new examples, README.md is added to explain the what the example does, 
the source of the dataset, expected performance on test set and reference to 
the original paper if applicable
   - Check the API doc at 
https://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html
   - [ ] To the best of my knowledge, examples are either not affected by this 
change, or have been fixed to be compatible with this change
   
   ### Changes ###
   - [ ] Feature1, tests, (and when applicable, API doc)
   - [ ] Feature2, tests, (and when applicable, API doc)
   
   ## Comments ##
   - If this change is a backward incompatible change, why must this change be 
made.
   - Interesting edge cases to note here
   



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] mxnet-bot commented on pull request #18257: Fixing #16655

[GitBox]

mxnet-bot commented on pull request #18257:
URL: https://github.com/apache/incubator-mxnet/pull/18257#issuecomment-625383183


   Hey @JinboCi , Thanks for submitting the PR 
   All tests are already queued to run once. If tests fail, you can trigger one 
or more tests again with the following commands: 
   - To trigger all jobs: @mxnet-bot run ci [all] 
   - To trigger specific jobs: @mxnet-bot run ci [job1, job2] 
   *** 
   **CI supported jobs**: [website, edge, sanity, unix-cpu, centos-cpu, clang, 
windows-cpu, unix-gpu, miscellaneous, windows-gpu, centos-gpu]
   *** 
   _Note_: 
Only following 3 categories can trigger CI :PR Author, MXNet Committer, 
Jenkins Admin. 
   All CI tests must pass before the PR can be merged. 
   



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ChaiBapchya commented on a change in pull request #18186: Update unix gpu toolchain

[GitBox]

ChaiBapchya commented on a change in pull request #18186:
URL: https://github.com/apache/incubator-mxnet/pull/18186#discussion_r421651875



##
File path: tests/python/unittest/test_numpy_op.py
##
@@ -2220,6 +,7 @@ def hybrid_forward(self, F, x):
 assert same(ret_mx.asnumpy(), ret_np)
 
 
+@unittest.skip("Flaky test 
https://github.com/apache/incubator-mxnet/issues/17840;)

Review comment:
   Alright, removing all the flaky test skips then





This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ChaiBapchya commented on pull request #17959: [MKLDNN] Add LSTMP to v1.6.x

[GitBox]

ChaiBapchya commented on pull request #17959:
URL: https://github.com/apache/incubator-mxnet/pull/17959#issuecomment-625373529


   @TaoLv Makes sense. Thanks for the clarification. Let's circle back to this 
point if rc1 is required.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] flygragon commented on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

[GitBox]

flygragon commented on issue #13773:
URL: 
https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625364871


   > is 7.3.8 your default gcc version? Or you can try downgrade your gcc and 
g++.
   > 
   > And check if there is enough swap memory.
   
   My gcc version is 7.5.0, RAM is 8G. I've tried again without graphic 
desktop, but it didn't work reporting the error of out of memory. I'm goting to 
try to downgrade the gcc version and enlarge the swap file.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] kohillyang edited a comment on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

[GitBox]

kohillyang edited a comment on issue #13773:
URL: 
https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625356733


   is 7.3.8 your default gcc version? Or you can try downgrade your gcc and g++.
   
   And check if there is enough swap memory.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] leezu commented on issue #18255: MKL_USE_STATIC_LIBS broken

[GitBox]

leezu commented on issue #18255:
URL: 
https://github.com/apache/incubator-mxnet/issues/18255#issuecomment-625357850


   mkl is used if available by default. 



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] kohillyang commented on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

[GitBox]

kohillyang commented on issue #13773:
URL: 
https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625356733


   is 7.3.8 your default gcc version? Or you can try downgrade your gcc and g++.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] flygragon commented on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

[GitBox]

flygragon commented on issue #13773:
URL: 
https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625351814


   > > Howerver, today I encountered the same issue, and always with the same 
file, after a carton of the machine. I was working with Ubuntu 18.04 with 8G 
mem. Could some one give me a way to solve it?
   > > Error msg is following.
   > > Building CXX object 
CMakeFiles/mxnet_static.dir/src/operator/tensor/indexing_op.cc.o
   > > c++: internal compiler error: killed (program cc1plus)
   > > Please submit a full bug report,
   > > with preprocessed source if appropriate.
   > 
   > Please try "make -j1" instead of "make -j".
   
   I've tried to reboot the OS and make -j1,but it doesn't work. What can I do ?



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] kohillyang commented on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

[GitBox]

kohillyang commented on issue #13773:
URL: 
https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625348429


   > Howerver, today I encountered the same issue, and always with the same 
file, after a carton of the machine. I was working with Ubuntu 18.04 with 8G 
mem. Could some one give me a way to solve it?
   > Error msg is following.
   > Building CXX object 
CMakeFiles/mxnet_static.dir/src/operator/tensor/indexing_op.cc.o
   > c++: internal compiler error: killed (program cc1plus)
   > Please submit a full bug report,
   > with preprocessed source if appropriate.
   Please try "make -j1" instead of "make -j".



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] kohillyang edited a comment on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

[GitBox]

kohillyang edited a comment on issue #13773:
URL: 
https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625348429


   > Howerver, today I encountered the same issue, and always with the same 
file, after a carton of the machine. I was working with Ubuntu 18.04 with 8G 
mem. Could some one give me a way to solve it?
   > Error msg is following.
   > Building CXX object 
CMakeFiles/mxnet_static.dir/src/operator/tensor/indexing_op.cc.o
   > c++: internal compiler error: killed (program cc1plus)
   > Please submit a full bug report,
   > with preprocessed source if appropriate.
   
   Please try "make -j1" instead of "make -j".



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] flygragon edited a comment on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

[GitBox]

flygragon edited a comment on issue #13773:
URL: 
https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625344577


   Howerver, today I encountered the same issue, and always with the same file, 
after a carton of the machine. I was working with Ubuntu 18.04 with 8G mem. 
Could some one give me a way to solve it?
   Error msg is following.
   Building CXX object 
CMakeFiles/mxnet_static.dir/src/operator/tensor/indexing_op.cc.o
   c++: internal compiler error: killed (program cc1plus)
   Please submit a full bug report,
   with preprocessed source if appropriate.
   



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] flygragon commented on issue #13773: Compiler error with g++ (Ubuntu 7.3.0-27ubuntu1~18.04) 7.3.0

[GitBox]

flygragon commented on issue #13773:
URL: 
https://github.com/apache/incubator-mxnet/issues/13773#issuecomment-625344577


   Howerver, today I encountered the same issue, and always with the same file, 
after a carton of the machine. I was working with Ubuntu 18.04 with 8G mem. 
Could some one give me a way to solve it?
   Building CXX object 
CMakeFiles/mxnet_static.dir/src/operator/tensor/indexing_op.cc.o
   c++: internal compiler error: killed (program cc1plus)
   Please submit a full bug report,
   with preprocessed source if appropriate.
   



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] TaoLv commented on pull request #17959: [MKLDNN] Add LSTMP to v1.6.x

[GitBox]

TaoLv commented on pull request #17959:
URL: https://github.com/apache/incubator-mxnet/pull/17959#issuecomment-625330433


   @ChaiBapchya @ciyongch As far as I know, the feature lives on 1.6.x branch 
but is not released in any release yet. Given the code of 1.7.0 has been frozen 
for a while and to pay respect to the effort from release manager, I would 
suggest to not include this feature in the coming rc0. We can reconsider the 
request if we will have another rc1. The pined version of oneDNN is also a 
valid concern.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] mikeobr commented on issue #16173: Saving and loading cudNN autotune and graph optimization

[GitBox]

mikeobr commented on issue #16173:
URL: 
https://github.com/apache/incubator-mxnet/issues/16173#issuecomment-625302116


   @QueensGambit I'm a bit of a noob level understanding around the internals, 
but if that allows us to load and start doing inference without taking the 
tuning hit, then it will help us out. 
   
   Generally my team generally uses models as SymbolBlocks and load them like `
   `gluon.SymbolBlock.imports(...``, so if we're able use an imported executor 
alongside our params and symbols, we're good.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] TaoLv commented on issue #18255: MKL_USE_STATIC_LIBS broken

[GitBox]

TaoLv commented on issue #18255:
URL: 
https://github.com/apache/incubator-mxnet/issues/18255#issuecomment-625295428


   Hi @leezu, I would expect the flag `-DMKL_USE_STATIC_LIBS` should take 
effect only when `-DUSE_BLAS` is set to `mkl`. That means, with your command 
line, mkl libraries should not be linked at all.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] kmfrick edited a comment on issue #18057: Edge device documentation

[GitBox]

kmfrick edited a comment on issue #18057:
URL: 
https://github.com/apache/incubator-mxnet/issues/18057#issuecomment-625291919


   Piggybacking on this: the pre-built library linked [here, section "Install 
MXNet for Python"](https://mxnet.apache.org/get_started/jetson_setup) is 
outdated  as JetPack 4.3 bundles CUDA 10.2 and that library is linked against 
CUDA 10.0 so as soon as you run any script that uses MX you get linker errors. 
   
   Right now I'm building branch 1.4.x from source on my Jetson Nano, as well 
as cross-compiling master on my laptop; let me know if there is anything I can 
do to help.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] kmfrick edited a comment on issue #18057: Edge device documentation

[GitBox]

kmfrick edited a comment on issue #18057:
URL: 
https://github.com/apache/incubator-mxnet/issues/18057#issuecomment-625291919


   Piggybacking on this: the pre-built library linked [here, section "Install 
MXNet for Python"](https://mxnet.apache.org/get_started/jetson_setup) is 
outdated  as JetPack 4.3 bundles CUDA 10.2 and that library is linked against 
CUDA 10.0 so as soon as you run any script that uses MX you get linker errors. 
   
   Right now I'm building branch 1.4.x from source on my Jetson Nano, let me 
know if there is anything I can do to help.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] kmfrick commented on issue #18057: Edge device documentation

[GitBox]

kmfrick commented on issue #18057:
URL: 
https://github.com/apache/incubator-mxnet/issues/18057#issuecomment-625291919


   Piggybacking on this: the pre-built library linked [here, section "Install 
MXNet for Python"](https://mxnet.apache.org/get_started/jetson_setup) is 
outdated  as JetPack 4.3 bundles CUDA 10.2 and that library is linked against 
CUDA 10.0 so as soon as you run any script that uses MX you get linker errors. 
Right now I'm building from source on my Jetson (branch v1.4.x), let me know if 
there is anything I can do to help.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.

[aaronmarkham]

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 edf7f6a  Bump the publish timestamp.
edf7f6a is described below

commit edf7f6acc52fdf3720d59ece651f872b0ff6dfa1
Author: mxnet-ci 
AuthorDate: Thu May 7 12:48:29 2020 +

Bump the publish timestamp.
---
 date.txt | 1 +
 1 file changed, 1 insertion(+)

diff --git a/date.txt b/date.txt
new file mode 100644
index 000..e0cf04a
--- /dev/null
+++ b/date.txt
@@ -0,0 +1 @@
+Thu May  7 12:48:29 UTC 2020

[GitHub] [incubator-mxnet] ciyongch commented on pull request #17959: [MKLDNN] Add LSTMP to v1.6.x

[GitBox]

ciyongch commented on pull request #17959:
URL: https://github.com/apache/incubator-mxnet/pull/17959#issuecomment-625112554


   >Backport to 1.6 should be included by default in 1.7 right?
   
   Hi @ChaiBapchya , not sure if it's a common practice to do so. As there're 
several new features including (oneDNN) FP32 LSTMP/INT8 LSTM/INT8 LSTMP 
introduced in v1.6.x branch, are they all needed to go into 1.7.0 or only FP32 
LSTMP is good enough? As currently INT8 LSTMP relies on non-release version of 
oneDNN which might be involved till it's matured and included in a formal 
release version. Previously MXNet is always pin to a formal release version of 
oneDNN.
   cc @TaoLv to see if there's any other comments.
   Thanks,
   Ciyong.
   
   
   



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] pasa13142 commented on issue #15553: Installing mxnet on PowerPC

[GitBox]

pasa13142 commented on issue #15553:
URL: 
https://github.com/apache/incubator-mxnet/issues/15553#issuecomment-625089795


   @leezu @Talgin no idea guys how can i fix this ? :(



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] pasa13142 edited a comment on issue #15553: Installing mxnet on PowerPC

[GitBox]

pasa13142 edited a comment on issue #15553:
URL: 
https://github.com/apache/incubator-mxnet/issues/15553#issuecomment-625089795


   @leezu @Talgin no idea how can i fix this ? :(



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ChaiBapchya commented on pull request #17959: [MKLDNN] Add LSTMP to v1.6.x

[GitBox]

ChaiBapchya commented on pull request #17959:
URL: https://github.com/apache/incubator-mxnet/pull/17959#issuecomment-625084536


   @ciyongch While the customer request is resolved with 1.6.x backport, it 
made sense to include it in upcoming minor release. Backport to 1.6 should be 
included by default in 1.7 right?



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] leezu commented on issue #18184: Flaky Incorrect Results test_numpy_op.py:: test_np_linalg_slogdet

[GitBox]

leezu commented on issue #18184:
URL: 
https://github.com/apache/incubator-mxnet/issues/18184#issuecomment-625081994


   Did you try on G3 instances? The other question is why the `rtol=atol=1e-1` 
and not lower



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.

[aaronmarkham]

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 93fe491  Bump the publish timestamp.
93fe491 is described below

commit 93fe491b983913bb9df96dda86e6ecdfca8051a3
Author: mxnet-ci 
AuthorDate: Thu May 7 06:48:03 2020 +

Bump the publish timestamp.
---
 date.txt | 1 +
 1 file changed, 1 insertion(+)

diff --git a/date.txt b/date.txt
new file mode 100644
index 000..e9b6b66
--- /dev/null
+++ b/date.txt
@@ -0,0 +1 @@
+Thu May  7 06:48:03 UTC 2020

[GitHub] [incubator-mxnet] yzhliu commented on issue #16357: Numpy Operator Inconsistency

[GitBox]

yzhliu commented on issue #16357:
URL: 
https://github.com/apache/incubator-mxnet/issues/16357#issuecomment-625062177


   @Yiyan66 is looking into it.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] ciyongch commented on pull request #17959: [MKLDNN] Add LSTMP to v1.6.x

[GitBox]

ciyongch commented on pull request #17959:
URL: https://github.com/apache/incubator-mxnet/pull/17959#issuecomment-625061521


   I suggest to leave this new feature to the next release version (which 
requires upgrade oneDNN (ex: DNNL)), as the current version 1.7.0 is code 
freeze and suppose to only accept bug fixes, besides this feature is also 
included in master branch. The reason for applying it in 1.6.x branch is it's 
required by customer. 
   @ChaiBapchya is there any special requirement for this feature in 1.7.0?



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] sxjscience edited a comment on issue #18256: put more effort over 'easy to use'

[GitBox]

sxjscience edited a comment on issue #18256:
URL: 
https://github.com/apache/incubator-mxnet/issues/18256#issuecomment-625050992


   @leezu Previously, we have one benchmark by @Jerryzcn : 
https://github.com/apache/incubator-mxnet/issues/17335 . According to my recent 
coding experience with the new DeepNumpy interface of MXNet, I do recognize 
that the usability of Gluon HybridBlock is worse than the PyTorch Module. When 
implementing some slightly more complicated structures like TransformerXL or 
Transformer for NMT, it's easy to write a PyTorch-based implementation that can 
fully utilize the GPUs. On the other hand, it is not so trivial to do so in 
MXNet. One observation is that the MXNet program will occupy lots of CPU cores. 
I'm still trying to figure out the cause.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] sxjscience edited a comment on issue #18256: put more effort over 'easy to use'

[GitBox]

sxjscience edited a comment on issue #18256:
URL: 
https://github.com/apache/incubator-mxnet/issues/18256#issuecomment-625050992


   @leezu Previously, we have one benchmark by @Jerryzcn : 
https://github.com/apache/incubator-mxnet/issues/17335 . According to my recent 
coding experience with the new DeepNumpy interface of MXNet, I do recognize 
that the usability of Gluon HybridBlock is worse than the PyTorch Module. When 
implementing some slightly more advanced networks like TransformerXL, it's easy 
to write a PyTorch-based version that can fully utilize the GPU. On the other 
hand, it is not so trivial to do so in MXNet. I'm still trying to figure out 
the cause.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] sxjscience commented on issue #18256: put more effort over 'easy to use'

[GitBox]

sxjscience commented on issue #18256:
URL: 
https://github.com/apache/incubator-mxnet/issues/18256#issuecomment-625050992


   @leezu Previously, we have one benchmark by @Jerryzcn : 
https://github.com/apache/incubator-mxnet/issues/17335 . According to my recent 
coding experience with the new DeepNumpy interface of MXNet, I do recognize 
that the usability of Gluon HybridBlock is worse than the Module of PyTorch. 
When implementing some slightly more advanced networks like TransformerXL, it's 
easy to write a PyTorch-based version that can fully utilize the GPU. On the 
other hand, it is not so trivial to do so in MXNet. I'm still trying to figure 
out the cause.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] vexilligera edited a comment on issue #18184: Flaky Incorrect Results test_numpy_op.py:: test_np_linalg_slogdet

[GitBox]

vexilligera edited a comment on issue #18184:
URL: 
https://github.com/apache/incubator-mxnet/issues/18184#issuecomment-625047127


   Hi, I tried to reproduce the error by running
   `pytest tests/python/gpu/test_operator_gpu.py::test_np_linalg_slogdet`
   for 20 times and didn't run into the error.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] vexilligera commented on issue #18184: Flaky Incorrect Results test_numpy_op.py:: test_np_linalg_slogdet

[GitBox]

vexilligera commented on issue #18184:
URL: 
https://github.com/apache/incubator-mxnet/issues/18184#issuecomment-625047127


   Hi, I tried to reproduce the error by running
   `pytest tests/python/gpu/test_operator_gpu.py::test_np_linalg_slogdet`
   for 20 times both on CPU and GPU, and didn't run into the error.



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

[GitHub] [incubator-mxnet] leezu commented on issue #18256: put more effort over 'easy to use'

[GitBox]

leezu commented on issue #18256:
URL: 
https://github.com/apache/incubator-mxnet/issues/18256#issuecomment-625046431


   I agree that many use-cases can be covered in a simpler fashion without 
hybridization. One question here is if "unhybridized mxnet" may be slower than 
"hybridized mxnet" and only the latter competitive with pytorch. I think 
@sxjscience is preparing some benchmarks. 



This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org