wkcn commented on issue #14526: Fixes for CI downloads (v1.4.x)
URL: https://github.com/apache/incubator-mxnet/pull/14526#issuecomment-476727533
The PR has been merged. Thank you for the fix!
This is an automated message from
wkcn merged pull request #14526: Fixes for CI downloads (v1.4.x)
URL: https://github.com/apache/incubator-mxnet/pull/14526
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
drivanov closed pull request #14443: Mxnet allclose
URL: https://github.com/apache/incubator-mxnet/pull/14443
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and us
drivanov commented on issue #14443: Mxnet allclose
URL: https://github.com/apache/incubator-mxnet/pull/14443#issuecomment-476727363
Failing tests are not related to the changes of this PR. I will launch CI
one more time.
Thi
gigasquid commented on issue #14531: Chouffe/clojure fix tests
URL: https://github.com/apache/incubator-mxnet/pull/14531#issuecomment-476713430
Thanks for fixing this 💯 - Will review shortly
This is an automated message from t
Chouffe opened a new pull request #14531: Chouffe/clojure fix tests
URL: https://github.com/apache/incubator-mxnet/pull/14531
## Description ##
Fixes some clojure tests that were not being run because the `is` macro from
`clojure.test` was returning true.
Tests were defined like s
jlcontreras commented on a change in pull request #14525: Fixes for CI
downloads (v1.3.x)
URL: https://github.com/apache/incubator-mxnet/pull/14525#discussion_r269172808
##
File path: cpp-package/example/feature_extract/run.sh
##
@@ -17,7 +17,12 @@
# Downloading the dat
perdasilva edited a comment on issue #14502: [Test Failure] GPU Test failures
across different CUDA versions
URL:
https://github.com/apache/incubator-mxnet/issues/14502#issuecomment-476694909
I've figured out why the test_lstmp test was failing and have submitted a PR
(#14529) to fix that
perdasilva commented on issue #14502: [Test Failure] GPU Test failures across
different CUDA versions
URL:
https://github.com/apache/incubator-mxnet/issues/14502#issuecomment-476694909
I've figured out why the test_lstmp test was failing and have submitted a PR
(#14529) to fix that issue.
lebeg commented on issue #14529: Fix lstmp GPU test
URL: https://github.com/apache/incubator-mxnet/pull/14529#issuecomment-476694089
It would be great if you could mention `test_lstmp` in the title or
description explicitly to ease the search.
--
lebeg opened a new pull request #14530: [WIP] (Do not merge) Removed
conflicting GitHub status update
URL: https://github.com/apache/incubator-mxnet/pull/14530
## Description ##
(Brief description on what this PR is about)
## Checklist ##
### Essentials ###
Please feel free
lebeg commented on issue #14529: Fix lstmp GPU test
URL: https://github.com/apache/incubator-mxnet/pull/14529#issuecomment-476688964
lgtm
This is an automated message from the Apache Git Service.
To respond to the message, ple
perdasilva opened a new pull request #14529: Fix lstmp GPU test
URL: https://github.com/apache/incubator-mxnet/pull/14529
## Description ##
The way CI is at the moment, some failing GPU tests are going undetected.
I'm putting together a PR (#14513) to update CI do CUDA v10 and fix a
perdasilva closed pull request #14527: Fixes hidden gpu test failure
URL: https://github.com/apache/incubator-mxnet/pull/14527
This is an automated message from the Apache Git Service.
To respond to the message, please log on
pengzhao-intel commented on a change in pull request #14528: Performance
improving for MKL-DNN Quantized FullyConnected
URL: https://github.com/apache/incubator-mxnet/pull/14528#discussion_r269140994
##
File path: src/operator/subgraph/mkldnn/mkldnn_fc.cc
##
@@ -195,6 +195
pengzhao-intel commented on a change in pull request #14528: Performance
improving for MKL-DNN Quantized FullyConnected
URL: https://github.com/apache/incubator-mxnet/pull/14528#discussion_r269139721
##
File path: src/operator/nn/fully_connected-inl.h
##
@@ -48,6 +48,12 @@
lihaofd commented on a change in pull request #14476: Change RNN OP to stateful
URL: https://github.com/apache/incubator-mxnet/pull/14476#discussion_r269112331
##
File path: src/operator/rnn-inl.h
##
@@ -436,387 +566,897 @@ class RNNOp : public Operator{
if (param_.sta
ciyongch opened a new pull request #14528: Performance improving for MKL-DNN
Quantized FullyConnected
URL: https://github.com/apache/incubator-mxnet/pull/14528
## Description ##
The patch is mainly for improving the performance of MKL-DNN quantized
FullyConnected.
@pengzhao-intel
wkcn edited a comment on issue #14522: mx.nd.Custom conflicts with memory
management
URL:
https://github.com/apache/incubator-mxnet/issues/14522#issuecomment-47516
I see. In `python/mxnet/ndarray/ndarray.py`, the function `def copyto(self,
other):` calls `_internal._copyto(self, out=o
wkcn commented on issue #14522: mx.nd.Custom conflicts with memory management
URL:
https://github.com/apache/incubator-mxnet/issues/14522#issuecomment-47516
I see. In `python/mxnet/ndarray/ndarray.py', the function `def copyto(self,
other):` calls `_internal._copyto(self, out=other)`,
wkcn edited a comment on issue #14522: mx.nd.Custom conflicts with memory
management
URL:
https://github.com/apache/incubator-mxnet/issues/14522#issuecomment-476660491
reproduce the bug in CPU environment, it get stuck too.
I use GDB to print the backtrace:
```
#0 0x7
wkcn commented on issue #14522: mx.nd.Custom conflicts with memory management
URL:
https://github.com/apache/incubator-mxnet/issues/14522#issuecomment-476660491
reproduce the bug in CPU environment, it get stuck too.
This
lihaofd commented on a change in pull request #14476: Change RNN OP to stateful
URL: https://github.com/apache/incubator-mxnet/pull/14476#discussion_r269112331
##
File path: src/operator/rnn-inl.h
##
@@ -436,387 +566,897 @@ class RNNOp : public Operator{
if (param_.sta
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 ce4a511 Bump the publish ti
perdasilva opened a new pull request #14527: Fixes hidden gpu test failure
URL: https://github.com/apache/incubator-mxnet/pull/14527
## Description ##
The way CI is at the moment, some failing GPU tests are going undetected.
I'm putting together a PR (#14513) to update CI do CUDA v10
Bumblebee1964 commented on issue #14116: Failure in generated op.h in version
1.3.1
URL:
https://github.com/apache/incubator-mxnet/issues/14116#issuecomment-476603510
It looks whether the name of the operator (and probably definition) that
caused the problem has been changed. From my curr
wkcn edited a comment on issue #14522: mx.nd.Custom conflicts with memory
management
URL:
https://github.com/apache/incubator-mxnet/issues/14522#issuecomment-476598361
I mean that does this issue exist before the PR
https://github.com/apache/incubator-mxnet/pull/14363 ?
-
wkcn commented on issue #14522: mx.nd.Custom conflicts with memory management
URL:
https://github.com/apache/incubator-mxnet/issues/14522#issuecomment-476598361
I mean that does this issue exists before the PR
https://github.com/apache/incubator-mxnet/pull/14363 ?
YutingZhang commented on issue #14522: mx.nd.Custom conflicts with memory
management
URL:
https://github.com/apache/incubator-mxnet/issues/14522#issuecomment-476589331
@wkcn "the bug in old version": Do you mean the deadlock in subprocess? I
thought https://github.com/apache/incubator-mxn
smissan commented on issue #14116: Failure in generated op.h in version 1.3.1
URL:
https://github.com/apache/incubator-mxnet/issues/14116#issuecomment-476566585
The script did not change in 2 months and is still not working
On Tue., Mar. 26, 2019, 5:21 a.m. Fred Hugen,
wrote:
lebeg commented on issue #14162: [WIP] Quantize/digitize operator
URL: https://github.com/apache/incubator-mxnet/pull/14162#issuecomment-476551712
Closing this as no capacity to work on this right now.
This is an automated mes
lebeg closed pull request #14162: [WIP] Quantize/digitize operator
URL: https://github.com/apache/incubator-mxnet/pull/14162
This is an automated message from the Apache Git Service.
To respond to the message, please log on t
lebeg closed pull request #13574: Replaced rand_r with std:: random generation
URL: https://github.com/apache/incubator-mxnet/pull/13574
This is an automated message from the Apache Git Service.
To respond to the message, ple
lebeg commented on issue #13574: Replaced rand_r with std:: random generation
URL: https://github.com/apache/incubator-mxnet/pull/13574#issuecomment-476551370
Closing this for now, no capacity to work on this right now.
This i
lebeg commented on issue #14504: Fixes for CI downloads
URL: https://github.com/apache/incubator-mxnet/pull/14504#issuecomment-476548989
Created ports for release branches:
* [#14525](https://github.com/apache/incubator-mxnet/pull/14525) Fixes for
CI downloads (v1.3.x)
* [#14526](
lebeg opened a new pull request #14526: Fixes for CI downloads (v1.4.x)
URL: https://github.com/apache/incubator-mxnet/pull/14526
## Description ##
Port of https://github.com/apache/incubator-mxnet/pull/14504 to the v1.4.x
branch.
## Checklist ##
### Essentials ###
Pleas
lebeg opened a new pull request #14525: Fixes for CI downloads (v1.3.x)
URL: https://github.com/apache/incubator-mxnet/pull/14525
## Description ##
Port of https://github.com/apache/incubator-mxnet/pull/14504 to the v1.3.x
branch.
## Checklist ##
### Essentials ###
Pleas
kohillyang removed a comment on issue #9686: APIs that might be a good idea to
break in 2.0
URL:
https://github.com/apache/incubator-mxnet/issues/9686#issuecomment-476283396
And, there seem many functions to load parameters from a pre-trained file,
but for nn.Block, there is no public way
kohillyang removed a comment on issue #9686: APIs that might be a good idea to
break in 2.0
URL:
https://github.com/apache/incubator-mxnet/issues/9686#issuecomment-476279328
I think currently CustomOp may need to be redesigned, they cannot be
executed in multi-threads, and even in multi-p
lebeg commented on issue #14234: Backwards compatibility out of bounds for 1.2.1
URL:
https://github.com/apache/incubator-mxnet/issues/14234#issuecomment-476539211
Probably this issue can be closed.
This is an automated messa
lebeg commented on issue #14234: Backwards compatibility out of bounds for 1.2.1
URL:
https://github.com/apache/incubator-mxnet/issues/14234#issuecomment-476529891
Created separate issue https://github.com/apache/incubator-mxnet/issues/14524
beingjoey commented on issue #14508: python import models include mxnet cause a
segmentation error
URL:
https://github.com/apache/incubator-mxnet/issues/14508#issuecomment-476529911
I reproduced this problem on two different computers, with different version
of these models.
I found I
lebeg opened a new issue #14524: Flaky backwards compatibility with 1.2.1 for
lenet_gluon_save_params_api
URL: https://github.com/apache/incubator-mxnet/issues/14524
## Description
Several failures for
[backwards-compatibility-checker](http://jenkins.mxnet-ci.amazon-ml.com/job/restr
mxnet-label-bot commented on issue #14524: Flaky backwards compatibility with
1.2.1 for lenet_gluon_save_params_api
URL:
https://github.com/apache/incubator-mxnet/issues/14524#issuecomment-476529727
Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and s
IvyGongoogle commented on issue #14512: the inference speed using C++ API with
mxnet of higher version is slower than lower mxnet
URL:
https://github.com/apache/incubator-mxnet/issues/14512#issuecomment-476528740
@wkcn thanks for your reply,
```
mxnet v0.8:
initialization time(
lebeg commented on issue #14234: Backwards compatibility out of bounds for 1.2.1
URL:
https://github.com/apache/incubator-mxnet/issues/14234#issuecomment-476527904
Another failure:
http://jenkins.mxnet-ci.amazon-ml.com/job/restricted-backwards-compatibility-checker/420/console
beingjoey commented on issue #14508: python import models include mxnet cause a
segmentation error
URL:
https://github.com/apache/incubator-mxnet/issues/14508#issuecomment-476526022
yes, I just remove
from skimage import transform
--
lobanov-m edited a comment on issue #13226: [MXNet-1211] Factor and "Like"
modes in BilinearResize2D operator
URL: https://github.com/apache/incubator-mxnet/pull/13226#issuecomment-476522994
Error in build doesn't connected with pull request:
```
Rscript -e "library(devtools); library
lobanov-m commented on issue #13226: [MXNet-1211] Factor and "Like" modes in
BilinearResize2D operator
URL: https://github.com/apache/incubator-mxnet/pull/13226#issuecomment-476522994
```
Rscript -e "library(devtools); library(methods);
options(repos=c(CRAN='https://cloud.r-project.org/
Bumblebee1964 commented on issue #14116: Failure in generated op.h in version
1.3.1
URL:
https://github.com/apache/incubator-mxnet/issues/14116#issuecomment-476519772
It seems fixed in the latest version. I downloaded the code 2 weeks ago, and
it seemed to be resolved.
--
szha commented on a change in pull request #14476: Change RNN OP to stateful
URL: https://github.com/apache/incubator-mxnet/pull/14476#discussion_r268979558
##
File path: src/operator/rnn-inl.h
##
@@ -436,387 +566,897 @@ class RNNOp : public Operator{
if (param_.state_
wkcn commented on issue #14522: mx.nd.Custom conflicts with memory management
URL:
https://github.com/apache/incubator-mxnet/issues/14522#issuecomment-476513543
Does it reproduce the bug in old version of MXNet?
This is an au
pengzhao-intel commented on issue #14440: Seg fault during mxnet.Convolution()
after reshape()
URL:
https://github.com/apache/incubator-mxnet/issues/14440#issuecomment-476512957
ping @Shumakriss
This is an automated message
pengzhao-intel merged pull request #14277: Enhance PartitionGraph
URL: https://github.com/apache/incubator-mxnet/pull/14277
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
This is an automated email from the ASF dual-hosted git repository.
patriczhao 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 f8a0dbc Enhance PartitionGraph (#1
higumachan opened a new pull request #14523: [FIX] replace shuffle method
URL: https://github.com/apache/incubator-mxnet/pull/14523
## Description ##
I replaced shuffle method `std::random_shuffle` to `std::shuffle` because It
can not build with c++17.
-
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 705e117 Bump the publish ti
YutingZhang commented on issue #14522: mx.nd.Custom conflicts with memory
management
URL:
https://github.com/apache/incubator-mxnet/issues/14522#issuecomment-476501439
@arcadiaphy @wkcn @anirudh2290 Interested in looking into this?
-
mxnet-label-bot commented on issue #14522: mx.nd.Custom conflicts with memory
management
URL:
https://github.com/apache/incubator-mxnet/issues/14522#issuecomment-476501131
Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some labels so
that
YutingZhang opened a new issue #14522: mx.nd.Custom conflicts with memory
management
URL: https://github.com/apache/incubator-mxnet/issues/14522
When training/running a large neural network with CustomOP, mxnet can get
stuck. My speculation is that if memory management (e.g.,
releasing/re
101 - 160 of 160 matches
Mail list logo