lupesko commented on issue #9378: Mxnet C API CPU Mode segmentation fault
URL:
https://github.com/apache/incubator-mxnet/issues/9378#issuecomment-366889618
@sandeep-krishnamurthy can you please close?
This is an automated mes
lupesko commented on issue #9378: Mxnet C API CPU Mode segmentation fault
URL:
https://github.com/apache/incubator-mxnet/issues/9378#issuecomment-366889574
@pitLog I verified today that the C API with the example code works
perfectly well.
It is most likely that the problem is in your m
rimusolem opened a new issue #9836: Allow -2, -3, and -4 as a shape for
NDArray.reshape
URL: https://github.com/apache/incubator-mxnet/issues/9836
This is a feature request. `mxnet.ndarray.reshape` and
`mxnet.symbol.reshape` allow 0, -1, -2, -3, and -4 as valid shapes with special
meanin
szha commented on issue #9835: Failed to export gluon model zoo vision model to
symbol file
URL:
https://github.com/apache/incubator-mxnet/issues/9835#issuecomment-366882796
@zheng-da would you take a look?
This is an automa
marcoabreu commented on issue #9705: Added unittest for benchmarking metric
performance
URL: https://github.com/apache/incubator-mxnet/pull/9705#issuecomment-366881982
It's planned for end of Q1
Am 20.02.2018 5:47 vorm. schrieb "Sheng Zha" :
> When will nightly tests be moved
kevinthesun opened a new issue #9835: Failed to export gluon model zoo vision
model to symbol file
URL: https://github.com/apache/incubator-mxnet/issues/9835
## Description
Gluon block export function returned error while trying to export mobilenet,
alexnet, densenet, squeezenet and res
anirudh2290 commented on issue #9826: Crash Mxnet: Error in `python3':
corrupted double-linked list: 0x7f1c4b2e09d0
URL:
https://github.com/apache/incubator-mxnet/issues/9826#issuecomment-366877966
What version of mxnet are you using ? Please provide commit hash ? Please
provide repro
rafaelsimonmaia commented on issue #9217: Installing GPU support on Mac
URL:
https://github.com/apache/incubator-mxnet/issues/9217#issuecomment-366876266
Hi @helloniklas, could you remove the space after `USE_CUDA = 1`. It is
just a minor detail that may make it fail to compile if someone
cjolivier01 commented on a change in pull request #9832: Ability to not build
examples during the build of the cpp package.
URL: https://github.com/apache/incubator-mxnet/pull/9832#discussion_r169219022
##
File path: cpp-package/CMakeLists.txt
##
@@ -1,21 +1,23 @@
if(USE_
szha commented on issue #9705: Added unittest for benchmarking metric
performance
URL: https://github.com/apache/incubator-mxnet/pull/9705#issuecomment-366865940
When will nightly tests be moved to public CI?
This is an autom
szha commented on issue #9805: Enable the reporting of cross-entropy or nll
loss value when training CNN network using the models defined by
example/image-classification
URL: https://github.com/apache/incubator-mxnet/pull/9805#issuecomment-366863435
Yes, that test is a flaky test #9820
-
juliusshufan commented on issue #9805: Enable the reporting of cross-entropy or
nll loss value when training CNN network using the models defined by
example/image-classification
URL: https://github.com/apache/incubator-mxnet/pull/9805#issuecomment-366863185
I rerun the failed GPU case "tes
eric-haibin-lin commented on a change in pull request #9747: Add
contrib.rand_zipfian
URL: https://github.com/apache/incubator-mxnet/pull/9747#discussion_r169206557
##
File path: python/mxnet/ndarray/contrib.py
##
@@ -18,9 +18,81 @@
# coding: utf-8
# pylint: disable=wild
szha closed pull request #9834: minor spelling tweaks for docs
URL: https://github.com/apache/incubator-mxnet/pull/9834
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 pull req
This is an automated email from the ASF dual-hosted git repository.
zhasheng 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 af0c3b4 minor spelling tweaks for do
Author: liuyizhi
Date: Tue Feb 20 01:37:51 2018
New Revision: 25153
Log:
1.1.0 release moving from dev to release repo
Added:
release/incubator/mxnet/1.1.0/
- copied from r25152, dev/incubator/mxnet/1.1.0/
Removed:
dev/incubator/mxnet/1.1.0/
Author: liuyizhi
Date: Tue Feb 20 01:32:40 2018
New Revision: 25152
Log:
Add mxnet-1.1.0
Added:
dev/incubator/mxnet/1.1.0/
dev/incubator/mxnet/1.1.0/apache-mxnet-src-1.1.0-incubating.tar.gz (with
props)
dev/incubator/mxnet/1.1.0/apache-mxnet-src-1.1.0-incubating.tar.gz.asc
dev/
szha commented on a change in pull request #9833: [Metric] Accelerate the
calculation of F1
URL: https://github.com/apache/incubator-mxnet/pull/9833#discussion_r169199762
##
File path: python/mxnet/metric.py
##
@@ -510,16 +510,10 @@ def update_binary_stats(self, label, pre
pengzhao-intel commented on issue #9828: Building with MKL fails on OSX
URL:
https://github.com/apache/incubator-mxnet/issues/9828#issuecomment-366838868
@TaoLv @ashokei, please take look this compatibility issue.
@marcoabreu I see there is **NO** OSX build and test in Jenkin.
It
pengzhao-intel commented on issue #9828: Building with MKL fails on OSX
URL:
https://github.com/apache/incubator-mxnet/issues/9828#issuecomment-366838868
@TaoLv @ashokei, please take look this compatibility issue.
@marcoabreu I see there is **NO** OSX build and test in Jenkin.
It
brettkoonce opened a new pull request #9834: minor spelling tweaks for docs
URL: https://github.com/apache/incubator-mxnet/pull/9834
This is an automated message from the Apache Git Service.
To respond to the message, please
rahul003 commented on issue #9831: add USE_MKLDNN to config.mk
URL: https://github.com/apache/incubator-mxnet/pull/9831#issuecomment-366836229
https://github.com/apache/incubator-mxnet/pull/9810 does this and more.
Closing
Th
rahul003 closed pull request #9831: add USE_MKLDNN to config.mk
URL: https://github.com/apache/incubator-mxnet/pull/9831
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 pull re
rahul003 commented on issue #9774: mx.io.ImageRecordIter does not respect dtype
argument
URL:
https://github.com/apache/incubator-mxnet/issues/9774#issuecomment-366835890
But for Resnet 110 on Cifar10, fp16 is much slower. Do you see something
fishy here? There are barely any operations w
rahul003 commented on issue #9774: mx.io.ImageRecordIter does not respect dtype
argument
URL:
https://github.com/apache/incubator-mxnet/issues/9774#issuecomment-366823518
Hey @KellenSunderland
I ran Resnet50 with Imagenet and got about 70% speedup. Some of the top ones
don't seem
sxjscience opened a new pull request #9833: [Metric] Accelerate the calculation
of F1
URL: https://github.com/apache/incubator-mxnet/pull/9833
## Description ##
Accelerate the calculation of F1 by removing the for-loop. I find that using
GPU will in fact make the code slower :sweat_smil
zhechen closed issue #9823: RCNN example fails for using latest mxnet
URL: https://github.com/apache/incubator-mxnet/issues/9823
This is an automated message from the Apache Git Service.
To respond to the message, please log
zhechen commented on issue #9823: RCNN example fails for using latest mxnet
URL:
https://github.com/apache/incubator-mxnet/issues/9823#issuecomment-366833092
I somehow found a solution to this. Since I observed that this issue is
caused by cudnn_softmax_activation function, both disabling
b0noI opened a new pull request #9832: Ability to not build examples during the
build of the cpp package.
URL: https://github.com/apache/incubator-mxnet/pull/9832
## Description ##
Not everyone need to build all the examples together with the cpp_package.
Not building examples saves tim
samhodge commented on issue #9813: Unable to save gluon model to symbolic
network : neural style
URL:
https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-366830758
thanks for the help, I will continue with this advice.
---
sxjscience commented on issue #9816: Dropout may mask values even when ratio=0.0
URL:
https://github.com/apache/incubator-mxnet/issues/9816#issuecomment-366827980
There was an issue in mshadow about this
https://github.com/dmlc/mshadow/issues/213. The cuda rng generates (0.0, 1.0].
--
anirudh2290 commented on issue #9813: Unable to save gluon model to symbolic
network : neural style
URL:
https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-366826156
@samhodge Can you just reuse the InstanceNorm in
python/mxnet/gluon/nn/basic_layers.py ? your Net class see
Roshrini commented on issue #9813: Unable to save gluon model to symbolic
network : neural style
URL:
https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-366824341
@samhodge Not sure if this helps but HybridBlock has export method
https://mxnet.incubator.apache.org/api/pyt
Roshrini commented on issue #9813: Unable to save gluon model to symbolic
network : neural style
URL:
https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-366824341
samhodge Not sure if this helps but HybridBlock has export method
https://mxnet.incubator.apache.org/api/pyth
Roshrini commented on issue #9813: Unable to save gluon model to symbolic
network : neural style
URL:
https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-366824341
Not sure if this helps but HybridBlock has export method
https://mxnet.incubator.apache.org/api/python/gluon/g
DickJC123 commented on issue #9816: Dropout may mask values even when ratio=0.0
URL:
https://github.com/apache/incubator-mxnet/issues/9816#issuecomment-366824118
I like making ratio=0 a special case with an identity pass-through. During
the same fix, how about making ratio=1 a special cas
rahul003 commented on issue #9774: mx.io.ImageRecordIter does not respect dtype
argument
URL:
https://github.com/apache/incubator-mxnet/issues/9774#issuecomment-366823518
Hey @KellenSunderland
These are the top 30 activities on the GPU. I'm running Resnet 50 on
imagenet data with
rahul003 commented on issue #9774: mx.io.ImageRecordIter does not respect dtype
argument
URL:
https://github.com/apache/incubator-mxnet/issues/9774#issuecomment-366788577
Thanks @KellenSunderland . I am unable to run nvprof . Looks like this is a
known issue
(https://devtalk.nvidia.com/d
Roshrini commented on issue #9804: [R] mx.io.arrayiter shuffing is disabled
URL:
https://github.com/apache/incubator-mxnet/issues/9804#issuecomment-366818160
@thirdwing Can you please take a look?
This is an automated message
rahul003 opened a new pull request #9831: add USE_MKLDNN to config.mk
URL: https://github.com/apache/incubator-mxnet/pull/9831
## Description ##
(Brief description on what this PR is about)
## Checklist ##
### Essentials ###
- [ ] Passed code style checking (`make lint`)
-
szha commented on issue #9830: Macro F1 score depends on the size of the
minibatch
URL:
https://github.com/apache/incubator-mxnet/issues/9830#issuecomment-366817125
Fixing this requires api change, so we should do that in 2.0. #9686
---
marcoabreu commented on issue #9492: fix print_summary bug and add groups of
convolution
URL: https://github.com/apache/incubator-mxnet/pull/9492#issuecomment-366816544
Wait, in how far does this have test coverage?
This is
marcoabreu commented on a change in pull request #9809: register optimizers
only once in CPP-Package
URL: https://github.com/apache/incubator-mxnet/pull/9809#discussion_r169180595
##
File path: cpp-package/example/test_optimizer.cpp
##
@@ -0,0 +1,33 @@
+/*
+ * Licensed to
sxjscience opened a new issue #9830: Macro F1 score depends on the size of the
minibatch
URL: https://github.com/apache/incubator-mxnet/issues/9830
I find our current implementation of the macro F1 score is dependent to the
size of the minibatch. Our strategy is to average the F1 scores of
chsin commented on issue #9809: register optimizers only once in CPP-Package
URL: https://github.com/apache/incubator-mxnet/pull/9809#issuecomment-366808870
@marcoabreu, test was made, checks passes, now what?
This is an autom
piiswrong commented on issue #9816: Dropout may mask values even when ratio=0.0
URL:
https://github.com/apache/incubator-mxnet/issues/9816#issuecomment-366808616
This needs to be fix.
One possible fix is to use identity pass through when rate=0
--
piiswrong closed pull request #9492: fix print_summary bug and add groups of
convolution
URL: https://github.com/apache/incubator-mxnet/pull/9492
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 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 f14b2eb fix print_summary bug and add gr
rahul003 commented on issue #9774: mx.io.ImageRecordIter does not respect dtype
argument
URL:
https://github.com/apache/incubator-mxnet/issues/9774#issuecomment-366788577
Thanks @KellenSunderland . I am unable to run nvprof . Looks like this is a
known issue
(https://devtalk.nvidia.com/d
piiswrong commented on issue #9718: organized installation instructions to fix
render issue
URL: https://github.com/apache/incubator-mxnet/pull/9718#issuecomment-366790805
ping
This is an automated message from the Apache Git
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 50f326f Use get_bz2_data in test_utils (
piiswrong closed pull request #9776: Use get_bz2_data from test_utils for
sparse_op script
URL: https://github.com/apache/incubator-mxnet/pull/9776
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:
rahul003 commented on issue #9774: mx.io.ImageRecordIter does not respect dtype
argument
URL:
https://github.com/apache/incubator-mxnet/issues/9774#issuecomment-366788577
Thanks @KellenSunderland . I am unable to run nvprof . Looks like this is a
known issue
(https://devtalk.nvidia.com/d
piiswrong closed pull request #9790: make array.reshape compatible with numpy
URL: https://github.com/apache/incubator-mxnet/pull/9790
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 f
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 9348a3a make array.reshape compatible wi
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 d048615 temporary solution for instancen
piiswrong closed pull request #9807: temporary solution for instancenorm, will
refactor using backend
URL: https://github.com/apache/incubator-mxnet/pull/9807
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 pr
terrytangyuan commented on issue #9829: Simplify R package installation
instruction
URL: https://github.com/apache/incubator-mxnet/pull/9829#issuecomment-366786645
Actually this messes up with other default repos. Closing this now.
-
terrytangyuan closed pull request #9829: Simplify R package installation
instruction
URL: https://github.com/apache/incubator-mxnet/pull/9829
This is an automated message from the Apache Git Service.
To respond to the messag
terrytangyuan commented on issue #9829: Simplify R package installation
instruction
URL: https://github.com/apache/incubator-mxnet/pull/9829#issuecomment-366786334
Because it's shorter and you don't have to instantiate a variable
piiswrong commented on issue #9829: Simplify R package installation instruction
URL: https://github.com/apache/incubator-mxnet/pull/9829#issuecomment-366785886
why?
This is an automated message from the Apache Git Service.
To
samhodge commented on issue #9813: Unable to save gluon model to symbolic
network : neural style
URL:
https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-366778617
I have continued on further see
`python main.py train --dataset dataset --style-folder images/styles
samhodge commented on issue #9813: Unable to save gluon model to symbolic
network : neural style
URL:
https://github.com/apache/incubator-mxnet/issues/9813#issuecomment-366778617
I have continued on further see
`python main.py train --dataset dataset --style-folder images/styles
marcoabreu commented on issue #9828: Building with MKL fails on OSX
URL:
https://github.com/apache/incubator-mxnet/issues/9828#issuecomment-366763226
@zheng-da
This is an automated message from the Apache Git Service.
To res
shivonkar commented on issue #8575: mxnet multicore on LInux in R
URL:
https://github.com/apache/incubator-mxnet/issues/8575#issuecomment-366755177
Here is solution.
First set the environment variable
Sys.setenv(MXNET_CPU_WORKER_NTHREADS = core);
Use ctx= mx.cpu() dependin
terrytangyuan opened a new pull request #9829: Simplify R package installation
instruction
URL: https://github.com/apache/incubator-mxnet/pull/9829
This is an automated message from the Apache Git Service.
To respond to the
lebeg commented on a change in pull request #9799: Cleaned up image
classification cpp example
URL: https://github.com/apache/incubator-mxnet/pull/9799#discussion_r169129782
##
File path: example/image-classification/predict-cpp/CMakeLists.txt
##
@@ -1,34 +1,28 @@
-if(USE_
stonedl3 closed issue #9766: DeepLearning on Imagenet with mxnet issues
translating .lst to .rec files
URL: https://github.com/apache/incubator-mxnet/issues/9766
This is an automated message from the Apache Git Service.
To r
sbodenstein opened a new issue #9828: Building with MKL fails on OSX
URL: https://github.com/apache/incubator-mxnet/issues/9828
## Description
When building MXNet on OSX with the flag `USE_MKLDNN=1` in the `config.m`
file, I get a build failure.
## Environment info (Required)
marcoabreu commented on a change in pull request #9799: Cleaned up image
classification cpp example
URL: https://github.com/apache/incubator-mxnet/pull/9799#discussion_r169122011
##
File path: example/image-classification/predict-cpp/CMakeLists.txt
##
@@ -1,34 +1,28 @@
-if
lebeg commented on issue #9799: Cleaned up image classification cpp example
URL: https://github.com/apache/incubator-mxnet/pull/9799#issuecomment-366740410
Added windows to tested platforms
This is an automated message from th
jmacglashan commented on issue #9822: gluon HybridBlock wrapper of constant
nd.array, is it possible?
URL:
https://github.com/apache/incubator-mxnet/issues/9822#issuecomment-366731099
I would like better constant support for symbols (and consequently,
`HybridBlock`s) as well.
A wor
jmacglashan commented on issue #9822: gluon HybridBlock wrapper of constant
nd.array, is it possible?
URL:
https://github.com/apache/incubator-mxnet/issues/9822#issuecomment-366731099
I would like better constant support for symbols (and consequently,
`HybridBlock`s) as well.
A wor
stonedl3 commented on issue #9766: DeepLearning on Imagenet with mxnet issues
translating .lst to .rec files
URL:
https://github.com/apache/incubator-mxnet/issues/9766#issuecomment-366729741
I have resolved the issue. I had used resize=256 while my training script
was using 3x227x227 ima
gigasquid commented on issue #9794: MKL linker error on macOS 10.13 despite
using configuration without MKL
URL:
https://github.com/apache/incubator-mxnet/issues/9794#issuecomment-366727882
fwiw - I just ran into this trying out the gpu build as well
--
jmacglashan opened a new issue #9827: Request: to_bytes and from_bytes variant
of ndarray.save and ndarray.load.
URL: https://github.com/apache/incubator-mxnet/issues/9827
Mxnet has convenient `ndarray` (and `ParameterDict`) save/load functions,
but they always go through files specified b
KellenSunderland commented on issue #9753: Make fail for MXNET
URL:
https://github.com/apache/incubator-mxnet/issues/9753#issuecomment-366725803
I usually create a softlink from openblas to blas. Should be something like
```bash
ln - s /usr/lib/libopenblas.so /usr/lib/libcblas.so
marcoabreu commented on issue #9794: MKL linker error on macOS 10.13 despite
using configuration without MKL
URL:
https://github.com/apache/incubator-mxnet/issues/9794#issuecomment-366725166
@zheng-da could you take a look?
KellenSunderland commented on issue #9762: Should MXNet also support logical
operations?
URL:
https://github.com/apache/incubator-mxnet/issues/9762#issuecomment-366725239
+1. This comes in quite handy for NLP problems where the sequence length is
variable.
--
edmBernard opened a new issue #9826: Crash Mxnet: Error in `python3': corrupted
double-linked list: 0x7f1c4b2e09d0
URL: https://github.com/apache/incubator-mxnet/issues/9826
I try to test Mtcnn face detection from this repository:
https://github.com/pangyupo/mxnet_mtcnn_face_detection
lebeg opened a new issue #9825: Google index points to old version of install
instructions
URL: https://github.com/apache/incubator-mxnet/issues/9825
## Description
![screenshot from 2018-02-19
15-36-47](https://user-images.githubusercontent.com/1753787/36383002-1a4d3dd0-158b-11e8-8
lebeg opened a new issue #9824: Windows build instructions broken
URL: https://github.com/apache/incubator-mxnet/issues/9824
## Description
Build from source instructions for Windows point to MAC OS build:
https://mxnet.incubator.apache.org/versions/master/install/index.html
-
stonedl3 commented on issue #9766: DeepLearning on Imagenet with mxnet issues
translating .lst to .rec files
URL:
https://github.com/apache/incubator-mxnet/issues/9766#issuecomment-366689907
I have oosted the script.
On Feb 13, 2018 7:48 PM, "David Stone" wrote:
> I am trave
szha commented on issue #8630: How to reassign a new input symbol when load
symbol from json file
URL:
https://github.com/apache/incubator-mxnet/issues/8630#issuecomment-366678989
@apache/mxnet-committers: This issue has been inactive for the past 90 days.
It has no label and needs triage
gigasquid commented on issue #8971: Clojure Library for mxnet
URL:
https://github.com/apache/incubator-mxnet/issues/8971#issuecomment-366543008
An update - I've ported over the MnistModule example to clojure - yay! ?
https://github.com/gigasquid/incubator-mxnet/blob/clojure-package/clo
welnaseth commented on issue #621: Support for other Device Types, OpenCL AMD
GPU
URL: https://github.com/apache/incubator-mxnet/issues/621#issuecomment-366623893
Any updates on this? @kpot makes a good point above, tvm (and nnvm due to
being built off of it) supports opencl, so to me it s
86 matches
Mail list logo
