szha commented on issue #7847: Bug on Multi GPU : probably Invalid
initialization of optimizer
URL:
https://github.com/apache/incubator-mxnet/issues/7847#issuecomment-332426955
#8054 prevents the edge case from happening. However, I agree with
@piiswrong that "load_param should work even
szha commented on issue #8054: throw err message for wrong order of
load_param/init_optimizer in module
URL: https://github.com/apache/incubator-mxnet/pull/8054#issuecomment-332426414
Agreed. Let's track this in #7847. In the meantime, I suggest we apply this
patch to prevent this bug sile
piiswrong commented on issue #8054: throw err message for wrong order of
load_param/init_optimizer in module
URL: https://github.com/apache/incubator-mxnet/pull/8054#issuecomment-332425301
load_param should work even after optimizer is initialized. This is a bug
that needs to be fixed
--
szha opened a new pull request #8056: rename
URL: https://github.com/apache/incubator-mxnet/pull/8056
@zhreshold
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and u
tornadomeet commented on issue #7631: windows MXNET_CUDNN_AUTOTUNE_DEFAULT = 0
no work.
URL:
https://github.com/apache/incubator-mxnet/issues/7631#issuecomment-332422880
same here.
This is an automated message from the Apac
tornadomeet commented on issue #7631: windows MXNET_CUDNN_AUTOTUNE_DEFAULT = 0
no work.
URL:
https://github.com/apache/incubator-mxnet/issues/7631#issuecomment-332422880
same here, ubuntu
This is an automated message from t
dougsm closed issue #7006: Prediction fails with reused module
URL: https://github.com/apache/incubator-mxnet/issues/7006
This is an automated message from the Apache Git Service.
To respond to the message, please log on Git
reminisce opened a new pull request #8055: Skip adding control_deps if nodes
are the same
URL: https://github.com/apache/incubator-mxnet/pull/8055
The following code would fail before this PR due to introducing a cycle into
the gradient graph. See this issue for analysis:
https://github.c
eric-haibin-lin opened a new pull request #8054: throw err message for wrong
order of load_param/init_optimizer in module
URL: https://github.com/apache/incubator-mxnet/pull/8054
#7847
@szha
This is an automated mes
piiswrong commented on a change in pull request #8050: fix symbolblock
URL: https://github.com/apache/incubator-mxnet/pull/8050#discussion_r141248254
##
File path: python/mxnet/gluon/block.py
##
@@ -487,8 +487,8 @@ def __init__(self, outputs, inputs, params=None):
sergeykolychev opened a new pull request #8053: [Perl] do not merge, debugging
ci failure of perl gpu tests.
URL: https://github.com/apache/incubator-mxnet/pull/8053
This is an automated message from the Apache Git Service.
wut0n9 commented on a change in pull request #8030: add cnn + highway network
architecture for Chinese text classification example
URL: https://github.com/apache/incubator-mxnet/pull/8030#discussion_r141241847
##
File path: example/cnn_chinese_text_classification/data_helpers.py
##
wut0n9 commented on a change in pull request #8030: add cnn + highway network
architecture for Chinese text classification example
URL: https://github.com/apache/incubator-mxnet/pull/8030#discussion_r141241847
##
File path: example/cnn_chinese_text_classification/data_helpers.py
##
eric-haibin-lin commented on a change in pull request #7911: More sparse
related docs
URL: https://github.com/apache/incubator-mxnet/pull/7911#discussion_r141240857
##
File path: python/mxnet/test_utils.py
##
@@ -134,22 +134,32 @@ def _get_uniform_dataset_csr(num_rows, num
szha commented on a change in pull request #8050: fix symbolblock
URL: https://github.com/apache/incubator-mxnet/pull/8050#discussion_r141240853
##
File path: python/mxnet/gluon/block.py
##
@@ -487,8 +487,8 @@ def __init__(self, outputs, inputs, params=None):
self.
szha commented on a change in pull request #8030: add cnn + highway network
architecture for Chinese text classification example
URL: https://github.com/apache/incubator-mxnet/pull/8030#discussion_r141240736
##
File path: example/cnn_chinese_text_classification/data_helpers.py
szha commented on a change in pull request #7911: More sparse related docs
URL: https://github.com/apache/incubator-mxnet/pull/7911#discussion_r141240533
##
File path: python/mxnet/test_utils.py
##
@@ -134,22 +134,32 @@ def _get_uniform_dataset_csr(num_rows, num_cols,
dens
piiswrong commented on a change in pull request #8050: fix symbolblock
URL: https://github.com/apache/incubator-mxnet/pull/8050#discussion_r141240150
##
File path: python/mxnet/gluon/block.py
##
@@ -487,8 +487,8 @@ def __init__(self, outputs, inputs, params=None):
szha commented on a change in pull request #8050: fix symbolblock
URL: https://github.com/apache/incubator-mxnet/pull/8050#discussion_r141239287
##
File path: python/mxnet/gluon/block.py
##
@@ -487,8 +487,8 @@ def __init__(self, outputs, inputs, params=None):
self.
jeremiedb commented on issue #5796: Problem with installing mxnet on R v.3.3.3
URL:
https://github.com/apache/incubator-mxnet/issues/5796#issuecomment-332400372
drat is no longer in use and the pre-compiled libraries are now stored on
S3. Below is the procedure, taken from the updated docu
szha commented on issue #7319: [RoadMap] Legacy issue resolution before 1.0
release
URL:
https://github.com/apache/incubator-mxnet/issues/7319#issuecomment-332400119
Many examples are outdated or don't uphold the style standard. Duplicates of
the same or similar (most popular being MNIST
qingzhouzhen commented on issue #7957: add densenet
URL: https://github.com/apache/incubator-mxnet/pull/7957#issuecomment-332397878
Is it required to upload pretrained weight then to accept my PR?, because
the pretrained weight is about 62M, which is restricted to upload in my
company, and
sbodenstein commented on a change in pull request #8013: Add ability to query
MXNet version to C API
URL: https://github.com/apache/incubator-mxnet/pull/8013#discussion_r141234680
##
File path: src/c_api/c_api.cc
##
@@ -136,6 +136,8 @@ int MXSetNumOMPThreads(int thread_num
tqchen commented on issue #8029: Bug on sum
URL:
https://github.com/apache/incubator-mxnet/issues/8029#issuecomment-332394392
Yes, I think we do not have to add the control deps when they are the same.
Thanks for catching the problem!
-
reminisce commented on issue #8029: Bug on sum
URL:
https://github.com/apache/incubator-mxnet/issues/8029#issuecomment-332393859
@tqchen It was found that the control_deps of a gradient node is the
gradient node itself. For a graph like this: `sym = data + data`, where `data`
is a symboli
reminisce commented on issue #7956: simple_bind error when using rotation
matrix & batch_dot
URL:
https://github.com/apache/incubator-mxnet/issues/7956#issuecomment-332393922
Root cause found and tracked in this issue.
https://github.com/apache/incubator-mxnet/issues/8029#issuecomment-332
reminisce commented on issue #8029: Bug on sum
URL:
https://github.com/apache/incubator-mxnet/issues/8029#issuecomment-332393859
@tqchen It was found that the control_deps of a gradient node is the
gradient node itself. For a graph like this: `sym = data + data`, where `data`
is a symboli
bejjani closed issue #7998: Failing to build scalapkg with USE_DIST_KVSTORE=1
and USE_OPENCV=1 on Amazon Linux (RHEL)
URL: https://github.com/apache/incubator-mxnet/issues/7998
This is an automated message from the Apache G
KeyKy opened a new issue #8052: I need to add a name prefix in resnet.py
URL: https://github.com/apache/incubator-mxnet/issues/8052
I found the symbol's name in resnet.py are fixed, for example (conv0):
```
body = mx.sym.Convolution(data=data, num_filter=filter_list[0], kernel=(7,
bejjani commented on issue #7998: Failing to build scalapkg with
USE_DIST_KVSTORE=1 and USE_OPENCV=1 on Amazon Linux (RHEL)
URL:
https://github.com/apache/incubator-mxnet/issues/7998#issuecomment-332391275
@cjolivier01, Spot on. OpenCV3 was installing protobuf as well causing the
error.
jeremiedb commented on issue #7968: [R] Transfer Learning using VGG-16
URL:
https://github.com/apache/incubator-mxnet/issues/7968#issuecomment-332385636
There are different ways to approach the fine tuning, which depends on
whether there are changes to only the weights parameters or also t
KeyKy commented on issue #8018: How to put 2 networks in different gpus
(resnet152 in gpu0 and resnet50 in gpu1) ?
URL:
https://github.com/apache/incubator-mxnet/issues/8018#issuecomment-332384936
@solin319 thanks. They share the same loss layer. The whole network is
similar to a two stre
muchuanyun commented on issue #6915: Training Faster R-CNN on own dataset
URL:
https://github.com/apache/incubator-mxnet/issues/6915#issuecomment-332384154
What do you mean by "getting no results testing the model"?
If you use "test.py", it should be fed with network, end_epoch and m
mbaijal commented on a change in pull request #8051: Issue #7750: MXNet 0.11.0
Release Feedback: README File
URL: https://github.com/apache/incubator-mxnet/pull/8051#discussion_r141226022
##
File path: README.md
##
@@ -33,26 +33,26 @@ What's New
* [MKLDNN for Faster CPU P
mbaijal opened a new pull request #8051: Issue #7750: MXNet 0.11.0 Release
Feedback: README File
URL: https://github.com/apache/incubator-mxnet/pull/8051
Refer to Issue
[#7750](https://github.com/apache/incubator-mxnet/issues/7750) -
1. Installation instructions are already included und
solin319 commented on issue #8018: How to put 2 networks in different gpus
(resnet152 in gpu0 and resnet50 in gpu1) ?
URL:
https://github.com/apache/incubator-mxnet/issues/8018#issuecomment-332382160
Is there any relationship between the two network, such as they will share
some parameter
piiswrong opened a new pull request #8050: fix symbolblock
URL: https://github.com/apache/incubator-mxnet/pull/8050
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub an
ray3572 commented on issue #8039: MXPredCreate() costs all my memory(24G), and
shows "terminate called after throwing an instance of 'std::bad_alloc"
URL:
https://github.com/apache/incubator-mxnet/issues/8039#issuecomment-332379917
@chinakook but I installed from apt too!
---
cjolivier01 commented on issue #7998: Failing to build scalapkg with
USE_DIST_KVSTORE=1 on ami-a4c7edb2
URL:
https://github.com/apache/incubator-mxnet/issues/7998#issuecomment-332379190
You may have (aka probably) conflicting versions of either protobuf or
OpenCV installed. The build proc
solin319 commented on issue #8023: Error in continuing training a model loaded
from file
URL:
https://github.com/apache/incubator-mxnet/issues/8023#issuecomment-332379092
It seems that, the label used in metric has a different shape with
'lenet_model._label_shapes' used in bind.
You ca
chinakook commented on issue #8039: MXPredCreate() costs all my memory(24G),
and shows "terminate called after throwing an instance of 'std::bad_alloc"
URL:
https://github.com/apache/incubator-mxnet/issues/8039#issuecomment-332378737
openblas installed from apt is OK.
---
dmadeka commented on issue #8048: Bug in backprop for the transpose of a
transpose
URL:
https://github.com/apache/incubator-mxnet/issues/8048#issuecomment-332373231
Current hack (made compatible with 0.10.0):
``` python
def ttake( x, i ):
""" Take from axis 1 instead o
ray3572 commented on issue #8039: MXPredCreate() costs all my memory(24G), and
shows "terminate called after throwing an instance of 'std::bad_alloc"
URL:
https://github.com/apache/incubator-mxnet/issues/8039#issuecomment-332376902
@chinakook there's something wrong with openblas?
--
solin319 commented on issue #8042: Convert network written in json format to
mxnet symbol code?
URL:
https://github.com/apache/incubator-mxnet/issues/8042#issuecomment-332376924
In python, "mxnet.symbol.load_json()" can load a json str to a mxnet symbol
 costs all my memory(24G),
and shows "terminate called after throwing an instance of 'std::bad_alloc"
URL:
https://github.com/apache/incubator-mxnet/issues/8039#issuecomment-332376512
You may turn on MKL_ML or something, It's full of bugs a
ray3572 commented on issue #8039: MXPredCreate() costs all my memory(24G), and
shows "terminate called after throwing an instance of 'std::bad_alloc"
URL:
https://github.com/apache/incubator-mxnet/issues/8039#issuecomment-332376822
@piiswrong It's sure that
--
chinakook commented on issue #8039: MXPredCreate() costs all my memory(24G),
and shows "terminate called after throwing an instance of 'std::bad_alloc"
URL:
https://github.com/apache/incubator-mxnet/issues/8039#issuecomment-332376512
You may turn on MKL_ML or something, It's full of bugs.
solin319 commented on issue #8005: add warmup lr_scheduler
URL: https://github.com/apache/incubator-mxnet/pull/8005#issuecomment-332376292
Yes, I will combine warmup and multi step scheduler in a single class.
This is an auto
anirudh2290 commented on issue #7814: Pooling operator expects kernel size even
when 'global_pool' is set to True
URL:
https://github.com/apache/incubator-mxnet/issues/7814#issuecomment-332375603
The kernel parameter is required to determine whether the Pooling Operation
is 1d or 2d or 3d
anirudh2290 commented on issue #7814: Pooling operator expects kernel size even
when 'global_pool' is set to True
URL:
https://github.com/apache/incubator-mxnet/issues/7814#issuecomment-332375603
The kernel parameter is required to determine whether the Pooling Operation
is 1d or 2d or 3d
dmadeka commented on issue #8048: Bug in backprop for the transpose of a
transpose
URL:
https://github.com/apache/incubator-mxnet/issues/8048#issuecomment-332373231
Current hack:
``` python
def ttake( x, i ):
""" Take from axis 1 instead of 0.
"""
a = mx.
christophebjjani commented on issue #7998: Failing to build scalapkg with
USE_DIST_KVSTORE=1 on ami-a4c7edb2
URL:
https://github.com/apache/incubator-mxnet/issues/7998#issuecomment-332372922
if I disable opencv in the build, it works
```
echo "USE_OPENCV=0" >>config.mk
```
gautamkmr opened a new issue #8049: Enable the test test_ndarray.test_cached
after fixing the crash.
URL: https://github.com/apache/incubator-mxnet/issues/8049
Currently the test_cached crashing in mklml_cpu run unit test
## Environment info
Operating System: linux
Compiler
eric-haibin-lin commented on a change in pull request #7911: More sparse
related docs
URL: https://github.com/apache/incubator-mxnet/pull/7911#discussion_r141218375
##
File path: python/mxnet/ndarray/sparse.py
##
@@ -88,6 +88,8 @@ def _new_alloc_handle(stype, shape, ctx, d
dmadeka opened a new issue #8048: Issue in backprop for the transpose of a
transpose
URL: https://github.com/apache/incubator-mxnet/issues/8048
For bugs or installation issues, please provide the following information.
The more information you provide, the more likely people will be able
mbaijal opened a new pull request #8047: DO NOT MERGE: This is just to test the
new changes made in PR #8034
URL: https://github.com/apache/incubator-mxnet/pull/8047
This is an automated message from the Apache Git Service.
eric-haibin-lin commented on a change in pull request #7947: [WIP] Refactor
infer storage function for sparse operators.
URL: https://github.com/apache/incubator-mxnet/pull/7947#discussion_r141210016
##
File path: src/imperative/imperative_utils.h
##
@@ -551,17 +539,44 @@
eric-haibin-lin commented on issue #5783: Float64 support for symbolic dot
operator
URL:
https://github.com/apache/incubator-mxnet/issues/5783#issuecomment-332362797
Should be fixed by https://github.com/apache/incubator-mxnet/pull/6756
--
eric-haibin-lin closed issue #5783: Float64 support for symbolic dot operator
URL: https://github.com/apache/incubator-mxnet/issues/5783
This is an automated message from the Apache Git Service.
To respond to the message, pl
piiswrong closed pull request #8034: Kill running PR builds when a new build is
triggered for the same PR
URL: https://github.com/apache/incubator-mxnet/pull/8034
This is an automated message from the Apache Git Service.
To
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 4c0df62 Kill running PR builds when a ne
nswamy commented on a change in pull request #8034: Kill running PR builds when
a new build is triggered for the same PR
URL: https://github.com/apache/incubator-mxnet/pull/8034#discussion_r141207412
##
File path: Jenkinsfile
##
@@ -119,6 +143,9 @@ try {
stage("Sanity
cmwetherell commented on issue #5796: Problem with installing mxnet on R v.3.3.3
URL:
https://github.com/apache/incubator-mxnet/issues/5796#issuecomment-332356458
The URL from @jeremiedb worked well for me, whereas drat did not work. I am
running Microsoft R Open 3.3.3. I think this is sim
cmwetherell commented on issue #5796: Problem with installing mxnet on R v.3.3.3
URL:
https://github.com/apache/incubator-mxnet/issues/5796#issuecomment-332356458
The URL from @jeremiedb worked well for me, whereas drat did not work. I am
running Microsoft R Open 3.3.3. Thanks!
-
eric-haibin-lin commented on a change in pull request #7947: [WIP] Refactor
infer storage function for sparse operators.
URL: https://github.com/apache/incubator-mxnet/pull/7947#discussion_r141201987
##
File path: include/mxnet/op_attr_types.h
##
@@ -228,17 +241,24 @@ usin
eric-haibin-lin commented on a change in pull request #7947: [WIP] Refactor
infer storage function for sparse operators.
URL: https://github.com/apache/incubator-mxnet/pull/7947#discussion_r141200581
##
File path: src/c_api/c_api_function.cc
##
@@ -107,6 +107,24 @@ void Ba
szha commented on issue #7993: The Mulit-task learning example can not run
URL:
https://github.com/apache/incubator-mxnet/issues/7993#issuecomment-332352902
Fixed. You can checkout the latest master branch and give the example
another try. Let me know if there's still issue.
szha closed issue #7993: The Mulit-task learning example can not run
URL: https://github.com/apache/incubator-mxnet/issues/7993
This is an automated message from the Apache Git Service.
To respond to the message, please log
szha closed pull request #8006: fix example
URL: https://github.com/apache/incubator-mxnet/pull/8006
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL a
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 3ba0c22 fix example (#8006)
3ba0c22
mbaijal commented on a change in pull request #8034: Kill running PR builds
when a new build is triggered for the same PR
URL: https://github.com/apache/incubator-mxnet/pull/8034#discussion_r141198673
##
File path: Jenkinsfile
##
@@ -119,6 +143,9 @@ try {
stage("Sanit
mbaijal commented on issue #8034: Kill running PR builds when a new build is
triggered for the same PR
URL: https://github.com/apache/incubator-mxnet/pull/8034#issuecomment-332350218
ReviewComments:
1. Does not abort on branch master;
2. Added Comments
3. checks for build# < curr
mseeger commented on issue #7966: Operator linalg_syevd: Symmetric
eigendecomposition
URL: https://github.com/apache/incubator-mxnet/pull/7966#issuecomment-332341057
OK, I fixed this. Now, all unit tests use linalg.XXX instead of linalg_XXX.
---
cjolivier01 commented on issue #7854: Basic CPU Kernel OMP selection based upon
whether GPU has been used
URL: https://github.com/apache/incubator-mxnet/pull/7854#issuecomment-332339670
I am going on like day 4 of weird build breaks that have nothing to do with
my code...
---
zhreshold commented on issue #8005: add warmup lr_scheduler
URL: https://github.com/apache/incubator-mxnet/pull/8005#issuecomment-332332580
Warmup behavior is nice.
Do you mind adding a combined warmup and multi step scheduler which will be
more useful IMO?
gautamkmr commented on issue #8045: Disable the test test_batchnorm_training
URL: https://github.com/apache/incubator-mxnet/pull/8045#issuecomment-332330287
@piiswrong The issue is still open and needed specific Visual studio to
debug. Since the test only failed on windows machine. CC'ing
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 08c15fc Fix a mistake in example/ssd/dem
piiswrong closed pull request #8022: Fix a mistake in example/ssd/demo.py
URL: https://github.com/apache/incubator-mxnet/pull/8022
This is an automated message from the Apache Git Service.
To respond to the message, please l
piiswrong commented on issue #8045: Disable the test test_batchnorm_training
URL: https://github.com/apache/incubator-mxnet/pull/8045#issuecomment-332328716
What's the status on fixing the last skipped test?
This is an automa
piiswrong closed pull request #7939: Random refactor
URL: https://github.com/apache/incubator-mxnet/pull/7939
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use
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 0ad14b6 Random refactor (#7939)
0ad14b6
piiswrong closed pull request #8046: Issue #7748: Update the Copyright years in
NOTICE file
URL: https://github.com/apache/incubator-mxnet/pull/8046
This is an automated message from the Apache Git Service.
To respond to th
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 e040610 Issue #7748: Update the Copyrigh
mbaijal opened a new pull request #8046: Issue #7748: Update the Copyright
years in NOTICE file
URL: https://github.com/apache/incubator-mxnet/pull/8046
Fixing Issue #7748
As discussed on slack channel "builds"
This is a
piiswrong commented on issue #8027: Optional reshape of predictions in
Perplexity metric
URL: https://github.com/apache/incubator-mxnet/pull/8027#issuecomment-332314548
This is fine but in case you care about performance pick actually supports
picking on axes in the middle.
It works
gautamkmr opened a new pull request #8045: Disable the test
test_batchnorm_training
URL: https://github.com/apache/incubator-mxnet/pull/8045
@piiswrong @rahul003
This is an automated message from the Apache Git Service.
To
gautamkmr opened a new issue #8044: Enable Test
URL: https://github.com/apache/incubator-mxnet/issues/8044
For bugs or installation issues, please provide the following information.
The more information you provide, the more likely people will be able to
help you.
## Environment i
sergeykolychev commented on issue #8017: Add cuda rtc module
URL: https://github.com/apache/incubator-mxnet/pull/8017#issuecomment-332310738
libcuda.so does not exist in the /usr/local/cuda/lib , it only exists in
/usr/local/cuda/lib/stubs/
libcuda.so.1 does not exist in /usr/local/cuda
piiswrong commented on issue #8017: Add cuda rtc module
URL: https://github.com/apache/incubator-mxnet/pull/8017#issuecomment-332308621
I think libcuda.so is necessary. Does it compile without libcuda?
This is an automated me
mbaijal commented on a change in pull request #8034: Kill running PR builds
when a new build is triggered for the same PR
URL: https://github.com/apache/incubator-mxnet/pull/8034#discussion_r141158264
##
File path: Jenkinsfile
##
@@ -11,14 +11,36 @@ max_time = 120
// assi
sergeykolychev commented on issue #8017: Add cuda rtc module
URL: https://github.com/apache/incubator-mxnet/pull/8017#issuecomment-332306959
@piiswrong yeah it's weird but is not weird that ldd libmxnet.so shows
libcuda.so.1 as linked lib and docker container does not have this file ? If
y
sergeykolychev commented on issue #8017: Add cuda rtc module
URL: https://github.com/apache/incubator-mxnet/pull/8017#issuecomment-332306959
@piiswrong yeah it's weird but is not weird that ldd libmxnet.so shows
libcuda.so.1 as linked lib and docker container does not have this file ? If
y
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 a36bf57 Add cuda rtc module (#8017)
a36b
piiswrong commented on issue #8017: Add cuda rtc module
URL: https://github.com/apache/incubator-mxnet/pull/8017#issuecomment-332304003
It's weird to install a specific version of cuda inside the install script.
I'll mark perl gpu failures as ignored and merge this first.
--
piiswrong closed pull request #8017: Add cuda rtc module
URL: https://github.com/apache/incubator-mxnet/pull/8017
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and
piiswrong closed pull request #8043: Issue #7748: MXNet 0.11.0 Release
Feedback: NOTICE file
URL: https://github.com/apache/incubator-mxnet/pull/8043
This is an automated message from the Apache Git Service.
To respond to t
mbaijal opened a new pull request #8043: Issue #7748: MXNet 0.11.0 Release
Feedback: NOTICE file
URL: https://github.com/apache/incubator-mxnet/pull/8043
As discussed in [this
issue](https://github.com/apache/incubator-mxnet/issues/7748)
--
jmerkow commented on issue #7949: account for batch padding when updating
metrics
URL: https://github.com/apache/incubator-mxnet/pull/7949#issuecomment-332302224
@piiswrong I think that @mikerossgithub's approach could work pretty
seamlessly.
Though I think padding needs to be addressed
nswamy commented on a change in pull request #8034: Kill running PR builds when
a new build is triggered for the same PR
URL: https://github.com/apache/incubator-mxnet/pull/8034#discussion_r141152379
##
File path: Jenkinsfile
##
@@ -11,14 +11,36 @@ max_time = 120
// assig
nswamy commented on a change in pull request #8034: Kill running PR builds when
a new build is triggered for the same PR
URL: https://github.com/apache/incubator-mxnet/pull/8034#discussion_r141152726
##
File path: Jenkinsfile
##
@@ -11,14 +11,36 @@ max_time = 120
// assig
1 - 100 of 137 matches
Mail list logo
