[GitHub] lxn2 commented on issue #7379: Add license header

2017-08-08 Thread git
lxn2 commented on issue #7379: Add license header URL: https://github.com/apache/incubator-mxnet/pull/7379#issuecomment-321092807 Can we merge? This is an automated message from the Apache Git Service. To respond to the

[GitHub] eric-haibin-lin commented on issue #7082: Sparse Tensor: request for reviews

2017-08-08 Thread git
eric-haibin-lin commented on issue #7082: Sparse Tensor: request for reviews URL: https://github.com/apache/incubator-mxnet/pull/7082#issuecomment-321096175 Hi everyone, I've updated the branch to include code changes to address comments from reviewers (except the discussion re:

[GitHub] eric-haibin-lin closed pull request #7015: RowSparse pull/push

2017-08-08 Thread git
eric-haibin-lin closed pull request #7015: RowSparse pull/push URL: https://github.com/apache/incubator-mxnet/pull/7015 This is an automated message from the Apache Git Service. To respond to the message, please log on

[GitHub] joey2014 commented on a change in pull request #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs

2017-08-08 Thread git
joey2014 commented on a change in pull request #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs URL: https://github.com/apache/incubator-mxnet/pull/7381#discussion_r132065167 ## File path: tools/caffe_converter/convert_symbol.py ## @@ -128,6

[incubator-mxnet] branch master updated: decouple record/train and add state readers (#7356)

2017-08-08 Thread jxie
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 07661ae decouple record/train and add

[GitHub] piiswrong closed pull request #7356: decouple record/train and add state readers

2017-08-08 Thread git
piiswrong closed pull request #7356: decouple record/train and add state readers URL: https://github.com/apache/incubator-mxnet/pull/7356 This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] piiswrong closed pull request #7388: Update conv_layers.py

2017-08-08 Thread git
piiswrong closed pull request #7388: Update conv_layers.py URL: https://github.com/apache/incubator-mxnet/pull/7388 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub

[incubator-mxnet] branch master updated: Update conv_layers.py (#7388)

2017-08-08 Thread jxie
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 4c13b96 Update conv_layers.py (#7388)

[GitHub] zhreshold commented on issue #7051: Non-maximum-suppression op

2017-08-08 Thread git
zhreshold commented on issue #7051: Non-maximum-suppression op URL: https://github.com/apache/incubator-mxnet/pull/7051#issuecomment-320128610 ``` cpp-package/include/mxnet-cpp/lr_scheduler.h:7: #ifndef header guard has wrong style, please use: MXNET_CPP_LR_SCHEDULER_H_

[GitHub] stefanhenneking commented on a change in pull request #7226: Extending the GPU dot operator

2017-08-08 Thread git
stefanhenneking commented on a change in pull request #7226: Extending the GPU dot operator URL: https://github.com/apache/incubator-mxnet/pull/7226#discussion_r132058753 ## File path: src/operator/tensor/dot-inl.cuh ## @@ -9,66 +9,163 @@ #include #include +#include

[GitHub] stefanhenneking commented on a change in pull request #7226: Extending the GPU dot operator

2017-08-08 Thread git
stefanhenneking commented on a change in pull request #7226: Extending the GPU dot operator URL: https://github.com/apache/incubator-mxnet/pull/7226#discussion_r132058753 ## File path: src/operator/tensor/dot-inl.cuh ## @@ -9,66 +9,163 @@ #include #include +#include

[GitHub] piiswrong commented on issue #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs

2017-08-08 Thread git
piiswrong commented on issue #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs URL: https://github.com/apache/incubator-mxnet/pull/7381#issuecomment-321118391 Thanks. Could you revert this PR https://github.com/apache/incubator-mxnet/pull/7369/files here to

[GitHub] qingzhouzhen closed issue #7386: How could I define function for GPU only?

2017-08-08 Thread git
qingzhouzhen closed issue #7386: How could I define function for GPU only? URL: https://github.com/apache/incubator-mxnet/issues/7386 This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] LakeCarrot commented on issue #5837: Error caused by two different dependencies in protobuf version when using both distributed mode and tensorboard

2017-08-08 Thread git
LakeCarrot commented on issue #5837: Error caused by two different dependencies in protobuf version when using both distributed mode and tensorboard URL: https://github.com/apache/incubator-mxnet/issues/5837#issuecomment-321120862 I met the same problem, MXNet v0.10.0-support and Ubuntu

[GitHub] szha commented on issue #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs

2017-08-08 Thread git
szha commented on issue #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs URL: https://github.com/apache/incubator-mxnet/pull/7381#issuecomment-321121243 Thanks, @joey2014. Please revert the removed test cases.

[GitHub] szha commented on issue #7264: gluon conv rnns

2017-08-08 Thread git
szha commented on issue #7264: gluon conv rnns URL: https://github.com/apache/incubator-mxnet/pull/7264#issuecomment-321104507 @ykim362 I tried adding dim=5 extension in MKL but didn't get it to work. For now I will use 1D conv as test.

[GitHub] piiswrong commented on issue #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs

2017-08-08 Thread git
piiswrong commented on issue #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs URL: https://github.com/apache/incubator-mxnet/pull/7381#issuecomment-321118391 Thanks. Could you revert this PR https://github.com/apache/incubator-mxnet/pull/7369/files to make

[GitHub] thirdwing opened a new pull request #7389: [R] im2rec in R. close #7273

2017-08-08 Thread git
thirdwing opened a new pull request #7389: [R] im2rec in R. close #7273 URL: https://github.com/apache/incubator-mxnet/pull/7389 This is an automated message from the Apache Git Service. To respond to the message, please

[GitHub] thirdwing commented on issue #7375: Can I set instance weight when training?

2017-08-08 Thread git
thirdwing commented on issue #7375: Can I set instance weight when training? URL: https://github.com/apache/incubator-mxnet/issues/7375#issuecomment-321101775 @regzhuce The output of Makeloss is the gradient. See

[incubator-mxnet] branch master updated: Clarify definition of cross-entropy metric in the documentation (clean up PR #7291) (#7365)

2017-08-08 Thread jxie
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 44c2bfe Clarify definition of

[GitHub] piiswrong closed pull request #7365: Clarify definition of cross-entropy metric in the documentation (clean up PR #7291)

2017-08-08 Thread git
piiswrong closed pull request #7365: Clarify definition of cross-entropy metric in the documentation (clean up PR #7291) URL: https://github.com/apache/incubator-mxnet/pull/7365 This is an automated message from the Apache

[GitHub] LakeCarrot commented on issue #5837: Error caused by two different dependencies in protobuf version when using both distributed mode and tensorboard

2017-08-08 Thread git
LakeCarrot commented on issue #5837: Error caused by two different dependencies in protobuf version when using both distributed mode and tensorboard URL: https://github.com/apache/incubator-mxnet/issues/5837#issuecomment-321120862 I met the same problem, with MXNet v0.10.0support and

[GitHub] joey2014 commented on issue #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs

2017-08-08 Thread git
joey2014 commented on issue #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs URL: https://github.com/apache/incubator-mxnet/pull/7381#issuecomment-321120896 Yes, I have test it locally. The fix works INFO:root:test performance of model: vgg-16

[GitHub] stefanhenneking commented on a change in pull request #7226: Extending the GPU dot operator

2017-08-08 Thread git
stefanhenneking commented on a change in pull request #7226: Extending the GPU dot operator URL: https://github.com/apache/incubator-mxnet/pull/7226#discussion_r132040459 ## File path: src/operator/tensor/dot-inl.cuh ## @@ -353,27 +640,308 @@ inline void

[GitHub] szha commented on issue #7264: gluon conv rnns

2017-08-08 Thread git
szha commented on issue #7264: gluon conv rnns URL: https://github.com/apache/incubator-mxnet/pull/7264#issuecomment-321091550 @ykim362 looks like a problem in MKL concat. The following code causes the same error when using MKL version to do 5 dim concat ```

[GitHub] nikrao opened a new pull request #7390: adding ranking metrics (precision/recall) at position K.

2017-08-08 Thread git
nikrao opened a new pull request #7390: adding ranking metrics (precision/recall) at position K. URL: https://github.com/apache/incubator-mxnet/pull/7390 The file takes in data in dictionary format, for both predictions and ground truth. Ranking problems are very common in

[GitHub] ZihengJiang commented on issue #7319: [RoadMap] Legacy issue resolution before 1.0 release

2017-08-08 Thread git
ZihengJiang commented on issue #7319: [RoadMap] Legacy issue resolution before 1.0 release URL: https://github.com/apache/incubator-mxnet/issues/7319#issuecomment-321099604 Should enable multiple times resource requests

[GitHub] piiswrong closed pull request #7379: Add license header

2017-08-08 Thread git
piiswrong closed pull request #7379: Add license header URL: https://github.com/apache/incubator-mxnet/pull/7379 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and

[GitHub] joey2014 commented on issue #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs

2017-08-08 Thread git
joey2014 commented on issue #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs URL: https://github.com/apache/incubator-mxnet/pull/7381#issuecomment-321120896 Yes, I have already tested it locally. The fix works INFO:root:test performance of model: vgg-16

[GitHub] LakeCarrot commented on issue #5837: Error caused by two different dependencies in protobuf version when using both distributed mode and tensorboard

2017-08-08 Thread git
LakeCarrot commented on issue #5837: Error caused by two different dependencies in protobuf version when using both distributed mode and tensorboard URL: https://github.com/apache/incubator-mxnet/issues/5837#issuecomment-321120862 I met the same problem, with MXNet v0.10.0-support and

[GitHub] szha commented on issue #7264: gluon conv rnns

2017-08-08 Thread git
szha commented on issue #7264: gluon conv rnns URL: https://github.com/apache/incubator-mxnet/pull/7264#issuecomment-320868244 Somehow the tests fail in MKLConcatOp in the MKL variant. @ykim362 any idea what this is? This

[GitHub] buaacszj opened a new issue #7383: How to build a library without the BLAS dependency?

2017-08-08 Thread git
buaacszj opened a new issue #7383: How to build a library without the BLAS dependency? URL: https://github.com/apache/incubator-mxnet/issues/7383 As I saw from mxnet document "Deep Learning in a Single File for Smart Devices", there is a word saying "MXNet provides an amalgamation script

[GitHub] edmBernard commented on issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler?

2017-08-08 Thread git
edmBernard commented on issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler? URL: https://github.com/apache/incubator-mxnet/issues/7380#issuecomment-320886272 I don't know for gpu driver and cuda version. But cudnn version

[GitHub] joey2014 commented on issue #7368: Caffe converter test fails, causing CI to halt for all PRs

2017-08-08 Thread git
joey2014 commented on issue #7368: Caffe converter test fails, causing CI to halt for all PRs URL: https://github.com/apache/incubator-mxnet/issues/7368#issuecomment-320869584 The layer input name in some cases is changed which causes the layer added to group by mistake, for example, in

[GitHub] joey2014 opened a new pull request #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs

2017-08-08 Thread git
joey2014 opened a new pull request #7381: fix [#7368] Caffe converter test fails, causing CI to halt for all PRs URL: https://github.com/apache/incubator-mxnet/pull/7381 This is an automated message from the Apache Git

[GitHub] peschn opened a new issue #7382: GRU not working with layout 'NTC'

2017-08-08 Thread git
peschn opened a new issue #7382: GRU not working with layout 'NTC' URL: https://github.com/apache/incubator-mxnet/issues/7382 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.

[GitHub] szha commented on issue #7356: decouple record/train and add state readers

2017-08-08 Thread git
szha commented on issue #7356: decouple record/train and add state readers URL: https://github.com/apache/incubator-mxnet/pull/7356#issuecomment-320867252 Doc for autograd can be found at http://mxnet-doc.s3-website-us-east-1.amazonaws.com/api/python/autograd.html

[GitHub] edmBernard commented on issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler?

2017-08-08 Thread git
edmBernard commented on issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler? URL: https://github.com/apache/incubator-mxnet/issues/7380#issuecomment-320881324 maybe a stupid question but did you write model checkpoint at the

[GitHub] coolbrain commented on issue #6999: bgr or rgb input to vgg16?

2017-08-08 Thread git
coolbrain commented on issue #6999: bgr or rgb input to vgg16? URL: https://github.com/apache/incubator-mxnet/issues/6999#issuecomment-320878728 I want to know this, too. Is the vgg16 model trained by mxnex or just converted from caffe model?

[GitHub] githubfragments commented on issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler?

2017-08-08 Thread git
githubfragments commented on issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler? URL: https://github.com/apache/incubator-mxnet/issues/7380#issuecomment-320883014 Model checkpoint is written every 10 epoch. Both of the hard

[GitHub] githubfragments opened a new issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler?

2017-08-08 Thread git
githubfragments opened a new issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler? URL: https://github.com/apache/incubator-mxnet/issues/7380 For bugs or installation issues, please provide the following information. The

[GitHub] githubfragments commented on issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler?

2017-08-08 Thread git
githubfragments commented on issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler? URL: https://github.com/apache/incubator-mxnet/issues/7380#issuecomment-320863746 CPU is OK. It is Intel? Xeon(R) CPU E5-2620 v4 @ 2.10GHz ? 16

[GitHub] novioleo commented on issue #5224: include/dmlc/logging.h:235: [00:59:21] src/io/local_filesys.cc:154: Check failed: allow_null LocalFileSystem: fail to open "model/det1-symbol.json"

2017-08-08 Thread git
novioleo commented on issue #5224: include/dmlc/logging.h:235: [00:59:21] src/io/local_filesys.cc:154: Check failed: allow_null LocalFileSystem: fail to open "model/det1-symbol.json" URL: https://github.com/apache/incubator-mxnet/issues/5224#issuecomment-320858716 @ysh329 i know it's

[GitHub] jeremiedb commented on issue #7375: Can I set instance weight when training?

2017-08-08 Thread git
jeremiedb commented on issue #7375: Can I set instance weight when training? URL: https://github.com/apache/incubator-mxnet/issues/7375#issuecomment-320860659 The default behavior when using `predict` function on a model with MakeLoss is to get the inference on the latest layer, which is

[GitHub] edmBernard commented on issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler?

2017-08-08 Thread git
edmBernard commented on issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler? URL: https://github.com/apache/incubator-mxnet/issues/7380#issuecomment-320862930 Are you sure the CPU is not the bottleneck ? I just see there

[GitHub] szha opened a new pull request #7392: update for cudnn7 for rnn changes

2017-08-08 Thread git
szha opened a new pull request #7392: update for cudnn7 for rnn changes URL: https://github.com/apache/incubator-mxnet/pull/7392 This is to reflect a change in `cudnnSetRNNDescriptor` in cudnn 7, and should enable building with cudnn7.

[GitHub] nikrao commented on issue #7390: adding ranking metrics (precision/recall) at position K.

2017-08-08 Thread git
nikrao commented on issue #7390: adding ranking metrics (precision/recall) at position K. URL: https://github.com/apache/incubator-mxnet/pull/7390#issuecomment-321143824 fixed. pylint checks pass now. This is an

[GitHub] szha commented on issue #7304: [WIP] gluon bce & ctc losses

2017-08-08 Thread git
szha commented on issue #7304: [WIP] gluon bce & ctc losses URL: https://github.com/apache/incubator-mxnet/pull/7304#issuecomment-321144365 Updating the interface now for potential integration with cudnn7. Do not merge yet.

[GitHub] piiswrong commented on a change in pull request #7347: Tensorcore conv deconv support

2017-08-08 Thread git
piiswrong commented on a change in pull request #7347: Tensorcore conv deconv support URL: https://github.com/apache/incubator-mxnet/pull/7347#discussion_r132088433 ## File path: src/operator/cudnn_algoreg-inl.h ## @@ -40,12 +65,17 @@ class CuDNNAlgoReg { oss <<

[GitHub] DickJC123 commented on a change in pull request #7347: Tensorcore conv deconv support

2017-08-08 Thread git
DickJC123 commented on a change in pull request #7347: Tensorcore conv deconv support URL: https://github.com/apache/incubator-mxnet/pull/7347#discussion_r132093608 ## File path: src/operator/cudnn_algoreg-inl.h ## @@ -40,12 +65,17 @@ class CuDNNAlgoReg { oss <<

[GitHub] piiswrong commented on a change in pull request #7393: add depthwise convolution's gpu version optimization

2017-08-08 Thread git
piiswrong commented on a change in pull request #7393: add depthwise convolution's gpu version optimization URL: https://github.com/apache/incubator-mxnet/pull/7393#discussion_r132093497 ## File path: src/operator/convolution-inl.h ## @@ -97,6 +98,8 @@ struct

[GitHub] piiswrong commented on a change in pull request #7393: add depthwise convolution's gpu version optimization

2017-08-08 Thread git
piiswrong commented on a change in pull request #7393: add depthwise convolution's gpu version optimization URL: https://github.com/apache/incubator-mxnet/pull/7393#discussion_r132094688 ## File path: tests/python/unittest/test_operator.py ## @@ -955,6 +955,40 @@ def

[GitHub] kkk669 opened a new pull request #7391: Update CONTRIBUTORS.md

2017-08-08 Thread git
kkk669 opened a new pull request #7391: Update CONTRIBUTORS.md URL: https://github.com/apache/incubator-mxnet/pull/7391 Because my patches have been merged in #6908 and #6927. This is an automated message from the Apache Git

[GitHub] piiswrong commented on a change in pull request #7347: Tensorcore conv deconv support

2017-08-08 Thread git
piiswrong commented on a change in pull request #7347: Tensorcore conv deconv support URL: https://github.com/apache/incubator-mxnet/pull/7347#discussion_r132088512 ## File path: src/operator/cudnn_algoreg-inl.h ## @@ -40,12 +65,17 @@ class CuDNNAlgoReg { oss <<

[GitHub] piiswrong commented on a change in pull request #7347: Tensorcore conv deconv support

2017-08-08 Thread git
piiswrong commented on a change in pull request #7347: Tensorcore conv deconv support URL: https://github.com/apache/incubator-mxnet/pull/7347#discussion_r132088433 ## File path: src/operator/cudnn_algoreg-inl.h ## @@ -40,12 +65,17 @@ class CuDNNAlgoReg { oss <<

[GitHub] piiswrong commented on issue #7322: add verification to gluon dataset

2017-08-08 Thread git
piiswrong commented on issue #7322: add verification to gluon dataset URL: https://github.com/apache/incubator-mxnet/pull/7322#issuecomment-321156648 Please add tests for mnist and cifar10 This is an automated message from

[GitHub] piiswrong closed pull request #7391: Update CONTRIBUTORS.md

2017-08-08 Thread git
piiswrong closed pull request #7391: Update CONTRIBUTORS.md URL: https://github.com/apache/incubator-mxnet/pull/7391 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub

[GitHub] piiswrong commented on a change in pull request #7393: add depthwise convolution's gpu version optimization

2017-08-08 Thread git
piiswrong commented on a change in pull request #7393: add depthwise convolution's gpu version optimization URL: https://github.com/apache/incubator-mxnet/pull/7393#discussion_r132093964 ## File path: src/operator/convolution.cu ## @@ -45,6 +47,18 @@ Operator*

[GitHub] zhreshold commented on issue #7376: MXbox -- a simple and flexible vision toolbox for mxnet framework.

2017-08-08 Thread git
zhreshold commented on issue #7376: MXbox -- a simple and flexible vision toolbox for mxnet framework. URL: https://github.com/apache/incubator-mxnet/issues/7376#issuecomment-321130007 Good intuition and hitting the point, but seems like there's a lot of duplicated function with gluon

[GitHub] eric-haibin-lin commented on a change in pull request #7390: adding ranking metrics (precision/recall) at position K.

2017-08-08 Thread git
eric-haibin-lin commented on a change in pull request #7390: adding ranking metrics (precision/recall) at position K. URL: https://github.com/apache/incubator-mxnet/pull/7390#discussion_r132081575 ## File path: python/mxnet/ranking_metrics.py ## @@ -0,0 +1,45 @@ +""" +A

[GitHub] szha closed pull request #7392: update for cudnn7 for rnn changes

2017-08-08 Thread git
szha closed pull request #7392: update for cudnn7 for rnn changes URL: https://github.com/apache/incubator-mxnet/pull/7392 This is an automated message from the Apache Git Service. To respond to the message, please log on

[GitHub] szha commented on issue #7392: update for cudnn7 for rnn changes

2017-08-08 Thread git
szha commented on issue #7392: update for cudnn7 for rnn changes URL: https://github.com/apache/incubator-mxnet/pull/7392#issuecomment-321150940 #7347 already covers this This is an automated message from the Apache Git

[GitHub] piiswrong commented on a change in pull request #7393: add depthwise convolution's gpu version optimization

2017-08-08 Thread git
piiswrong commented on a change in pull request #7393: add depthwise convolution's gpu version optimization URL: https://github.com/apache/incubator-mxnet/pull/7393#discussion_r132094422 ## File path: tests/python/unittest/test_operator.py ## @@ -955,6 +955,40 @@ def

[GitHub] crazy-cat opened a new pull request #7393: add depthwise convolution's gpu version optimization

2017-08-08 Thread git
crazy-cat opened a new pull request #7393: add depthwise convolution's gpu version optimization URL: https://github.com/apache/incubator-mxnet/pull/7393 As the cudnn is not optimized for depthwise convolution, we optimized the gpu version of depthwise 2D convolution. The training

[GitHub] piiswrong commented on a change in pull request #7393: add depthwise convolution's gpu version optimization

2017-08-08 Thread git
piiswrong commented on a change in pull request #7393: add depthwise convolution's gpu version optimization URL: https://github.com/apache/incubator-mxnet/pull/7393#discussion_r132094170 ## File path: tests/python/unittest/test_operator.py ## @@ -955,6 +955,40 @@ def

[GitHub] piiswrong commented on a change in pull request #7393: add depthwise convolution's gpu version optimization

2017-08-08 Thread git
piiswrong commented on a change in pull request #7393: add depthwise convolution's gpu version optimization URL: https://github.com/apache/incubator-mxnet/pull/7393#discussion_r132094838 ## File path: tests/python/unittest/test_operator.py ## @@ -955,6 +955,40 @@ def

[GitHub] KellenSunderland closed pull request #7188: Do not Docker cache git clones

2017-08-08 Thread git
KellenSunderland closed pull request #7188: Do not Docker cache git clones URL: https://github.com/apache/incubator-mxnet/pull/7188 This is an automated message from the Apache Git Service. To respond to the message, please

[GitHub] alfredtofu opened a new issue #7387: c_predict_api doesnot support multi predictor

2017-08-08 Thread git
alfredtofu opened a new issue #7387: c_predict_api doesnot support multi predictor URL: https://github.com/apache/incubator-mxnet/issues/7387 It seems that c_predict_api doesnot support create multi predictor on different devices(gpus) in one program and it will raise `invalid device

[GitHub] train-test-laura commented on issue #7364: MxnetR chunk-wise neural nets

2017-08-08 Thread git
train-test-laura commented on issue #7364: MxnetR chunk-wise neural nets URL: https://github.com/apache/incubator-mxnet/issues/7364#issuecomment-320941845 Thanks for your reply. I have found that page already pretty helpful, but there are a few questions I have - maybe you or

[GitHub] train-test-laura commented on issue #7364: MxnetR chunk-wise neural nets

2017-08-08 Thread git
train-test-laura commented on issue #7364: MxnetR chunk-wise neural nets URL: https://github.com/apache/incubator-mxnet/issues/7364#issuecomment-320941845 Thanks for your reply. I have found that page already pretty helpful, but there are a few questions I have - maybe you or

[GitHub] regzhuce commented on issue #7375: Can I set instance weight when training?

2017-08-08 Thread git
regzhuce commented on issue #7375: Can I set instance weight when training? URL: https://github.com/apache/incubator-mxnet/issues/7375#issuecomment-320948290 It's so trivial. Hopefully we can have more graceful approach encapuslated. That would be very nice.

[GitHub] solin319 opened a new issue #7384: Parameter 'dtype' in the function 'ImageRecordIter' doesn't work

2017-08-08 Thread git
solin319 opened a new issue #7384: Parameter 'dtype' in the function 'ImageRecordIter' doesn't work URL: https://github.com/apache/incubator-mxnet/issues/7384 In mxnet-0.10 version, I add 'dtype = float16', 'dtype = float64', 'dtype = uint8' in the function 'ImageRecordIter' in

[GitHub] rahul003 commented on issue #7321: fixes broken compilation by including tensor_blob

2017-08-08 Thread git
rahul003 commented on issue #7321: fixes broken compilation by including tensor_blob URL: https://github.com/apache/incubator-mxnet/pull/7321#issuecomment-320741353 @piiswrong So what needs to be done to ensure mxnet compiles fine?

[GitHub] eric-haibin-lin commented on a change in pull request #7226: Extending the GPU dot operator

2017-08-08 Thread git
eric-haibin-lin commented on a change in pull request #7226: Extending the GPU dot operator URL: https://github.com/apache/incubator-mxnet/pull/7226#discussion_r131993234 ## File path: src/operator/tensor/dot-inl.cuh ## @@ -353,27 +640,308 @@ inline void

[GitHub] zhanghang1989 commented on issue #7377: gluon.nn.Conv2D initialization int stride not handled correctly

2017-08-08 Thread git
zhanghang1989 commented on issue #7377: gluon.nn.Conv2D initialization int stride not handled correctly URL: https://github.com/apache/incubator-mxnet/issues/7377#issuecomment-321041311 ```python import mxnet as mx from mxnet import gluon from mxnet.gluon import nn

[GitHub] rahul003 commented on a change in pull request #7321: fixes broken compilation by including tensor_blob

2017-08-08 Thread git
rahul003 commented on a change in pull request #7321: fixes broken compilation by including tensor_blob URL: https://github.com/apache/incubator-mxnet/pull/7321#discussion_r131995423 ## File path: src/io/inst_vector.h ## @@ -12,6 +12,7 @@ #include #include #include

[GitHub] piiswrong opened a new pull request #7388: Update conv_layers.py

2017-08-08 Thread git
piiswrong opened a new pull request #7388: Update conv_layers.py URL: https://github.com/apache/incubator-mxnet/pull/7388 This is an automated message from the Apache Git Service. To respond to the message, please log on

[GitHub] zhanghang1989 closed issue #7377: gluon.nn.Conv2D initialization int stride not handled correctly

2017-08-08 Thread git
zhanghang1989 closed issue #7377: gluon.nn.Conv2D initialization int stride not handled correctly URL: https://github.com/apache/incubator-mxnet/issues/7377 This is an automated message from the Apache Git Service. To

[GitHub] zhanghang1989 commented on issue #7377: gluon.nn.Conv2D initialization int stride not handled correctly

2017-08-08 Thread git
zhanghang1989 commented on issue #7377: gluon.nn.Conv2D initialization int stride not handled correctly URL: https://github.com/apache/incubator-mxnet/issues/7377#issuecomment-321049677 Never mind, it is a typo. Should be ``strides=1`` instead of ``stride=1``

[GitHub] piiswrong commented on issue #7376: MXbox -- a simple and flexible vision toolbox for mxnet framework.

2017-08-08 Thread git
piiswrong commented on issue #7376: MXbox -- a simple and flexible vision toolbox for mxnet framework. URL: https://github.com/apache/incubator-mxnet/issues/7376#issuecomment-321037816 Thanks a lot for the contribution. This looks pretty similar to the recently introduced gluon

[GitHub] piiswrong commented on issue #7376: MXbox -- a simple and flexible vision toolbox for mxnet framework.

2017-08-08 Thread git
piiswrong commented on issue #7376: MXbox -- a simple and flexible vision toolbox for mxnet framework. URL: https://github.com/apache/incubator-mxnet/issues/7376#issuecomment-321037816 Thanks a lot for the contribution. This looks pretty similar to the recently introduced gluon

[GitHub] szha commented on issue #7382: GRU not working with layout 'NTC'

2017-08-08 Thread git
szha commented on issue #7382: GRU not working with layout 'NTC' URL: https://github.com/apache/incubator-mxnet/issues/7382#issuecomment-321041727 Should already be fixed by #7385 This is an automated message from the

[GitHub] LakeCarrot commented on issue #7341: Usage of Tensorboard in Distributed MXNet

2017-08-08 Thread git
LakeCarrot commented on issue #7341: Usage of Tensorboard in Distributed MXNet URL: https://github.com/apache/incubator-mxnet/issues/7341#issuecomment-321013823 @zihaolucky Good to know. Thanks a lot! This is an automated

[GitHub] LakeCarrot closed issue #7341: Usage of Tensorboard in Distributed MXNet

2017-08-08 Thread git
LakeCarrot closed issue #7341: Usage of Tensorboard in Distributed MXNet URL: https://github.com/apache/incubator-mxnet/issues/7341 This is an automated message from the Apache Git Service. To respond to the message, please

[GitHub] stefanhenneking commented on a change in pull request #7226: Extending the GPU dot operator

2017-08-08 Thread git
stefanhenneking commented on a change in pull request #7226: Extending the GPU dot operator URL: https://github.com/apache/incubator-mxnet/pull/7226#discussion_r131979993 ## File path: src/operator/tensor/dot-inl.cuh ## @@ -9,66 +9,163 @@ #include #include +#include

[GitHub] piiswrong commented on a change in pull request #7363: Add tensorboard configure into ./common/fit.py and ./train_mnist.py

2017-08-08 Thread git
piiswrong commented on a change in pull request #7363: Add tensorboard configure into ./common/fit.py and ./train_mnist.py URL: https://github.com/apache/incubator-mxnet/pull/7363#discussion_r131982184 ## File path: example/image-classification/common/fit.py ## @@ -168,10

[GitHub] piiswrong commented on a change in pull request #7363: Add tensorboard configure into ./common/fit.py and ./train_mnist.py

2017-08-08 Thread git
piiswrong commented on a change in pull request #7363: Add tensorboard configure into ./common/fit.py and ./train_mnist.py URL: https://github.com/apache/incubator-mxnet/pull/7363#discussion_r131981997 ## File path: example/image-classification/common/fit.py ## @@ -168,10

[GitHub] stefanhenneking commented on a change in pull request #7226: Extending the GPU dot operator

2017-08-08 Thread git
stefanhenneking commented on a change in pull request #7226: Extending the GPU dot operator URL: https://github.com/apache/incubator-mxnet/pull/7226#discussion_r131982060 ## File path: src/operator/tensor/dot-inl.cuh ## @@ -353,27 +640,308 @@ inline void

[GitHub] mli commented on a change in pull request #7379: Add license header

2017-08-08 Thread git
mli commented on a change in pull request #7379: Add license header URL: https://github.com/apache/incubator-mxnet/pull/7379#discussion_r131985346 ## File path: src/operator/softmax_activation.cu ## @@ -1,5 +1,23 @@ +/* + * Licensed to the Apache Software Foundation (ASF)

[GitHub] githubfragments commented on issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler?

2017-08-08 Thread git
githubfragments commented on issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler? URL: https://github.com/apache/incubator-mxnet/issues/7380#issuecomment-320909103 I have find where the bottleneck is. @edmBernard . I have

[GitHub] githubfragments closed issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler?

2017-08-08 Thread git
githubfragments closed issue #7380: strange phenomenon: 1080Ti is slower than maxwell gtx titan x. How to analysis the results of Profiler? URL: https://github.com/apache/incubator-mxnet/issues/7380 This is an automated

[GitHub] edmBernard commented on issue #6999: bgr or rgb input to vgg16?

2017-08-08 Thread git
edmBernard commented on issue #6999: bgr or rgb input to vgg16? URL: https://github.com/apache/incubator-mxnet/issues/6999#issuecomment-320911298 I think it's bgr as opencv convention. when i use my model I keep opencv color order

[GitHub] qingzhouzhen opened a new issue #7386: How could I define function for GPU only?

2017-08-08 Thread git
qingzhouzhen opened a new issue #7386: How could I define function for GPU only? URL: https://github.com/apache/incubator-mxnet/issues/7386 I need to define a function in .h file for GPU version only, then implenent it in .cu file, declaration in .h: ``` void

[GitHub] peschn opened a new pull request #7385: fix[#7382] fixes how the batch size is obtained in RNN forward pass.

2017-08-08 Thread git
peschn opened a new pull request #7385: fix[#7382] fixes how the batch size is obtained in RNN forward pass. URL: https://github.com/apache/incubator-mxnet/pull/7385 This is an automated message from the Apache Git

[GitHub] kimtang commented on issue #7273: [R] call im2rec in R

2017-08-08 Thread git
kimtang commented on issue #7273: [R] call im2rec in R URL: https://github.com/apache/incubator-mxnet/issues/7273#issuecomment-320959689 Yes, it should not be that difficult but im2rec depends on opencv and this is not available in R. Can someone please confirm? As far as I can see we

[GitHub] piiswrong closed pull request #7385: fix[#7382] fixes how the batch size is obtained in RNN forward pass.

2017-08-08 Thread git
piiswrong closed pull request #7385: fix[#7382] fixes how the batch size is obtained in RNN forward pass. URL: https://github.com/apache/incubator-mxnet/pull/7385 This is an automated message from the Apache Git Service.

[incubator-mxnet] branch master updated: fixes the incorrect retrieval of the batch size in the RNN forward pass. (#7385)

2017-08-08 Thread jxie
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 1a617fa fixes the incorrect retrieval

[GitHub] piiswrong commented on a change in pull request #7356: decouple record/train and add state readers

2017-08-08 Thread git
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_r131980722 ## File path: tests/python/unittest/test_autograd.py ## @@ -251,18 +251,49 @@ def

[incubator-mxnet] branch master updated: Docs for GAN (#7378)

2017-08-08 Thread jxie
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 d08abf5 Docs for GAN (#7378) d08abf5 is

[GitHub] piiswrong closed pull request #7378: Docs for GAN

2017-08-08 Thread git
piiswrong closed pull request #7378: Docs for GAN URL: https://github.com/apache/incubator-mxnet/pull/7378 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use

  1   2   >