ThomasDelteil commented on a change in pull request #10900: [MXNET-414]
Tutorial on visualizing CNN decisions using Grad-CAM
URL: https://github.com/apache/incubator-mxnet/pull/10900#discussion_r188721229
##
File path: docs/tutorials/vision/cnn_visualization.md
##
@@ -0,0
ThomasDelteil commented on a change in pull request #10900: [MXNET-414]
Tutorial on visualizing CNN decisions using Grad-CAM
URL: https://github.com/apache/incubator-mxnet/pull/10900#discussion_r188719470
##
File path: docs/tutorials/vision/cnn_visualization.md
##
@@ -0,0
ThomasDelteil commented on a change in pull request #10900: [MXNET-414]
Tutorial on visualizing CNN decisions using Grad-CAM
URL: https://github.com/apache/incubator-mxnet/pull/10900#discussion_r188723657
##
File path: docs/tutorials/vision/cnn_visualization.md
##
@@ -0,0
ThomasDelteil commented on a change in pull request #10900: [MXNET-414]
Tutorial on visualizing CNN decisions using Grad-CAM
URL: https://github.com/apache/incubator-mxnet/pull/10900#discussion_r188717193
##
File path: docs/tutorials/vision/cnn_visualization.md
##
@@ -0,0
zhanghang1989 commented on issue #10976: [BUG] MNIST LeNet forward fails on
master branch
URL:
https://github.com/apache/incubator-mxnet/issues/10976#issuecomment-389616700
Failed on commit 65061dc93710afce92edfe548aa3352473da3cdb,
but seems fixed on master now.
chaoyuaw commented on issue #10950: The negative samples shall be other classes?
URL:
https://github.com/apache/incubator-mxnet/issues/10950#issuecomment-389612204
@jonbakerfish Yes, I think you are right. Would you mind creating a PR for
this?
larroy opened a new pull request #10975: avoid importing docker_cache if
feature is not used, so boto3 and job…
URL: https://github.com/apache/incubator-mxnet/pull/10975
…lib are not needed
## Description ##
(see title)
## Checklist ##
### Essentials ###
Please feel
eric-haibin-lin commented on issue #10950: The negative samples shall be other
classes?
URL:
https://github.com/apache/incubator-mxnet/issues/10950#issuecomment-389604635
@chaoyuaw
This is an automated message from the
piiswrong commented on issue #10964: linalg_potri() Error: linalg_impl.h:606:
Check failed: ret == 0 (3 vs. 0) spotri failed in lapack on cpu.
URL:
https://github.com/apache/incubator-mxnet/issues/10964#issuecomment-389603387
@asmushetzel
eric-haibin-lin commented on issue #10969: Mxnet dynamic graphs achieve scoure
code
URL:
https://github.com/apache/incubator-mxnet/issues/10969#issuecomment-389602149
You can read the code in src/imperative folder, in particular related to
cached op for Gluon backend execution
eric-haibin-lin closed issue #10896:
test_sparse_operator.py:test_sparse_mathematical_core fails under scipy v1.1
URL: https://github.com/apache/incubator-mxnet/issues/10896
This is an automated message from the Apache Git
eric-haibin-lin commented on issue #10896:
test_sparse_operator.py:test_sparse_mathematical_core fails under scipy v1.1
URL:
https://github.com/apache/incubator-mxnet/issues/10896#issuecomment-389601341
Thanks for looking into this! Closed via
eric-haibin-lin commented on issue #10901: Broken
test_sparse_operator.test_sparse_mathematical_core with scipy 1.1.0
URL:
https://github.com/apache/incubator-mxnet/issues/10901#issuecomment-389601219
Fixed by https://github.com/apache/incubator-mxnet/pull/10961
eric-haibin-lin closed issue #10901: Broken
test_sparse_operator.test_sparse_mathematical_core with scipy 1.1.0
URL: https://github.com/apache/incubator-mxnet/issues/10901
This is an automated message from the Apache Git
piiswrong commented on a change in pull request #10957: Make inner transform
activation configurable for LSTMCell
URL: https://github.com/apache/incubator-mxnet/pull/10957#discussion_r188703571
##
File path: python/mxnet/gluon/rnn/rnn_cell.py
##
@@ -441,6 +441,9 @@ class
This is an automated email from the ASF dual-hosted git repository.
jxie 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 b845147 Fix
piiswrong closed pull request #10961: [MXNET-10896] [MXNET-10901] Fix
test_sparse_mathematical_core sensitivity to scipy v1.1
URL: https://github.com/apache/incubator-mxnet/pull/10961
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is
anirudhacharya commented on a change in pull request #10512: [MXNET-309]
[ONNX-MXNet] Model Metadata API
URL: https://github.com/apache/incubator-mxnet/pull/10512#discussion_r188696940
##
File path: python/mxnet/contrib/onnx/_import/import_onnx.py
##
@@ -132,6 +132,33 @@
anirudhacharya commented on a change in pull request #10512: [MXNET-309]
[ONNX-MXNet] Model Metadata API
URL: https://github.com/apache/incubator-mxnet/pull/10512#discussion_r188696940
##
File path: python/mxnet/contrib/onnx/_import/import_onnx.py
##
@@ -132,6 +132,33 @@
ThomasDelteil commented on issue #10962: CI not early stopping builds anymore
URL:
https://github.com/apache/incubator-mxnet/issues/10962#issuecomment-389588267
Ah, don't know how I came to be sure of that then. Well I'm thinking that
might be a good idea to do it in the future :)
ThomasDelteil closed issue #10962: CI not early stopping builds anymore
URL: https://github.com/apache/incubator-mxnet/issues/10962
This is an automated message from the Apache Git Service.
To respond to the message, please
marcoabreu commented on issue #10962: CI not early stopping builds anymore
URL:
https://github.com/apache/incubator-mxnet/issues/10962#issuecomment-389576164
Hi, this was never enabled. We always built all commits
This
marcoabreu commented on issue #10974: Cpp CI may be broken
URL:
https://github.com/apache/incubator-mxnet/issues/10974#issuecomment-389574263
GDB reports ```"During startup program terminated with signal SIGABRT,
Aborted."```. Unfortunately, I can't get more informations since the error
marcoabreu commented on issue #10974: Cpp CI may be broken
URL:
https://github.com/apache/incubator-mxnet/issues/10974#issuecomment-389532591
Confirmed. Steps to reproduce:
Instance: g3.8xlarge (don't forget nvidia-docker)
Build using
```
ci/build.py
wkcn commented on issue #10852: [MXNET-411] Add ROI Align
URL: https://github.com/apache/incubator-mxnet/pull/10852#issuecomment-389542747
Hi. I wrote a ROIAlign forward/backward test. It may be useful.
https://github.com/wkcn/MobulaOP/blob/master/tests/test_roi_align_op.py
And I
wkcn commented on issue #10852: [MXNET-411] Add ROI Align
URL: https://github.com/apache/incubator-mxnet/pull/10852#issuecomment-389542747
Hi. I wrote a ROIAlign forward/backward test. It may be useful.
https://github.com/wkcn/MobulaOP/blob/master/tests/test_roi_align_op.py
And I
marcoabreu commented on issue #10974: Cpp CI may be broken
URL:
https://github.com/apache/incubator-mxnet/issues/10974#issuecomment-389532591
Confirmed. Steps to reproduce:
Instance: g3.8xlarge (don't forget nvidia-docker)
Build using
```
ci/build.py
marcoabreu commented on issue #10974: Cpp CI may be broken
URL:
https://github.com/apache/incubator-mxnet/issues/10974#issuecomment-389532591
Confirmed. Steps to reproduce:
Instance: g3.8xlarge (don't forget nvidia-docker)
Build using
```ci/build.py --download-docker-cache
marcoabreu commented on issue #10974: Cpp CI may be broken
URL:
https://github.com/apache/incubator-mxnet/issues/10974#issuecomment-389532591
Confirmed. Steps to reproduce:
Instance: g3.8xlarge (don't forget nvidia-docker)
Build using
```ci/build.py --download-docker-cache
asitstands opened a new issue #10974: Cpp CI may be broken
URL: https://github.com/apache/incubator-mxnet/issues/10974
http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-10970/1/pipeline/717
asitstands opened a new issue #10973: Flaky test_deconvolution
URL: https://github.com/apache/incubator-mxnet/issues/10973
http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-10970/1/pipeline/724
```
liuzx32 commented on issue #10966: mx.kvstore.create('dist_sync') throw error.
Is there anything wrong with my environment?
URL:
https://github.com/apache/incubator-mxnet/issues/10966#issuecomment-389503364
Maybe mx.kvstore.create('dist_sync') can't run in python Shell.
liuzx32 closed issue #10966: mx.kvstore.create('dist_sync') throw error. Is
there anything wrong with my environment?
URL: https://github.com/apache/incubator-mxnet/issues/10966
This is an automated message from the Apache
threeleafzerg commented on issue #10696: [MXNET-366]Extend MXNet Distributed
Training by AllReduce
URL: https://github.com/apache/incubator-mxnet/pull/10696#issuecomment-389170346
@eric-haibin-lin
Hi haibin, I have already finished code modification according to your
comments. Any
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 8ebc050 [MXNET-416] Add docker
marcoabreu closed pull request #10917: [MXNET-416] Add docker cache
URL: https://github.com/apache/incubator-mxnet/pull/10917
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
rocketbear opened a new issue #10972: Gluon's PReLU is very slow and a fix to it
URL: https://github.com/apache/incubator-mxnet/issues/10972
I have experienced significantly slow training speed when using PReLU
activation instead of using ReLU activation with model composed using the gluon
QiXuanWang commented on issue #10805: SKIP RNN is incorrect in LSTnet
URL:
https://github.com/apache/incubator-mxnet/issues/10805#issuecomment-389487036
@diyang thanks very much for LSTnet-Attn mention. My current problem
actually could use this. Will try it out.
wythe51 commented on issue #5245: Training and validation accuracy very low in
Binary Classification
URL:
https://github.com/apache/incubator-mxnet/issues/5245#issuecomment-389466035
Maybe try `SoftmaxOutput` instead of `LogisticRegressionOutput` (and set
`num_hidden=2` ).
At least in
jinhuang415 commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r188569030
##
File path: include/mxnet/c_api.h
##
@@ -1423,13 +1423,15 @@ MXNET_DLL int
GSanchis closed issue #10971: Segmentation fault with custom operator
URL: https://github.com/apache/incubator-mxnet/issues/10971
This is an automated message from the Apache Git Service.
To respond to the message, please
GSanchis commented on issue #10971: Segmentation fault with custom operator
URL:
https://github.com/apache/incubator-mxnet/issues/10971#issuecomment-389456909
Ok, I believe I'm mixing things. Closing this for now, until I've understood
what's going on.
jinhuang415 commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r188534806
##
File path: src/operator/nn/mkldnn/mkldnn_convolution-inl.h
##
@@ -0,0 +1,100 @@
GSanchis opened a new issue #10971: Segmentation fault with custom operator
URL: https://github.com/apache/incubator-mxnet/issues/10971
## Description
Hi all. I've just recently stumbled into a segmentation fault, which I can't
really explain. Creating a new operator (even the softmax
asitstands opened a new pull request #10970: [MXNET-424] dtype option for
multinomial
URL: https://github.com/apache/incubator-mxnet/pull/10970
## Description ##
This PR adds `dtype` option to set the data type of the sample output array
of `random.multinomial`, which is fixed as
qichaotang opened a new issue #10969: Mxnet dynamic graphs achieve scoure code
URL: https://github.com/apache/incubator-mxnet/issues/10969
Hi, all hackers
I want to read Mxnet dynamic graphs source code , I know, gluon is front
end, I want to know how to achieve this in C++
lordofgod opened a new issue #10968: Discuss about NDArray::Save and
NDArray::Load
URL: https://github.com/apache/incubator-mxnet/issues/10968
There seems to be some mistakes in NDArray::Save(dmlc::Stream *strm)
(File: ndarray.cc, Line: 1587~1651) and NDArray::Load(dmlc::Stream *strm)
juliusshufan commented on issue #10921: Test cases improvement for MKLDNN on
Gluon
URL: https://github.com/apache/incubator-mxnet/pull/10921#issuecomment-389424039
@marcoabreu @szha I checked the preci logs, and looks it is not relevant to
my changes. if @zheng-da is fine with the
liuzx32 commented on issue #10936: aws-ec2 make -j4 and throw the error
URL:
https://github.com/apache/incubator-mxnet/issues/10936#issuecomment-389420067
Ok, I get it, maybe someting wrong with opencv dependencies. thanks very
much! But I hope that mxnet can give some examples for other
liuzx32 closed issue #10936: aws-ec2 make -j4 and throw the error
URL: https://github.com/apache/incubator-mxnet/issues/10936
This is an automated message from the Apache Git Service.
To respond to the message, please log on
liuzx32 opened a new issue #10967: Is the doc for launch.py parameters outdated？
URL: https://github.com/apache/incubator-mxnet/issues/10967
https://mxnet.incubator.apache.org/versions/master/faq/distributed_training.html#modeling
##
There are no instractions for option --cluster and
yajiedesign commented on issue #10481: [MXNET-302] added ability to set search
path for Accelerate library
URL: https://github.com/apache/incubator-mxnet/pull/10481#issuecomment-389417953
LGTM
This is an automated message
nihui commented on issue #10804: Use depthwise convolution(group convolution)
by cuDNNv7 if available
URL: https://github.com/apache/incubator-mxnet/pull/10804#issuecomment-389411880
hello
some feedback about the speed
hardware: tesla-m40 24G x 2
system: centos-7
nihui commented on issue #10804: Use depthwise convolution(group convolution)
by cuDNNv7 if available
URL: https://github.com/apache/incubator-mxnet/pull/10804#issuecomment-389411880
hello
some feedback about the speed
hardware: tesla-m40 24G x 2
system: centos-7
liuzx32 opened a new issue #10966: mx.kvstore.create('dist_sync') throw error.
Is there anything wrong with my environment?
URL: https://github.com/apache/incubator-mxnet/issues/10966
mx.kvstore.create('dist_sync')
mx.kvstore.create('dist_async')
##
MXNetError: [06:04:43]
jinhuang415 commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r188510242
##
File path: src/operator/quantization/requantize-inl.h
##
@@ -68,6 +68,24 @@
101 - 156 of 156 matches
Mail list logo
