ZiyueHuang commented on issue #7958: Engine shutdown
URL:
https://github.com/apache/incubator-mxnet/issues/7958#issuecomment-330762854
It seems not relevant to the engine. `src/io/image_io.cc:165: Check failed:
!dst.empty()`
---
x1year commented on issue #7934: Bug of group2ctx? Wrong device placement?
URL:
https://github.com/apache/incubator-mxnet/issues/7934#issuecomment-330759842
@formath Is executor just a scheduler? What does it mean by "executor run on
this context"? What executor runs? Parameter updatin
formath commented on issue #7934: Bug of group2ctx? Wrong device placement?
URL:
https://github.com/apache/incubator-mxnet/issues/7934#issuecomment-330759033
`exe = x.bind(mx.gpu(), {}, group2ctx={"a": mx.cpu(), "b": mx.gpu()})` The
default context just indicate this executor run on this c
formath commented on issue #7934: Bug of group2ctx? Wrong device placement?
URL:
https://github.com/apache/incubator-mxnet/issues/7934#issuecomment-330759033
`exe = x.bind(mx.gpu(), {}, group2ctx={"a": mx.cpu(), "b": mx.gpu()})` The
default context is just indicate this executor run on thi
KeyKy commented on issue #7899: Need help: numpy array to mxnet ndarray is too
slow.
URL:
https://github.com/apache/incubator-mxnet/issues/7899#issuecomment-330502892
@madjam image_data is (128, 3, 224, 224) label_data is (128,)
it needs to load batch of data before doing net training.
Harold-Zhang commented on issue #7958: Engine shutdown
URL:
https://github.com/apache/incubator-mxnet/issues/7958#issuecomment-330757502
@piiswrong @szha
This is an automated message from the Apache Git Service.
To respond t
x1year commented on issue #7934: Bug of group2ctx? Wrong device placement?
URL:
https://github.com/apache/incubator-mxnet/issues/7934#issuecomment-330757441
@formath So if a symbol a has input b, and b is assigned to cpu device, then
a will also be automatically assigned to cpu device
formath commented on issue #7913: what are the dependencies between each modules
URL:
https://github.com/apache/incubator-mxnet/issues/7913#issuecomment-330755045
Functionality is clear between modules but api is not. Some modules may use
inner variables or implicit functions in the other
formath commented on issue #7959: executor not expose its inner variables to
symbol
URL: https://github.com/apache/incubator-mxnet/pull/7959#issuecomment-330753643
@piiswrong
This is an automated message from the Apache Git
szha commented on issue #7960: Dropout layer crashes jupyter kernel
URL:
https://github.com/apache/incubator-mxnet/issues/7960#issuecomment-330751674
Please paste the full error message, including stack trace.
This is an aut
zjost opened a new issue #7960: Dropout layer crashes jupyter kernel
URL: https://github.com/apache/incubator-mxnet/issues/7960
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.
formath commented on issue #7934: Bug of group2ctx? Wrong device placement?
URL:
https://github.com/apache/incubator-mxnet/issues/7934#issuecomment-330749338
Depends on the device id of its first control node or input node. Your op
may not assigned by this.
--
formath commented on issue #7942: Adam optimizer consistent with paper
URL: https://github.com/apache/incubator-mxnet/pull/7942#issuecomment-330748814
@sergeykolychev Have been merged.
@piiswrong
This is an automated messa
formath opened a new pull request #7959: executor not expose its inner
variables to symbol
URL: https://github.com/apache/incubator-mxnet/pull/7959
This is an automated message from the Apache Git Service.
To respond to the
x1year commented on issue #7934: Bug of group2ctx? Wrong device placement?
URL:
https://github.com/apache/incubator-mxnet/issues/7934#issuecomment-330746889
@formath What do you mean by "if an op can't be assigned a device id"? How
op's device id is assigned?
formath closed pull request #7932: executor not expose its inner variables to
symbol
URL: https://github.com/apache/incubator-mxnet/pull/7932
This is an automated message from the Apache Git Service.
To respond to the messa
jeremiedb commented on issue #7951: Issues with the Regression example in MXNet
R package tutorial
URL:
https://github.com/apache/incubator-mxnet/issues/7951#issuecomment-330744391
You can notice the difference in the resulting models by looking at their
graph representation:
```
g
formath commented on issue #7934: Bug of group2ctx? Wrong device placement?
URL:
https://github.com/apache/incubator-mxnet/issues/7934#issuecomment-330742175
During `PlaceDevice`, if an `op` can't be assigned a device id, its device
id will be set `0`. And to assign a default context is do
Harold-Zhang opened a new issue #7958: Engine shutdown
URL: https://github.com/apache/incubator-mxnet/issues/7958
## Environment info
Operating System:
Ubuntu 14.04
Compiler:
gcc 4.8.4
Package used (Python/R/Scala/Julia):
Python 2.7
MXNet version:
The latest version
szha commented on issue #7957: add densenet
URL: https://github.com/apache/incubator-mxnet/pull/7957#issuecomment-330736157
The symbol-based image-classification is not connected with the gluon model
zoo yet and it still has a lot of users, so there probably is value in keeping
it. We will
solin319 commented on issue #7928: Cannot reproduce the results of faster-rcnn
URL:
https://github.com/apache/incubator-mxnet/issues/7928#issuecomment-330735101
I got mAP 0.69 ,too.
Why all the gamma and beta were fixed in config?
Why "use_global_stats" in "symbol_resnet.py" was alw
qingzhouzhen commented on issue #7957: add densenet
URL: https://github.com/apache/incubator-mxnet/pull/7957#issuecomment-330733686
do you suggest me to close this PR? @szha
This is an automated message from the Apache Git S
qingzhouzhen commented on issue #7957: add densenet
URL: https://github.com/apache/incubator-mxnet/pull/7957#issuecomment-330733686
do you suggest me to close this PR?
This is an automated message from the Apache Git Service.
szha commented on issue #7957: add densenet
URL: https://github.com/apache/incubator-mxnet/pull/7957#issuecomment-330732609
@qingzhouzhen thanks for the contribution. I also made a gluon version that
can be found
[here](https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/glu
qingzhouzhen opened a new pull request #7957: add densenet
URL: https://github.com/apache/incubator-mxnet/pull/7957
Adapted it form
[https://github.com/bruinxiong/densenet.mxnet/blob/master/symbol_densenet.py](https://github.com/bruinxiong/densenet.mxnet/blob/master/symbol_densenet.py)
--
pengwangucla opened a new issue #7956: simple_bind error when using rotation
matrix & batch_dot
URL: https://github.com/apache/incubator-mxnet/issues/7956
I met simple_bind fatal error
[19:24:05] src/core/graph.cc:50: Check failed: it != node2index_.end()
&& it->first == nptr.get()
solin319 commented on issue #7893: Add barriers in kvstore init
URL: https://github.com/apache/incubator-mxnet/pull/7893#issuecomment-330719121
1. I try to solve the problem in distribute training with fp16 #7554.
In the function Push_, the data will convert to real_t before push to
ps-l
zheng-da closed pull request #7948: Use nn interface for the activation
operator.
URL: https://github.com/apache/incubator-mxnet/pull/7948
This is an automated message from the Apache Git Service.
To respond to the message,
bigmw opened a new issue #7951: Issues with the Regression example in MXNet R
package tutorial
URL: https://github.com/apache/incubator-mxnet/issues/7951
I followed the example in Neural Network with MXNet in Five Minutes tutorial:
https://github.com/apache/incubator-mxnet/blob/master/R
kevinthesun commented on issue #7950: Changed How To String to FAQ and added
Discuss Tab to the Navigation bar
URL: https://github.com/apache/incubator-mxnet/pull/7950#issuecomment-330703094
Modify navbar also requires changes in
https://github.com/apache/incubator-mxnet/blob/master/docs/_
piiswrong closed pull request #7940: Update README.md
URL: https://github.com/apache/incubator-mxnet/pull/7940
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use
piiswrong commented on issue #7698: Second order gradient and Subgraph execution
URL: https://github.com/apache/incubator-mxnet/pull/7698#issuecomment-330701652
@jermainewang Once the column store is merged you can refactor this by
putting the attributes in the column store instead of plain
piiswrong closed pull request #7815: [Perl] reduce buffer copies in
NDArray<->PDL conversions
URL: https://github.com/apache/incubator-mxnet/pull/7815
This is an automated message from the Apache Git Service.
To respond 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 15a1eca [Perl] reduce buffer copies in N
thinksanky opened a new pull request #7950: Changed How To String to FAQ and
added Discuss Tab to the Navigation bar
URL: https://github.com/apache/incubator-mxnet/pull/7950
- Provide a Discuss Navigation tabbed link on the top navigation bar.
- Changed How To to FAQ.
-
mikerossgithub opened a new pull request #7949: account for batch padding in
when updating metrics
URL: https://github.com/apache/incubator-mxnet/pull/7949
Fixes executor_group.update_metric() to take account of batch padding. If a
batch has padding only the non padded data is used to upda
sergeykolychev commented on issue #7698: Second order gradient and Subgraph
execution
URL: https://github.com/apache/incubator-mxnet/pull/7698#issuecomment-330692381
@piiswrong what is your estimation as of when this will be merged into
master ?
I've finished on porting most of gluon in
zheng-da commented on a change in pull request #7948: Use nn interface for the
activation operator.
URL: https://github.com/apache/incubator-mxnet/pull/7948#discussion_r139821949
##
File path: src/operator/nn/activation-inl.h
##
@@ -0,0 +1,172 @@
+/*
+ * Licensed to the Ap
piiswrong commented on a change in pull request #7948: Use nn interface for the
activation operator.
URL: https://github.com/apache/incubator-mxnet/pull/7948#discussion_r139807758
##
File path: tests/python/unittest/test_operator.py
##
@@ -414,6 +414,38 @@ def fsigmoid(a):
piiswrong commented on a change in pull request #7948: Use nn interface for the
activation operator.
URL: https://github.com/apache/incubator-mxnet/pull/7948#discussion_r139807677
##
File path: tests/python/unittest/test_operator.py
##
@@ -414,6 +414,38 @@ def fsigmoid(a):
piiswrong commented on a change in pull request #7948: Use nn interface for the
activation operator.
URL: https://github.com/apache/incubator-mxnet/pull/7948#discussion_r139807406
##
File path: src/operator/nn/activation.cu
##
@@ -0,0 +1,38 @@
+/*
+ * Licensed to the Apach
piiswrong commented on a change in pull request #7948: Use nn interface for the
activation operator.
URL: https://github.com/apache/incubator-mxnet/pull/7948#discussion_r139806813
##
File path: src/operator/nn/activation.cu
##
@@ -0,0 +1,38 @@
+/*
+ * Licensed to the Apach
piiswrong commented on a change in pull request #7948: Use nn interface for the
activation operator.
URL: https://github.com/apache/incubator-mxnet/pull/7948#discussion_r139806262
##
File path: src/operator/nn/activation-inl.h
##
@@ -0,0 +1,172 @@
+/*
+ * Licensed to the A
piiswrong commented on a change in pull request #7948: Use nn interface for the
activation operator.
URL: https://github.com/apache/incubator-mxnet/pull/7948#discussion_r139806385
##
File path: src/operator/nn/activation-inl.h
##
@@ -0,0 +1,172 @@
+/*
+ * Licensed to the A
zheng-da opened a new pull request #7948: Use nn interface for the activation
operator.
URL: https://github.com/apache/incubator-mxnet/pull/7948
This commit only contains the basic implementation of the activation
operator. It has been tested on CPU and GPU. But it doesn't contain the cuDN
FrancisTse8 commented on issue #7852: Trouble installing MXNet on Raspberry Pi 3
URL:
https://github.com/apache/incubator-mxnet/issues/7852#issuecomment-330656399
Thanks @larroy. Seems so obvious once you pointed that out. I was just
following the instructions blindly. I redo the commands
mnikitin commented on issue #7592: About the train-accuracy between mxnet and
caffe
URL:
https://github.com/apache/incubator-mxnet/issues/7592#issuecomment-330638089
@idealboy
Have you figured out what was wrong? I have the same problem, my
caffe-trained resnet performs better tha
mnikitin commented on issue #7592: About the train-accuracy between mxnet and
caffe
URL:
https://github.com/apache/incubator-mxnet/issues/7592#issuecomment-330638089
@idealboy
Have you figured out what was wrong? I have the same problem, my
caffe-trained resnet performs better tha
rahul003 closed pull request #7896: fix batchnorm test failure
URL: https://github.com/apache/incubator-mxnet/pull/7896
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHu
Bumblebee1964 commented on issue #7865: How to load params with LoadToMap c++
api and set device contect to GPU?
URL:
https://github.com/apache/incubator-mxnet/issues/7865#issuecomment-330632889
Found following solution (maybe not very efficient):
```
m_params_map = NDArray::L
Bumblebee1964 closed issue #7865: How to load params with LoadToMap c++ api and
set device contect to GPU?
URL: https://github.com/apache/incubator-mxnet/issues/7865
This is an automated message from the Apache Git Service.
cjolivier01 opened a new pull request #7903: Refactor AdaGrad optimizer to
support sparse tensors
URL: https://github.com/apache/incubator-mxnet/pull/7903
Refactor AdaGrad optimizer to support sparse tensors
Add sparse support for _plus_scalar, _minus_scalar, clip
-
cjolivier01 closed pull request #7903: Refactor AdaGrad optimizer to support
sparse tensors
URL: https://github.com/apache/incubator-mxnet/pull/7903
This is an automated message from the Apache Git Service.
To respond to th
eric-haibin-lin opened a new pull request #7947: [WIP] Refactor infer storage
function for sparse operators.
URL: https://github.com/apache/incubator-mxnet/pull/7947
This is an automated message from the Apache Git Service.
eric-haibin-lin commented on a change in pull request #7903: Refactor AdaGrad
optimizer to support sparse tensors
URL: https://github.com/apache/incubator-mxnet/pull/7903#discussion_r139767993
##
File path: tests/python/unittest/test_module.py
##
@@ -462,74 +462,105 @@ def
piiswrong commented on a change in pull request #7892: allow users to specify
repo url for downloadable contents
URL: https://github.com/apache/incubator-mxnet/pull/7892#discussion_r139767382
##
File path: python/mxnet/gluon/model_zoo/vision/alexnet.py
##
@@ -76,9 +77,13 @
SumNeuron opened a new issue #7946: Defining Residual Convolution Structures in
Block from gluon Module
URL: https://github.com/apache/incubator-mxnet/issues/7946
Please see this [S.O.
post](https://stackoverflow.com/questions/46306782/mxnet-python3-defining-a-residual-convolution-structur
cjolivier01 commented on a change in pull request #7903: Refactor AdaGrad
optimizer to support sparse tensors
URL: https://github.com/apache/incubator-mxnet/pull/7903#discussion_r139765735
##
File path: tests/python/unittest/test_module.py
##
@@ -462,74 +462,105 @@ def tes
eric-haibin-lin commented on a change in pull request #7893: Add barriers in
kvstore init
URL: https://github.com/apache/incubator-mxnet/pull/7893#discussion_r139763399
##
File path: src/kvstore/kvstore_dist.h
##
@@ -144,9 +144,9 @@ class KVStoreDist : public KVStoreLocal
eric-haibin-lin commented on issue #7893: Add barriers in kvstore init
URL: https://github.com/apache/incubator-mxnet/pull/7893#issuecomment-330613450
Are you using multi-machine with GPUs? I'm not getting the setup here. Could
you post a code snippet for others to reproduce?
---
eric-haibin-lin commented on a change in pull request #7903: Refactor AdaGrad
optimizer to support sparse tensors
URL: https://github.com/apache/incubator-mxnet/pull/7903#discussion_r139760961
##
File path: tests/python/unittest/test_module.py
##
@@ -462,74 +462,105 @@ def
eric-haibin-lin commented on a change in pull request #7903: Refactor AdaGrad
optimizer to support sparse tensors
URL: https://github.com/apache/incubator-mxnet/pull/7903#discussion_r139760826
##
File path: tests/python/unittest/test_module.py
##
@@ -462,74 +462,105 @@ def
eric-haibin-lin commented on a change in pull request #7903: Refactor AdaGrad
optimizer to support sparse tensors
URL: https://github.com/apache/incubator-mxnet/pull/7903#discussion_r139759839
##
File path: python/mxnet/optimizer.py
##
@@ -664,27 +667,45 @@ class AdaGrad(O
eric-haibin-lin commented on a change in pull request #7903: Refactor AdaGrad
optimizer to support sparse tensors
URL: https://github.com/apache/incubator-mxnet/pull/7903#discussion_r139760494
##
File path: tests/python/unittest/test_sparse_operator.py
##
@@ -697,6 +697,19
sergeykolychev commented on issue #7942: Adam optimizer consistent with paper
URL: https://github.com/apache/incubator-mxnet/pull/7942#issuecomment-330611582
@formath please see https://github.com/formath/mxnet/pull/1
This is
Bumblebee1964 opened a new issue #7945: Has anyone a guide for installing MxNet
with Python on Windows 10 / VS2017 toolset (including Anaconda2)?
URL: https://github.com/apache/incubator-mxnet/issues/7945
For bugs or installation issues, please provide the following information.
The more
eric-haibin-lin commented on issue #7935: Zeroes CSR still needs a valid
row_pointer array.
URL: https://github.com/apache/incubator-mxnet/pull/7935#issuecomment-330608771
I agree that we should return a CSR array with valid indptr array for that
zeros() function. Should we also update the
cjolivier01 commented on a change in pull request #7903: Refactor AdaGrad
optimizer to support sparse tensors
URL: https://github.com/apache/incubator-mxnet/pull/7903#discussion_r139756939
##
File path: python/mxnet/optimizer.py
##
@@ -665,26 +667,46 @@ class AdaGrad(Optim
cjolivier01 commented on a change in pull request #7903: Refactor AdaGrad
optimizer to support sparse tensors
URL: https://github.com/apache/incubator-mxnet/pull/7903#discussion_r139756939
##
File path: python/mxnet/optimizer.py
##
@@ -665,26 +667,46 @@ class AdaGrad(Optim
Bumblebee1964 commented on issue #7835: C++ batchnorm CPU version training
well, but validation wrong
URL:
https://github.com/apache/incubator-mxnet/issues/7835#issuecomment-330607688
Having same problem. Noticed that the Mathematica version 11.2 (using MxNet
under the hood) is not able t
cjolivier01 commented on issue #7903: Refactor AdaGrad optimizer to support
sparse tensors
URL: https://github.com/apache/incubator-mxnet/pull/7903#issuecomment-330574013
different/more selective scatter behavior can be done in a different PR. I
don't have the bandwidth for that now.
---
cjolivier01 commented on a change in pull request #7903: Refactor AdaGrad
optimizer to support sparse tensors
URL: https://github.com/apache/incubator-mxnet/pull/7903#discussion_r139724976
##
File path: python/mxnet/optimizer.py
##
@@ -670,21 +672,37 @@ def __init__(self,
chaitanyabjoshi commented on issue #7524: Is there a tutorial of using mxnet R
LSTM for time series forecasting?
URL:
https://github.com/apache/incubator-mxnet/issues/7524#issuecomment-330571650
Unfortunately, poor documentation of mxnetR library is causing a lot of
trouble to many devel
kli-nlpr opened a new pull request #7944: Update softmax_output.cc
URL: https://github.com/apache/incubator-mxnet/pull/7944
I think that 'Computes the gradient of cross entropy loss with respect to
softmax output' is confusing.
So I change it to
'Computes the gradient of cross entropy
SumNeuron opened a new issue #7943: Request: Visualization for Gluon neural
networks
URL: https://github.com/apache/incubator-mxnet/issues/7943
[Here](https://mxnet.incubator.apache.org/how_to/visualize_graph.html) shows
how to visualize mx.Symbol networks.
`mx.viz.plot_network(net)
hhyasdf closed issue #7924: get stuck training on multiple machines error
URL: https://github.com/apache/incubator-mxnet/issues/7924
This is an automated message from the Apache Git Service.
To respond to the message, please
chaitanyabjoshi commented on issue #5147: Correct array shape() for mx.lstm
URL:
https://github.com/apache/incubator-mxnet/issues/5147#issuecomment-330542596
This issue should not be closed as documentation (even in version 0.11) is
not fixed and is so cryptic that no meaning can be made o
KeyKy commented on issue #7899: Need help: numpy array to mxnet ndarray is too
slow.
URL:
https://github.com/apache/incubator-mxnet/issues/7899#issuecomment-330502892
image_data is (128, 3, 224, 224) label_data is (128,)
it needs to load batch of data before doing net training. Forward
KeyKy commented on issue #7899: Need help: numpy array to mxnet ndarray is too
slow.
URL:
https://github.com/apache/incubator-mxnet/issues/7899#issuecomment-330502892
image_data is (128, 3, 224, 224) label_data is (128,)
it needs to load batch of data before doing net training. Forward
KeyKy commented on issue #7899: Need help: numpy array to mxnet ndarray is too
slow.
URL:
https://github.com/apache/incubator-mxnet/issues/7899#issuecomment-330502892
image_data is (128, 3, 224, 224) label_data is (128,)
it needs to load batch of data before doing net forward.
---
KeyKy commented on issue #7899: Need help: numpy array to mxnet ndarray is too
slow.
URL:
https://github.com/apache/incubator-mxnet/issues/7899#issuecomment-330502892
image_data is (128, 3, 224, 224) label_data is (128,)
Because it needs to load batch of data before doing net forward.
Prasad9 closed issue #7925: Resize image to fixed size
URL: https://github.com/apache/incubator-mxnet/issues/7925
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and
Prasad9 commented on issue #7925: Resize image to fixed size
URL:
https://github.com/apache/incubator-mxnet/issues/7925#issuecomment-330486842
Thanks @ZiyueHuang and @jiarenyf . Figured it out how to use it with
ImageIter. But for the particular usecase I wanted, Augmenter with ImageIter
formath opened a new pull request #7942: Adam optimizer consistent with paper
URL: https://github.com/apache/incubator-mxnet/pull/7942
This is an automated message from the Apache Git Service.
To respond to the message, plea
SumNeuron commented on issue #1281: how to prepare data for mxnet.io.NDArrayIter
URL:
https://github.com/apache/incubator-mxnet/issues/1281#issuecomment-330478164
I also am confused about prepping data for MXNet (although this is more
general)
Even the method you
[recommend](ht
SumNeuron commented on issue #7900: Request: finish python gpu enabled guide
for install
URL:
https://github.com/apache/incubator-mxnet/issues/7900#issuecomment-330477727
after running python3 setup.py develop --user things seem to work
---
SumNeuron closed issue #7900: Request: finish python gpu enabled guide for
install
URL: https://github.com/apache/incubator-mxnet/issues/7900
This is an automated message from the Apache Git Service.
To respond to the messa
87 matches
Mail list logo