piiswrong commented on a change in pull request #7345: add std_rgba to
normalize options
URL: https://github.com/apache/incubator-mxnet/pull/7345#discussion_r131536401
##
File path: src/io/iter_normalize.h
##
@@ -129,33 +129,58 @@ class ImageNormalizeIter : public IIterato
alues commented on issue #7309: Update Install script
URL: https://github.com/apache/incubator-mxnet/pull/7309#issuecomment-320495337
@piiswrong ??
This is an automated message from the Apache Git Service.
To respond to the m
firesaint opened a new issue #7353: AI-MXNet 'make test' failed.
URL: https://github.com/apache/incubator-mxnet/issues/7353
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.
jmacglashan closed issue #7312: gluon Trainer does not accept Optimizer object,
only str
URL: https://github.com/apache/incubator-mxnet/issues/7312
This is an automated message from the Apache Git Service.
To respond to the
jmacglashan commented on issue #7312: gluon Trainer does not accept Optimizer
object, only str
URL:
https://github.com/apache/incubator-mxnet/issues/7312#issuecomment-320511377
This appears to be fixed in the latest commit. Closing.
---
jmacglashan commented on issue #7351: The intellisence support for mxnet with
Python
URL:
https://github.com/apache/incubator-mxnet/issues/7351#issuecomment-320511578
I think this is because sym points to symbol and in general, many
references, like operators, currently don't show up in a
jmacglashan commented on issue #7351: The intellisence support for mxnet with
Python
URL:
https://github.com/apache/incubator-mxnet/issues/7351#issuecomment-320511578
I think this is because `sym` points to `symbol`, whereas `io` is the full
name of the module. In general, many references
kimtang commented on issue #7273: [R] call im2rec in R
URL:
https://github.com/apache/incubator-mxnet/issues/7273#issuecomment-320512362
This is indeed needed. Currently we can create the rec files using python
but to be able to stay completely in R is an advantage. All we need is a To-Do
glhfgg1024 opened a new issue #7354: Where is the "test_nn.py" file?
URL: https://github.com/apache/incubator-mxnet/issues/7354
Hi guys,
The `test_nn` is needed in the following file:
[https://github.com/apache/incubator-mxnet/blob/master/tests/python/gpu/test_operator_gpu.py#
sergeykolychev commented on issue #7353: AI-MXNet 'make test' failed.
URL:
https://github.com/apache/incubator-mxnet/issues/7353#issuecomment-320519400
@mtakanobu you're right, these tests became invalid because of the changes
made in upstream C++ and C code.
This pull https://github.co
sergeykolychev commented on issue #7353: AI-MXNet 'make test' failed.
URL:
https://github.com/apache/incubator-mxnet/issues/7353#issuecomment-320519400
@mtakanobu you're right, these tests became invalid because of the changes
made in upstream C++ and C code.
This pull https://github.co
sergeykolychev commented on issue #7353: AI-MXNet 'make test' failed.
URL:
https://github.com/apache/incubator-mxnet/issues/7353#issuecomment-320519400
@mtakanobu you're right, these test became invalid because of the changes
made in upstream C++ and C code.
This pull https://github.com
dma100180 commented on issue #7336: io.cc:54: Data and label shape in-consistent
URL:
https://github.com/apache/incubator-mxnet/issues/7336#issuecomment-320522494
Hi, yes, it works like this with an output field, but when they are 2 or
more, I change the fc4 layer with 2 neurons and train.
eric-haibin-lin commented on issue #7080: simple_bind elemwise_add with
group2ctx fails
URL:
https://github.com/apache/incubator-mxnet/issues/7080#issuecomment-320523747
@ptrendx are you referring to the place_device pass?
https://github.com/dmlc/nnvm/blob/master/src/pass/place_device.
eric-haibin-lin commented on issue #6998: 2 networks which shared same weight
on some layers
URL:
https://github.com/apache/incubator-mxnet/issues/6998#issuecomment-320524117
For param sharing I think you just need to use the shared_module option
during mod.bind when you have 2 modules sh
eric-haibin-lin commented on issue #7319: [RoadMap] Legacy issue resolution
before 1.0 release
URL:
https://github.com/apache/incubator-mxnet/issues/7319#issuecomment-320524304
kvstore has a new str interface, while the updater always uses int as the
key, which is not consistent.
https:/
eric-haibin-lin commented on issue #7034: How to switch a layer to cpu mode
while the other layers are in gpu mode.
URL:
https://github.com/apache/incubator-mxnet/issues/7034#issuecomment-320524720
You can try the `ctx2group` option during simple_bind(). For example, if you
want to specif
This is an automated email from the ASF dual-hosted git repository.
lxn2 pushed a change to branch asf-site
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git.
at ac1e48d Build on v0.10
This branch includes the following new commits:
new ac1e48d Build on v0
lxn2 closed pull request #7343: [Perl] Fix for CI
URL: https://github.com/apache/incubator-mxnet/pull/7343
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
This is an automated email from the ASF dual-hosted git repository.
lxn2 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 b0c8c6c [Perl] Fix for CI (#7343)
b0c8c6
thirdwing commented on issue #7336: io.cc:54: Data and label shape in-consistent
URL:
https://github.com/apache/incubator-mxnet/issues/7336#issuecomment-320531703
I think we have this kind of examples in our document. See
https://github.com/apache/incubator-mxnet/blob/master/R-package/vign
thirdwing commented on issue #7273: [R] call im2rec in R
URL:
https://github.com/apache/incubator-mxnet/issues/7273#issuecomment-320532040
I think this shouldn't be too hard. I always used the C++ version of im2rec
from https://github.com/apache/incubator-mxnet/blob/master/tools/im2rec.cc
mtakanobu closed issue #7353: AI-MXNet 'make test' failed.
URL: https://github.com/apache/incubator-mxnet/issues/7353
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub
mtakanobu commented on issue #7353: AI-MXNet 'make test' failed.
URL:
https://github.com/apache/incubator-mxnet/issues/7353#issuecomment-320532154
@sergeykolychev thanks.
This is an automated message from the Apache Git Serv
Lyken17 opened a new issue #7355: [API missing] mx.nd.stack & mx.sym.stack
URL: https://github.com/apache/incubator-mxnet/issues/7355
Hi there
I found `mx.nd.stack` description in [web
documents](http://mxnet.io/api/python/ndarray.html?highlight=stack#mxnet.ndarray.stack),
however,
idealboy commented on issue #7350: Multi-Training-Task on the same GPU card
URL:
https://github.com/apache/incubator-mxnet/issues/7350#issuecomment-320536048
when trainning, the following problem occurs:
?1?One trainning task on one gpu card(k40c, 12G)?batchsize=64??59 samples/ s
ec;
dokechin commented on issue #7334: cannot use ImageIter in Perl package
URL:
https://github.com/apache/incubator-mxnet/issues/7334#issuecomment-320537743
I tried and succeeded!
I appreciate prompt response.
This is an aut
beyondli commented on issue #7320: where PS-lite implementation of server
manager & hash ring ??
URL:
https://github.com/apache/incubator-mxnet/issues/7320#issuecomment-320538605
@formath , I see, thank you! hope mxnet can becom better and better!
piiswrong commented on issue #7355: [API missing] mx.nd.stack & mx.sym.stack
URL:
https://github.com/apache/incubator-mxnet/issues/7355#issuecomment-320541161
you need to install the nightly build with pip install --pre
This
piiswrong commented on issue #7347: Tensorcore conv deconv support
URL: https://github.com/apache/incubator-mxnet/pull/7347#issuecomment-320541240
Is this only enabled for fp16?
This is an automated message from the Apache Gi
szha opened a new pull request #7356: decouple record/train and add state
readers
URL: https://github.com/apache/incubator-mxnet/pull/7356
@piiswrong @zackchase
changes include:
- decouple training and testing
- add training/record state getters to allow model codes to change beha
Lyken17 commented on issue #7355: [API missing] mx.nd.stack & mx.sym.stack
URL:
https://github.com/apache/incubator-mxnet/issues/7355#issuecomment-320541536
@piiswrong Thanks. Perhaps it is better to indicate `nightly build` on web
documents? Find APIs on documents but not in code is kinda
Lyken17 commented on issue #7355: [API missing] mx.nd.stack & mx.sym.stack
URL:
https://github.com/apache/incubator-mxnet/issues/7355#issuecomment-320541536
@piiswrong Thanks. Perhaps it is better to indicate `nightly build` on web
documents? Finding APIs on documents but not in code is ki
Lyken17 commented on issue #7355: [API missing] mx.nd.stack & mx.sym.stack
URL:
https://github.com/apache/incubator-mxnet/issues/7355#issuecomment-320541536
@piiswrong Thanks. Perhaps it is better to indicate `nightly build` on web
documents? Finding APIs on documents but not in code is ki
sergeykolychev opened a new pull request #7357: [Perl] Bugfixes and sync with
python symbolic code
URL: https://github.com/apache/incubator-mxnet/pull/7357
1) Fixes for ImageIter
2) Convolutional RNN
3) Improved Visualization
4) PearsonCorrelation metric
5) Fixed tests.
--
szha opened a new pull request #7358: fix build
URL: https://github.com/apache/incubator-mxnet/pull/7358
Fix for issue reported in #7354
This is an automated message from the Apache Git Service.
To respond to the message, pl
sergeykolychev commented on issue #7334: cannot use ImageIter in Perl package
URL:
https://github.com/apache/incubator-mxnet/issues/7334#issuecomment-320542887
@dokechin Thank you, please report if you'll have any problems, I greatly
appreciate your testing.
I've just released fixed Ima
lnuchiyo opened a new issue #7359: Compile MXNet error: no matching function
for call to ?std::vector::push_back
URL: https://github.com/apache/incubator-mxnet/issues/7359
For bugs or installation issues, please provide the following information.
The more information you provide, the mor
amiol closed issue #7252: Compilation failure of module ctc_loss.cu
URL: https://github.com/apache/incubator-mxnet/issues/7252
This is an automated message from the Apache Git Service.
To respond to the message, please log o
amiol commented on issue #7252: Compilation failure of module ctc_loss.cu
URL:
https://github.com/apache/incubator-mxnet/issues/7252#issuecomment-320556729
I have been able to work around this issue by purging the APT CUDA packages
and downloading a fresh copy of CUDA 8.0 from the NVIDIA w
jjtav commented on issue #7246: [Scala] Make Module Api sync with Python
interface
URL: https://github.com/apache/incubator-mxnet/pull/7246#issuecomment-320559108
@Ldpe2G
Do you have an approximate timeframe for when you'll get to the
BucketingModule? I'll jump in on this unless you're
Ldpe2G commented on issue #7246: [Scala] Make Module Api sync with Python
interface
URL: https://github.com/apache/incubator-mxnet/pull/7246#issuecomment-320561826
@jjtav looking forward to your pr :)
This is an automated me
tianhaijie opened a new issue #7360: Compile MXNet error,I want wo checkout to
MXNet@(commit 62ecb60),by
URL: https://github.com/apache/incubator-mxnet/issues/7360
For bugs or installation issues, please provide the following information.
The more information you provide, the more likely
tornadomeet opened a new issue #7361: training speed of batch-norm is less than
batch-norm-v1
URL: https://github.com/apache/incubator-mxnet/issues/7361
if set `use_global_stats=True` then the training speed of batch-norm is less
than batch-norm-v1, so what's the change the newer batch-nor
tonyyangToBeNo1 closed issue #7280: BatchNorm compute problem
URL: https://github.com/apache/incubator-mxnet/issues/7280
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitH
edmBernard commented on issue #7360: I checkout to MXNet@(commit 62ecb60),then
compile mxnet error,but when i dont checkout to MXNet@(commit 62ecb60),compile
successfully!!
URL:
https://github.com/apache/incubator-mxnet/issues/7360#issuecomment-320574848
There is a issue to checkout an ol
edmBernard commented on issue #7350: Multi-Training-Task on the same GPU card
URL:
https://github.com/apache/incubator-mxnet/issues/7350#issuecomment-320575891
You have two training at 27 samples/sec each one ?
GPU have power processing so if one training use 100% the second need to
wai
piiswrong commented on a change in pull request #7356: decouple record/train
and add state readers
URL: https://github.com/apache/incubator-mxnet/pull/7356#discussion_r131578850
##
File path: python/mxnet/autograd.py
##
@@ -45,6 +45,28 @@ def set_training(is_train):
piiswrong commented on a change in pull request #7356: decouple record/train
and add state readers
URL: https://github.com/apache/incubator-mxnet/pull/7356#discussion_r131578832
##
File path: python/mxnet/autograd.py
##
@@ -10,21 +10,21 @@
from .symbol import _GRAD_REQ_MA
piiswrong commented on a change in pull request #7356: decouple record/train
and add state readers
URL: https://github.com/apache/incubator-mxnet/pull/7356#discussion_r131578883
##
File path: python/mxnet/autograd.py
##
@@ -45,6 +45,28 @@ def set_training(is_train):
piiswrong commented on a change in pull request #7356: decouple record/train
and add state readers
URL: https://github.com/apache/incubator-mxnet/pull/7356#discussion_r131578941
##
File path: python/mxnet/autograd.py
##
@@ -54,26 +76,28 @@ class RecordingStateScope(object)
piiswrong commented on a change in pull request #7356: decouple record/train
and add state readers
URL: https://github.com/apache/incubator-mxnet/pull/7356#discussion_r131579043
##
File path: python/mxnet/autograd.py
##
@@ -112,6 +137,34 @@ def pause(is_train=False):
piiswrong commented on a change in pull request #7356: decouple record/train
and add state readers
URL: https://github.com/apache/incubator-mxnet/pull/7356#discussion_r131579066
##
File path: python/mxnet/autograd.py
##
@@ -54,26 +76,28 @@ class RecordingStateScope(object)
piiswrong commented on a change in pull request #7356: decouple record/train
and add state readers
URL: https://github.com/apache/incubator-mxnet/pull/7356#discussion_r131579089
##
File path: python/mxnet/autograd.py
##
@@ -45,6 +45,28 @@ def set_training(is_train):
piiswrong closed pull request #7358: fix build
URL: https://github.com/apache/incubator-mxnet/pull/7358
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
UR
This is an automated email from the ASF dual-hosted git repository.
jxie pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from b0c8c6c [Perl] Fix for CI (#7343)
add 2f257e8 fix build (#7358)
No new revisions were added by this up
szha commented on issue #7358: fix build
URL: https://github.com/apache/incubator-mxnet/pull/7358#issuecomment-320580751
I wasn't able to figure out the caffe converter error yet. It succeeds for
the googlenet but fails for vgg16 and resnet50.
-
jeremiedb opened a new issue #5488: [WIP] RNN with bucketing and mask - R
package
URL: https://github.com/apache/incubator-mxnet/issues/5488
Since there wasn't an API yet for bucketing and masking in the R-package, I
started building some functionnalities to address this. Inference from it
tianhaijie commented on issue #7360: I checkout to MXNet@(commit 62ecb60),then
compile mxnet error,but when i dont checkout to MXNet@(commit 62ecb60),compile
successfully!!
URL:
https://github.com/apache/incubator-mxnet/issues/7360#issuecomment-320584501
Hi, @edmBernard Actually,when i
59 matches
Mail list logo