zheng-da commented on issue #12380: Manually track num_max_thread
URL: https://github.com/apache/incubator-mxnet/pull/12380#issuecomment-441422070
This PR is problematic. This causes significant performance regression in
spmv in the multi-GPU training. It seems the PR affects the number of
eric-haibin-lin opened a new pull request #13398: Use dynamic omp schedule for
sparse dot with large matrix
URL: https://github.com/apache/incubator-mxnet/pull/13398
## Description ##
The default kernel launch method doesn't use the dynamic openmp scheduling
primitive, which suffers
This is an automated email from the ASF dual-hosted git repository.
zhasheng 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 5f1447e Bump the publish
dbsxdbsx opened a new issue #13397: Gluon: name issue about example"Actor
Critic"
URL: https://github.com/apache/incubator-mxnet/issues/13397
I am referring to the [gluon example:actor
vdantu commented on issue #13395: Cannot import alexnet model from pytorch with
ONNX
URL:
https://github.com/apache/incubator-mxnet/issues/13395#issuecomment-441416516
@mxnet-label-bot add [question, onnx]
This is an
vdantu commented on issue #13391: Gluon: Per-layer learning rate for fine
tuning a pretrained network
URL:
https://github.com/apache/incubator-mxnet/issues/13391#issuecomment-441416339
@mxnet-label-bot add [gluon, question]
vdantu commented on issue #13389: how to carry out unit test on essential
classes
URL:
https://github.com/apache/incubator-mxnet/issues/13389#issuecomment-441416259
@mxnet-label-bot add [c++, question, test]
This is an
vdantu edited a comment on issue #13381: [Feature Request] Support CSRNDArray
in kv_store
URL:
https://github.com/apache/incubator-mxnet/issues/13381#issuecomment-441415943
@mxnet-label-bot add [KVStore, Feature request, Question]
vdantu commented on issue #13381: [Feature Request] Support CSRNDArray in
kv_store
URL:
https://github.com/apache/incubator-mxnet/issues/13381#issuecomment-441415943
@mxnet-label-bot add [KVStore, Feature request]
This is
marcoabreu commented on a change in pull request #13310: [MXNET-703] Update to
TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#discussion_r236059363
##
File path: src/operator/contrib/nnvm_to_onnx.cc
##
@@ -78,8 +84,13
marcoabreu commented on issue #13310: [MXNET-703] Update to TensorRT 5, ONNX IR
3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#issuecomment-441411854
For the record:
marcoabreu commented on issue #13310: [MXNET-703] Update to TensorRT 5, ONNX IR
3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#issuecomment-441411835
Feel free to go ahead at your discretion. Can you drop me an email to
investigate why your run for
marcoabreu commented on a change in pull request #13310: [MXNET-703] Update to
TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#discussion_r236059280
##
File path: src/operator/contrib/nnvm_to_onnx.cc
##
@@ -248,8
BullDemonKing opened a new pull request #13396: add adjacency operator.
URL: https://github.com/apache/incubator-mxnet/pull/13396
## Description ##
The adjacency operator is implemented for DGL that converts a CSR matrix
with edge Id to an adjacency matrix for SpMV.
## Checklist
pengzhao-intel commented on issue #11658: Optimize cached op static memory
allocation
URL: https://github.com/apache/incubator-mxnet/pull/11658#issuecomment-441411422
@ZhennanQin to aware the changes since we are WIP for Cached OP too.
KellenSunderland edited a comment on issue #13310: [MXNET-703] Update to
TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#issuecomment-441409851
@marcoabreu The windows GPU tests are continually failing with softmin
errors. Should we
KellenSunderland edited a comment on issue #13310: [MXNET-703] Update to
TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#issuecomment-441409851
@marcoabreu The windows GPU tests are continually failing with softmin
errors. Should we
KellenSunderland commented on issue #13310: [MXNET-703] Update to TensorRT 5,
ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#issuecomment-441409851
@marcoabreu The windows GPU tests are continually failing with softmin
errors. Should we ignore
KellenSunderland commented on a change in pull request #13310: [MXNET-703]
Update to TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#discussion_r236058587
##
File path: src/operator/contrib/nnvm_to_onnx.cc
##
@@ -78,8
KellenSunderland commented on a change in pull request #13310: [MXNET-703]
Update to TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#discussion_r236058587
##
File path: src/operator/contrib/nnvm_to_onnx.cc
##
@@ -78,8
KellenSunderland commented on a change in pull request #13310: [MXNET-703]
Update to TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#discussion_r236058512
##
File path: src/operator/contrib/nnvm_to_onnx.cc
##
@@ -248,8
KellenSunderland commented on a change in pull request #13310: [MXNET-703]
Update to TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#discussion_r236058512
##
File path: src/operator/contrib/nnvm_to_onnx.cc
##
@@ -248,8
marcoabreu commented on a change in pull request #13310: [MXNET-703] Update to
TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#discussion_r236058517
##
File path: ci/docker/Dockerfile.build.ubuntu_gpu_tensorrt
##
@@
KellenSunderland commented on a change in pull request #13310: [MXNET-703]
Update to TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#discussion_r236058494
##
File path: ci/docker/Dockerfile.build.ubuntu_gpu_tensorrt
##
marcoabreu commented on a change in pull request #13310: [MXNET-703] Update to
TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#discussion_r236058383
##
File path: src/operator/contrib/nnvm_to_onnx.cc
##
@@ -248,8
marcoabreu commented on a change in pull request #13310: [MXNET-703] Update to
TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#discussion_r236058327
##
File path: ci/docker/Dockerfile.build.ubuntu_gpu_tensorrt
##
@@
marcoabreu commented on a change in pull request #13310: [MXNET-703] Update to
TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#discussion_r236058387
##
File path: src/operator/contrib/nnvm_to_onnx.cc
##
@@ -382,31
marcoabreu commented on a change in pull request #13310: [MXNET-703] Update to
TensorRT 5, ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#discussion_r236058362
##
File path: src/operator/contrib/nnvm_to_onnx.cc
##
@@ -78,8 +84,13
This is an automated email from the ASF dual-hosted git repository.
zhasheng 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 232b6b2 Bump the publish
ieee8023 opened a new issue #13395: Cannot import alexnet model from pytorch
with ONNX
URL: https://github.com/apache/incubator-mxnet/issues/13395
I cannot import the alexnet model from pytorch. I receive this error:
```
File "test.py", line 5, in
sym, arg, aux =
This is an automated email from the ASF dual-hosted git repository.
zhasheng 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 f5eb47f Bump the publish
anirudhacharya commented on a change in pull request #13390: Onnx multi output
URL: https://github.com/apache/incubator-mxnet/pull/13390#discussion_r236052117
##
File path: tests/python/unittest/test_onnx.py
##
@@ -0,0 +1,95 @@
+# Licensed to the Apache Software Foundation
anirudhacharya commented on a change in pull request #13390: Onnx multi output
URL: https://github.com/apache/incubator-mxnet/pull/13390#discussion_r236052119
##
File path: python/mxnet/contrib/onnx/mx2onnx/export_onnx.py
##
@@ -242,6 +242,14 @@ def
KellenSunderland closed pull request #13394: [DO NOT MERGE] Install CUDA 10
compatible cmake
URL: https://github.com/apache/incubator-mxnet/pull/13394
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of
gigasquid commented on issue #13388: [Clojure] - Add Clojure to the main
website docs install
URL:
https://github.com/apache/incubator-mxnet/issues/13388#issuecomment-441388585
Awesome! Thank you
This is an automated
gigasquid commented on issue #13393: Fixes #13388 Adds Clojure to MXNet
installation docs
URL: https://github.com/apache/incubator-mxnet/pull/13393#issuecomment-441388503
Awesome! Thanks for this contribution. It will help Clojure users to get
started much easier than before. 磊
This is an automated email from the ASF dual-hosted git repository.
zhasheng 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 e94 Bump the publish
KellenSunderland commented on issue #13389: how to carry out unit test on
essential classes
URL:
https://github.com/apache/incubator-mxnet/issues/13389#issuecomment-441387974
Agree with Marco's points, but just wanted to mention that we're certainly
open to adding C++ tests for classes
KellenSunderland opened a new pull request #13394: [DO NOT MERGE][MXNET-703]
Install CUDA 10 compatible cmake
URL: https://github.com/apache/incubator-mxnet/pull/13394
Triggering build for aba34 to verify atomicity.
This is
KellenSunderland commented on issue #13310: [MXNET-703] Update to TensorRT 5,
ONNX IR 3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#issuecomment-441387052
@NRauschmayr would appreciate a review.
KellenSunderland commented on issue #13310: [MXNET-703] Update TensorRT to v5
ONNX IR to v3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#issuecomment-441386047
In fact I'm going to rework the commit history here such that we have a
single atomic commit
KellenSunderland edited a comment on issue #13310: [MXNET-703] Update TensorRT
to v5 ONNX IR to v3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#issuecomment-441385511
@marcoabreu Would appreciate it if you could have a quick look at the
Jenkins
KellenSunderland edited a comment on issue #13310: [MXNET-703] Update TensorRT
to v5 ONNX IR to v3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#issuecomment-441385511
@marcoabreu Would appreciate it if you could have a quick look at the
Jenkins
KellenSunderland commented on issue #13310: [MXNET-703] Update TensorRT to v5
ONNX IR to v3. Fix inference bugs.
URL: https://github.com/apache/incubator-mxnet/pull/13310#issuecomment-441385511
@marcoabreu Would appreciate it if you could have a quick look at the
Jenkins modifications
anujsrc commented on issue #13388: [Clojure] - Add Clojure to the main website
docs install
URL:
https://github.com/apache/incubator-mxnet/issues/13388#issuecomment-441385205
@gigasquid - I have submitted
https://github.com/apache/incubator-mxnet/pull/13393 and in the description, I
anujsrc opened a new pull request #13393: Fixes #13388 Adds Clojure to MXNet
installation docs
URL: https://github.com/apache/incubator-mxnet/pull/13393
## Description ##
Adds Clojure installation docs to MXNet website
## Checklist ##
### Essentials ###
Please feel free to
This is an automated email from the ASF dual-hosted git repository.
anirudh2290 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 3d64d15 Fix/env disable mkldnn
anirudh2290 closed pull request #13324: Fix/env disable mkldnn cache map
URL: https://github.com/apache/incubator-mxnet/pull/13324
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As this is a
lanking520 commented on a change in pull request #13289: Initial website
documentation for Java API
URL: https://github.com/apache/incubator-mxnet/pull/13289#discussion_r236049060
##
File path: docs/install/java_setup.md
##
@@ -0,0 +1,116 @@
+# Setup the MXNet Package for
lanking520 commented on issue #13105: [MXNET-1158] JVM Memory Management
Documentation
URL: https://github.com/apache/incubator-mxnet/pull/13105#issuecomment-441382292
Hi @nswamy could you please address the change and let's get this PR forward
marcoabreu commented on issue #13380: mxnet bind fails in nvidia jetson tx2
(jetpack 3.3, cuda 9.0)
URL:
https://github.com/apache/incubator-mxnet/issues/13380#issuecomment-441382216
Great news. Would you mind explaining what you had to change so other users
who face the same issue know
lanking520 commented on issue #13368: Java doc
URL: https://github.com/apache/incubator-mxnet/pull/13368#issuecomment-441382200
@zachgk Could you please trigger the CI test again?
This is an automated message from the Apache
lanking520 closed pull request #13361: Updated the paths for images for java
tutorial
URL: https://github.com/apache/incubator-mxnet/pull/13361
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As
lanking520 commented on issue #13361: Updated the paths for images for java
tutorial
URL: https://github.com/apache/incubator-mxnet/pull/13361#issuecomment-441381968
Thanks for your contribution @piyushghai , now merge it
This is an automated email from the ASF dual-hosted git repository.
lanking 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 6b5cf4f Updated the paths for images
safrooze commented on a change in pull request #13390: Onnx multi output
URL: https://github.com/apache/incubator-mxnet/pull/13390#discussion_r236046096
##
File path: python/mxnet/contrib/onnx/mx2onnx/export_onnx.py
##
@@ -242,6 +242,14 @@ def create_onnx_graph_proto(self,
safrooze commented on a change in pull request #13390: Onnx multi output
URL: https://github.com/apache/incubator-mxnet/pull/13390#discussion_r236046096
##
File path: python/mxnet/contrib/onnx/mx2onnx/export_onnx.py
##
@@ -242,6 +242,14 @@ def create_onnx_graph_proto(self,
safrooze commented on a change in pull request #13390: Onnx multi output
URL: https://github.com/apache/incubator-mxnet/pull/13390#discussion_r236046200
##
File path: tests/python/unittest/test_onnx.py
##
@@ -0,0 +1,95 @@
+# Licensed to the Apache Software Foundation (ASF)
safrooze commented on a change in pull request #13390: Onnx multi output
URL: https://github.com/apache/incubator-mxnet/pull/13390#discussion_r236046096
##
File path: python/mxnet/contrib/onnx/mx2onnx/export_onnx.py
##
@@ -242,6 +242,14 @@ def create_onnx_graph_proto(self,
safrooze commented on a change in pull request #13390: Onnx multi output
URL: https://github.com/apache/incubator-mxnet/pull/13390#discussion_r236046096
##
File path: python/mxnet/contrib/onnx/mx2onnx/export_onnx.py
##
@@ -242,6 +242,14 @@ def create_onnx_graph_proto(self,
muhyun commented on issue #13380: mxnet bind fails in nvidia jetson tx2
(jetpack 3.3, cuda 9.0)
URL:
https://github.com/apache/incubator-mxnet/issues/13380#issuecomment-441375204
I found that it is caused by mis-configured JetPack. So, this can be closed
as resolved.
aksnzhy opened a new pull request #13392: add csr sample op
URL: https://github.com/apache/incubator-mxnet/pull/13392
This pr add the graph sampling op, which will be used in the DGL project.
@zheng-da Please review this pr. Thanks!
For example:
```
shape = (5, 5)
pengzhao-intel commented on a change in pull request #12922: Support Quantized
Fully Connected by INT8 GEMM
URL: https://github.com/apache/incubator-mxnet/pull/12922#discussion_r236042654
##
File path: src/operator/quantization/quantized_fully_connected.cc
##
@@ -79,6
MXLHELLO commented on issue #6785: Error importing mxnet in python, installed
using pip
URL:
https://github.com/apache/incubator-mxnet/issues/6785#issuecomment-441366725
> Hi, I suppose you have cuda 8.0 and used the latest version of mxnet.
> I ran into the same problem, and it turns
MXLHELLO removed a comment on issue #6785: Error importing mxnet in python,
installed using pip
URL:
https://github.com/apache/incubator-mxnet/issues/6785#issuecomment-441366687
> 嗨,我想你有cuda 8.0并使用了最新版本的mxnet。
> 我遇到了同样的问题,结果发现cublasCgemm3mStridedBatched是cuda
MXLHELLO commented on issue #6785: Error importing mxnet in python, installed
using pip
URL:
https://github.com/apache/incubator-mxnet/issues/6785#issuecomment-441366687
> 嗨,我想你有cuda 8.0并使用了最新版本的mxnet。
> 我遇到了同样的问题,结果发现cublasCgemm3mStridedBatched是cuda
This is an automated email from the ASF dual-hosted git repository.
zhasheng 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 0375335 Bump the publish
sxjscience commented on issue #12821: [MXNET-1095] [WIP] Support higher order
gradient for cos,sin,negative,mul
URL: https://github.com/apache/incubator-mxnet/pull/12821#issuecomment-441362898
I'll continue working on it next week.
marcoabreu commented on a change in pull request #13390: Onnx multi output
URL: https://github.com/apache/incubator-mxnet/pull/13390#discussion_r236039991
##
File path: tests/python/unittest/test_onnx.py
##
@@ -0,0 +1,95 @@
+# Licensed to the Apache Software Foundation
marcoabreu commented on issue #13389: how to carry out unit test on essential
classes
URL:
https://github.com/apache/incubator-mxnet/issues/13389#issuecomment-441361814
Hello,
that's correct. We mainly use Python tests to validate our backend. We also
have some C++ tests, but for
JWarlock opened a new issue #13391: Gluon: Per-layer learning rate for fine
tuning a pretrained network
URL: https://github.com/apache/incubator-mxnet/issues/13391
Hi!
I want to train a pretrained model, for example, ResNet, where I want to
limit the first two layers’ learning rate to
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 79857a4 Fixes #13386 - Refer
marcoabreu closed pull request #13387: Fixes #13386 - Refer Warnings
URL: https://github.com/apache/incubator-mxnet/pull/13387
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As this is a foreign
marcoabreu closed issue #13386: [Clojure] - Remove refer warning
URL: https://github.com/apache/incubator-mxnet/issues/13386
This is an automated message from the Apache Git Service.
To respond to the message, please log on
jasonyu1996 commented on issue #12484: [Feature Request] Assertion operators
URL:
https://github.com/apache/incubator-mxnet/issues/12484#issuecomment-441359853
It seems that this is indeed a feature in need. Could you please provide
some instructions on how this should be implemented?
yemenr commented on issue #12484: [Feature Request] Assertion operators
URL:
https://github.com/apache/incubator-mxnet/issues/12484#issuecomment-441351108
@jasonyu1996 Same question...
This is an automated message from the
lupesko commented on issue #13138: MXNet crashes while trying load ONNX model
URL:
https://github.com/apache/incubator-mxnet/issues/13138#issuecomment-441350927
Was able to reproduce with latest MXNet (1.4.0b20181123) and ONNX 1.3.0.
Bouncing this @Roshrini @vandanavk can you guys
77 matches
Mail list logo