svn commit: r26915 - /dev/incubator/mxnet/1.2.0.rc3/

Author: anirudh2290 Date: Tue May 15 05:56:05 2018 New Revision: 26915 Log: Remove mxnet-1.2.0.rc3 Removed: dev/incubator/mxnet/1.2.0.rc3/

[GitHub] eric-haibin-lin opened a new issue #10947: Flaky test: test_tutorials.test_onnx_inference_on_onnx_model

eric-haibin-lin opened a new issue #10947: Flaky test: test_tutorials.test_onnx_inference_on_onnx_model URL: https://github.com/apache/incubator-mxnet/issues/10947 http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-10913/12/pipeline ```

svn commit: r26914 - in /dev/incubator/mxnet/1.2.0.rc3: ./ apache-mxnet-src-1.2.0.rc3-incubating.tar.gz apache-mxnet-src-1.2.0.rc3-incubating.tar.gz.asc apache-mxnet-src-1.2.0.rc3-incubating.tar.gz.sh

Author: anirudh2290 Date: Tue May 15 05:52:04 2018 New Revision: 26914 Log: Add mxnet-1.2.0.rc3 Added: dev/incubator/mxnet/1.2.0.rc3/ dev/incubator/mxnet/1.2.0.rc3/apache-mxnet-src-1.2.0.rc3-incubating.tar.gz (with props)

svn commit: r26913 - /dev/incubator/mxnet/KEYS

Author: anirudh2290 Date: Tue May 15 05:43:12 2018 New Revision: 26913 Log: Add KEYS Modified: dev/incubator/mxnet/KEYS Modified: dev/incubator/mxnet/KEYS == --- dev/incubator/mxnet/KEYS (original) +++

[incubator-mxnet] branch master updated: fix symbolblock save_params (#10748)

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 0fb57ff fix symbolblock save_params

[GitHub] piiswrong closed pull request #10748: fix symbolblock save_params

piiswrong closed pull request #10748: fix symbolblock save_params URL: https://github.com/apache/incubator-mxnet/pull/10748 This is a PR merged from a forked repository. As GitHub hides the original diff on merge, it is displayed below for the sake of provenance: As this is a foreign

svn commit: r26912 - /dev/incubator/mxnet/1.2.0.rc3/

Author: anirudh2290 Date: Tue May 15 05:20:39 2018 New Revision: 26912 Log: Remove rc3 Removed: dev/incubator/mxnet/1.2.0.rc3/

[GitHub] juliusshufan commented on a change in pull request #10921: Test cases improvement for MKLDNN on Gluon

juliusshufan commented on a change in pull request #10921: Test cases improvement for MKLDNN on Gluon URL: https://github.com/apache/incubator-mxnet/pull/10921#discussion_r188166489 ## File path: tests/python/mkl/test_mkldnn.py ## @@ -95,122 +110,1147 @@ def

svn commit: r26910 - in /dev/incubator/mxnet/1.2.0.rc3: ./ apache-mxnet-src-1.2.0.rc3-incubating.tar.gz apache-mxnet-src-1.2.0.rc3-incubating.tar.gz.asc apache-mxnet-src-1.2.0.rc3-incubating.tar.gz.sh

Author: anirudh2290 Date: Tue May 15 04:43:18 2018 New Revision: 26910 Log: Add mxnet-1.2.0.rc3 Added: dev/incubator/mxnet/1.2.0.rc3/ dev/incubator/mxnet/1.2.0.rc3/apache-mxnet-src-1.2.0.rc3-incubating.tar.gz (with props)

[GitHub] chinakook commented on issue #10937: c++ API, fixed batch size for predictor.

chinakook commented on issue #10937: c++ API, fixed batch size for predictor. URL: https://github.com/apache/incubator-mxnet/issues/10937#issuecomment-389040792 MXPredReshape However, reshape may bring overhead. This is

[GitHub] chinakook commented on issue #10937: c++ API, fixed batch size for predictor.

chinakook commented on issue #10937: c++ API, fixed batch size for predictor. URL: https://github.com/apache/incubator-mxnet/issues/10937#issuecomment-389040792 MXPredReshape This is an automated message from the Apache Git

[GitHub] eric-haibin-lin opened a new pull request #10946: Remove kvstore calls from FM example

eric-haibin-lin opened a new pull request #10946: Remove kvstore calls from FM example URL: https://github.com/apache/incubator-mxnet/pull/10946 ## Description ## Update the example with module.prepare. Thanks @yifeim for reporting. ## Checklist ## ### Essentials ###

[GitHub] juliusshufan commented on a change in pull request #10921: Test cases improvement for MKLDNN on Gluon

juliusshufan commented on a change in pull request #10921: Test cases improvement for MKLDNN on Gluon URL: https://github.com/apache/incubator-mxnet/pull/10921#discussion_r188162283 ## File path: tests/python/mkl/test_mkldnn.py ## @@ -95,122 +110,1147 @@ def

[GitHub] juliusshufan commented on a change in pull request #10921: Test cases improvement for MKLDNN on Gluon

juliusshufan commented on a change in pull request #10921: Test cases improvement for MKLDNN on Gluon URL: https://github.com/apache/incubator-mxnet/pull/10921#discussion_r188160329 ## File path: tests/python/mkl/test_mkldnn.py ## @@ -29,6 +29,21 @@

[GitHub] juliusshufan commented on a change in pull request #10921: Test cases improvement for MKLDNN on Gluon

juliusshufan commented on a change in pull request #10921: Test cases improvement for MKLDNN on Gluon URL: https://github.com/apache/incubator-mxnet/pull/10921#discussion_r188159032 ## File path: tests/python/mkl/test_mkldnn.py ## @@ -95,122 +110,1147 @@ def

[GitHub] haojin2 opened a new issue #10945: Flaky test:

haojin2 opened a new issue #10945: Flaky test: URL: https://github.com/apache/incubator-mxnet/issues/10945 http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-10939/4/pipeline ```

[GitHub] haojin2 commented on issue #10920: Flaky test test_operator_gpu.test_sparse_dot

haojin2 commented on issue #10920: Flaky test test_operator_gpu.test_sparse_dot URL: https://github.com/apache/incubator-mxnet/issues/10920#issuecomment-389031418 Met this again: http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-10931/7/pipeline

[GitHub] jonbakerfish commented on issue #10935: When are the images converted to range(0, 1) in pre-trained resnet50_v2 on imagenet?

jonbakerfish commented on issue #10935: When are the images converted to range(0,1) in pre-trained resnet50_v2 on imagenet? URL: https://github.com/apache/incubator-mxnet/issues/10935#issuecomment-389029276 Hi @chinakook , Thanks for pointing it out. Also, I found that the

[GitHub] jonbakerfish closed issue #10935: When are the images converted to range(0, 1) in pre-trained resnet50_v2 on imagenet?

jonbakerfish closed issue #10935: When are the images converted to range(0,1) in pre-trained resnet50_v2 on imagenet? URL: https://github.com/apache/incubator-mxnet/issues/10935 This is an automated message from the Apache

[GitHub] jonbakerfish commented on issue #10935: When are the images converted to range(0, 1) in pre-trained resnet50_v2 on imagenet?

jonbakerfish commented on issue #10935: When are the images converted to range(0,1) in pre-trained resnet50_v2 on imagenet? URL: https://github.com/apache/incubator-mxnet/issues/10935#issuecomment-389029276 Hi @chinakook , Thanks for pointing out. Also, I found that the

[GitHub] liuzx32 commented on issue #10936: aws-ec2 make -j4 and throw the error

liuzx32 commented on issue #10936: aws-ec2 make -j4 and throw the error URL: https://github.com/apache/incubator-mxnet/issues/10936#issuecomment-389027700 aws ec2: uname -a # Linux ip-172-31-8-144 3.14.27-25.47.amzn1.x86_64 #1 SMP Wed Dec 17 18:36:15 UTC 2014 x86_64 x86_64 x86_64

[GitHub] vdantu commented on issue #10897: performance and error issue with INT8 quantization example

vdantu commented on issue #10897: performance and error issue with INT8 quantization example URL: https://github.com/apache/incubator-mxnet/issues/10897#issuecomment-389027680 @sandeep-krishnamurthy : Could you please tag this as "Flaky", "Python", "Example", "Question".

[GitHub] vdantu commented on issue #10910: Feature Request: loading gluon trainer states with different contexts from saved states

vdantu commented on issue #10910: Feature Request: loading gluon trainer states with different contexts from saved states URL: https://github.com/apache/incubator-mxnet/issues/10910#issuecomment-389027113 @sandeep-krishnamurthy : Could you please tag this issue as a "Gluon" and "feature

[GitHub] vdantu commented on issue #10884: mx.model.load_checkpoint error?

vdantu commented on issue #10884: mx.model.load_checkpoint error? URL: https://github.com/apache/incubator-mxnet/issues/10884#issuecomment-389026272 @lovehuanhuan : It's an interesting question. Could you also post this on discuss.mxnet.io? You would receive a wider audience. Also

[GitHub] vdantu commented on issue #10904: switch_bucket() sets grad_req to 'write' by default and overwrites other settings

vdantu commented on issue #10904: switch_bucket() sets grad_req to 'write' by default and overwrites other settings URL: https://github.com/apache/incubator-mxnet/issues/10904#issuecomment-389025477 @sandeep-krishnamurthy : Could you please tag this "Question", "Python" @ArneNx :

[GitHub] vdantu commented on issue #10907: #10820 NaiveEngine regression

vdantu commented on issue #10907: #10820 NaiveEngine regression URL: https://github.com/apache/incubator-mxnet/issues/10907#issuecomment-389024968 @sandeep-krishnamurthy : Could you please tag this as "performance" This is

[GitHub] vdantu commented on issue #10925: problem installing mxnet in R 3.5

vdantu commented on issue #10925: problem installing mxnet in R 3.5 URL: https://github.com/apache/incubator-mxnet/issues/10925#issuecomment-389024657 @sandeep-krishnamurthy : Could you tag this question with "R", "Installation", "Question"

[GitHub] vdantu commented on issue #10928: Optimizers memory usage

vdantu commented on issue #10928: Optimizers memory usage URL: https://github.com/apache/incubator-mxnet/issues/10928#issuecomment-389024353 @sandeep-krishnamurthy : Could you please tag this as "question", "R" @jeremiedb : You could also post this on "discuss.mxnet.io". You would

[GitHub] vdantu commented on issue #10932: about the data translating tool im2rec, is there any option to add new images to the existing rec file?

vdantu commented on issue #10932: about the data translating tool im2rec, is there any option to add new images to the existing rec file? URL: https://github.com/apache/incubator-mxnet/issues/10932#issuecomment-389023980 @GoodJoey : Could you also post this on discuss.mxnet.io? You will

[GitHub] liuzx32 commented on issue #10936: aws-ec2 make -j4 and throw the error

liuzx32 commented on issue #10936: aws-ec2 make -j4 and throw the error URL: https://github.com/apache/incubator-mxnet/issues/10936#issuecomment-389023566 @vdantu thank you! 1. git clone https://github.com/apache/incubator-mxnet 2. git submodule update --init --recursive 3. cd

[GitHub] vdantu commented on issue #10936: aws-ec2 make -j4 and throw the error

vdantu commented on issue #10936: aws-ec2 make -j4 and throw the error URL: https://github.com/apache/incubator-mxnet/issues/10936#issuecomment-388875612 @liuzx32 : Could you provide minimum reproducible steps for this failure. @sandeep-krishnamurthy : Could you tag this "build",

[GitHub] liuzx32 commented on issue #10936: aws-ec2 make -j4 and throw the error

liuzx32 commented on issue #10936: aws-ec2 make -j4 and throw the error URL: https://github.com/apache/incubator-mxnet/issues/10936#issuecomment-389023566 @vdantu thank you! 1. git clone https://github.com/apache/incubator-mxnet 2. git submodule update --init --recursive 3. make

[GitHub] vdantu commented on issue #10937: c++ API, fixed batch size for predictor.

vdantu commented on issue #10937: c++ API, fixed batch size for predictor. URL: https://github.com/apache/incubator-mxnet/issues/10937#issuecomment-389023541 @ashkanaev : You could post this question on discuss.mxnet.io to get a wider audience. Also, are you seeing errors when you

[GitHub] feevos commented on issue #10944: Correction of gluon documentation in nn.md

feevos commented on issue #10944: Correction of gluon documentation in nn.md URL: https://github.com/apache/incubator-mxnet/pull/10944#issuecomment-389020749 Added Lambda too, sorry missed that! This is an automated message

[GitHub] Godricly opened a new pull request #10915: [MXNET-9704] An assertion check for invalid layout

Godricly opened a new pull request #10915: [MXNET-9704] An assertion check for invalid layout URL: https://github.com/apache/incubator-mxnet/pull/10915 ## Description ## a fix for issue #9704 ## Checklist ## ### Essentials ### Please feel free to remove inapplicable items for

[GitHub] Godricly commented on issue #10915: [MXNET-9704] An assertion check for invalid layout

Godricly commented on issue #10915: [MXNET-9704] An assertion check for invalid layout URL: https://github.com/apache/incubator-mxnet/pull/10915#issuecomment-389020691 @zhreshold The conv layers have a second round layout check, so I added the assertion according to that check. However,

[GitHub] Godricly closed pull request #10915: [MXNET-9704] An assertion check for invalid layout

Godricly closed pull request #10915: [MXNET-9704] An assertion check for invalid layout URL: https://github.com/apache/incubator-mxnet/pull/10915 This is a PR merged from a forked repository. As GitHub hides the original diff on merge, it is displayed below for the sake of provenance:

[GitHub] szha commented on a change in pull request #10944: Correction of gluon documentation in nn.md

szha commented on a change in pull request #10944: Correction of gluon documentation in nn.md URL: https://github.com/apache/incubator-mxnet/pull/10944#discussion_r188148161 ## File path: docs/api/python/gluon/nn.md ## @@ -23,6 +23,7 @@ This document lists the neural

[incubator-mxnet-site] branch asf-site updated: Bump the publish timestamp.

This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a commit to branch asf-site in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git The following commit(s) were added to refs/heads/asf-site by this push: new 673faa9 Bump the publish

[GitHub] eric-haibin-lin commented on issue #10095: BatchNorm on axis=-1 is very slower than axis=1

eric-haibin-lin commented on issue #10095: BatchNorm on axis=-1 is very slower than axis=1 URL: https://github.com/apache/incubator-mxnet/issues/10095#issuecomment-389017721 Yeah TF doesn’t use CUDNN. We’re waiting for CUDNN 7.2 which has lots of improvements on NHWC

[GitHub] feevos commented on issue #10930: gluon documentation bug

feevos commented on issue #10930: gluon documentation bug URL: https://github.com/apache/incubator-mxnet/issues/10930#issuecomment-389017529 Thanks @szha I am closing this and I hope I haven't created more trouble than help with my ignorance in contributing (getting there ... :) ).

[GitHub] feevos closed issue #10930: gluon documentation bug

feevos closed issue #10930: gluon documentation bug URL: https://github.com/apache/incubator-mxnet/issues/10930 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and

[GitHub] feevos opened a new pull request #10944: Correction of gluon documentation in nn.md

feevos opened a new pull request #10944: Correction of gluon documentation in nn.md URL: https://github.com/apache/incubator-mxnet/pull/10944 ## Description ## Added ```HybridLambda``` in the Basic Layers list documentation as suggested in #10930. By the way this is my

[GitHub] zhreshold opened a new pull request #10943: fix doc

zhreshold opened a new pull request #10943: fix doc URL: https://github.com/apache/incubator-mxnet/pull/10943 ## Description ## - Fix a missing link for mx.image.SequentialAug - add a header for how-to create record file for object detection so we can share ref links to this section.

[GitHub] eric-haibin-lin commented on a change in pull request #10924: [Sparse-Gluon] [WIP] embedding with sparse grad

eric-haibin-lin commented on a change in pull request #10924: [Sparse-Gluon] [WIP] embedding with sparse grad URL: https://github.com/apache/incubator-mxnet/pull/10924#discussion_r188142080 ## File path: python/mxnet/gluon/trainer.py ## @@ -110,16 +110,29 @@ def

[GitHub] szha commented on a change in pull request #10924: [Sparse-Gluon] [WIP] embedding with sparse grad

szha commented on a change in pull request #10924: [Sparse-Gluon] [WIP] embedding with sparse grad URL: https://github.com/apache/incubator-mxnet/pull/10924#discussion_r188142152 ## File path: python/mxnet/gluon/trainer.py ## @@ -110,16 +110,29 @@ def

[GitHub] eric-haibin-lin commented on a change in pull request #10924: [Sparse-Gluon] [WIP] embedding with sparse grad

eric-haibin-lin commented on a change in pull request #10924: [Sparse-Gluon] [WIP] embedding with sparse grad URL: https://github.com/apache/incubator-mxnet/pull/10924#discussion_r188142020 ## File path: python/mxnet/gluon/trainer.py ## @@ -110,16 +110,29 @@ def

[GitHub] szha commented on a change in pull request #10924: [Sparse-Gluon] [WIP] embedding with sparse grad

szha commented on a change in pull request #10924: [Sparse-Gluon] [WIP] embedding with sparse grad URL: https://github.com/apache/incubator-mxnet/pull/10924#discussion_r188141280 ## File path: python/mxnet/gluon/trainer.py ## @@ -110,16 +110,29 @@ def

[GitHub] chinakook commented on issue #10935: When are the images converted to range(0, 1) in pre-trained resnet50_v2 on imagenet?

chinakook commented on issue #10935: When are the images converted to range(0,1) in pre-trained resnet50_v2 on imagenet? URL: https://github.com/apache/incubator-mxnet/issues/10935#issuecomment-389010262 ```image = mx.nd.image.to_tensor(image)``` will do that.

[GitHub] eric-haibin-lin commented on a change in pull request #10939: [MXNET-420] broadcast_mul/div between csr and 1D dense on GPU

eric-haibin-lin commented on a change in pull request #10939: [MXNET-420] broadcast_mul/div between csr and 1D dense on GPU URL: https://github.com/apache/incubator-mxnet/pull/10939#discussion_r188141258 ## File path: src/operator/tensor/elemwise_binary_broadcast_op_basic.cu

[GitHub] szha commented on a change in pull request #10924: [Sparse-Gluon] [WIP] embedding with sparse grad

szha commented on a change in pull request #10924: [Sparse-Gluon] [WIP] embedding with sparse grad URL: https://github.com/apache/incubator-mxnet/pull/10924#discussion_r188141166 ## File path: python/mxnet/gluon/trainer.py ## @@ -110,16 +110,29 @@ def

[GitHub] chinakook commented on issue #10095: BatchNorm on axis=-1 is very slower than axis=1

chinakook commented on issue #10095: BatchNorm on axis=-1 is very slower than axis=1 URL: https://github.com/apache/incubator-mxnet/issues/10095#issuecomment-389009456 I got it. As I've tested, Tensorflow's batchnorm on the last axis is OK. It may not use cudnn.

[GitHub] eric-haibin-lin commented on issue #3946: When predicting, does mxnet provide thread-safe interface?

eric-haibin-lin commented on issue #3946: When predicting, does mxnet provide thread-safe interface? URL: https://github.com/apache/incubator-mxnet/issues/3946#issuecomment-389008247 Confirmed with @piiswrong offline that the dependency engine in C++ is actually thread safe.

[GitHub] eric-haibin-lin commented on issue #3946: When predicting, does mxnet provide thread-safe interface?

eric-haibin-lin commented on issue #3946: When predicting, does mxnet provide thread-safe interface? URL: https://github.com/apache/incubator-mxnet/issues/3946#issuecomment-389008247 Confirmed with @piiswrong offline that the dependency engine in C++ is thread safe.

[incubator-mxnet] branch v1.2.0 updated: add apache-release as parent to pom file (#10941)

This is an automated email from the ASF dual-hosted git repository. liuyizhi pushed a commit to branch v1.2.0 in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git The following commit(s) were added to refs/heads/v1.2.0 by this push: new 297c64f add apache-release as

[GitHub] yzhliu closed pull request #10941: add apache-release as parent to pom file

yzhliu closed pull request #10941: add apache-release as parent to pom file URL: https://github.com/apache/incubator-mxnet/pull/10941 This is a PR merged from a forked repository. As GitHub hides the original diff on merge, it is displayed below for the sake of provenance: As this is a

[GitHub] reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator

reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator URL: https://github.com/apache/incubator-mxnet/pull/10931#discussion_r188130642 ## File path: python/mxnet/ndarray/ndarray.py ## @@ -3740,3 +3740,32 @@ def empty(shape, ctx=None,

[GitHub] piiswrong commented on issue #10927: Feature Request: LSTM/GRU hard_sigmoid support

piiswrong commented on issue #10927: Feature Request: LSTM/GRU hard_sigmoid support URL: https://github.com/apache/incubator-mxnet/issues/10927#issuecomment-388996251 @szha This is an automated message from the Apache Git

[GitHub] piiswrong commented on issue #10919: gluon.Block load_params method does not accept python3 pathlib.Path

piiswrong commented on issue #10919: gluon.Block load_params method does not accept python3 pathlib.Path URL: https://github.com/apache/incubator-mxnet/issues/10919#issuecomment-388995506 I think we can remove the check. Python prefers duck typing anyway

[GitHub] eric-haibin-lin commented on issue #10095: BatchNorm on axis=-1 is very slower than axis=1

eric-haibin-lin commented on issue #10095: BatchNorm on axis=-1 is very slower than axis=1 URL: https://github.com/apache/incubator-mxnet/issues/10095#issuecomment-388993624 Yes, CUDNN has very different performance characteristics on these two formats. On 2018年5月13日, at 17:41,

[GitHub] haojin2 commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator

haojin2 commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator URL: https://github.com/apache/incubator-mxnet/pull/10931#discussion_r188126557 ## File path: python/mxnet/ndarray/ndarray.py ## @@ -3740,3 +3740,32 @@ def empty(shape, ctx=None,

[GitHub] haojin2 commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator

haojin2 commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator URL: https://github.com/apache/incubator-mxnet/pull/10931#discussion_r188126415 ## File path: python/mxnet/ndarray/ndarray.py ## @@ -3740,3 +3740,32 @@ def empty(shape, ctx=None,

[GitHub] haojin2 opened a new issue #10942: Flaky test test_operator_gpu.test_batchnorm_with_type

haojin2 opened a new issue #10942: Flaky test test_operator_gpu.test_batchnorm_with_type URL: https://github.com/apache/incubator-mxnet/issues/10942 http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-10939/2/pipeline ```

[GitHub] harryprince commented on issue #10890: [R] Is it possible to train Mxnet symbols and a GLM pretrained model as loss?

harryprince commented on issue #10890: [R] Is it possible to train Mxnet symbols and a GLM pretrained model as loss? URL: https://github.com/apache/incubator-mxnet/issues/10890#issuecomment-388989336 @some-guy1 it seems this is the answer:

[GitHub] eric-haibin-lin commented on a change in pull request #10913: Fix _backward_norm op registration

eric-haibin-lin commented on a change in pull request #10913: Fix _backward_norm op registration URL: https://github.com/apache/incubator-mxnet/pull/10913#discussion_r188121985 ## File path: src/operator/tensor/broadcast_reduce_op.h ## @@ -880,17 +878,30 @@ inline bool

[GitHub] eric-haibin-lin commented on a change in pull request #10913: Fix _backward_norm op registration

eric-haibin-lin commented on a change in pull request #10913: Fix _backward_norm op registration URL: https://github.com/apache/incubator-mxnet/pull/10913#discussion_r188121574 ## File path: src/operator/tensor/broadcast_reduce_op.h ## @@ -880,17 +878,30 @@ inline bool

[GitHub] reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator

reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator URL: https://github.com/apache/incubator-mxnet/pull/10931#discussion_r188116180 ## File path: python/mxnet/ndarray/ndarray.py ## @@ -3740,3 +3740,32 @@ def empty(shape, ctx=None,

[GitHub] reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator

reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator URL: https://github.com/apache/incubator-mxnet/pull/10931#discussion_r188119508 ## File path: src/operator/tensor/histogram-inl.h ## @@ -0,0 +1,178 @@ +/* + * Licensed to the

[GitHub] reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator

reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator URL: https://github.com/apache/incubator-mxnet/pull/10931#discussion_r188117569 ## File path: tests/python/unittest/test_operator.py ## @@ -5919,6 +5919,31 @@ def f(x, a, b, c):

[GitHub] reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator

reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator URL: https://github.com/apache/incubator-mxnet/pull/10931#discussion_r188118579 ## File path: python/mxnet/ndarray/ndarray.py ## @@ -3740,3 +3740,29 @@ def empty(shape, ctx=None,

[GitHub] reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator

reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator URL: https://github.com/apache/incubator-mxnet/pull/10931#discussion_r188119379 ## File path: src/operator/tensor/histogram-inl.h ## @@ -0,0 +1,178 @@ +/* + * Licensed to the

[GitHub] reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator

reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator URL: https://github.com/apache/incubator-mxnet/pull/10931#discussion_r188116305 ## File path: python/mxnet/ndarray/ndarray.py ## @@ -3740,3 +3740,32 @@ def empty(shape, ctx=None,

[GitHub] reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator

reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator URL: https://github.com/apache/incubator-mxnet/pull/10931#discussion_r188116437 ## File path: python/mxnet/ndarray/ndarray.py ## @@ -3740,3 +3740,32 @@ def empty(shape, ctx=None,

[GitHub] reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator

reminisce commented on a change in pull request #10931: [MXNET-349] [WIP] Histogram Operator URL: https://github.com/apache/incubator-mxnet/pull/10931#discussion_r188117066 ## File path: tests/python/unittest/test_operator.py ## @@ -5919,6 +5919,31 @@ def f(x, a, b, c):

[GitHub] zheng-da commented on a change in pull request #10913: Fix _backward_norm op registration

zheng-da commented on a change in pull request #10913: Fix _backward_norm op registration URL: https://github.com/apache/incubator-mxnet/pull/10913#discussion_r188119526 ## File path: src/operator/tensor/broadcast_reduce_op.h ## @@ -880,17 +878,30 @@ inline bool

[GitHub] ThomasDelteil commented on issue #10846: flaky tests in tutorial CI

ThomasDelteil commented on issue #10846: flaky tests in tutorial CI URL: https://github.com/apache/incubator-mxnet/issues/10846#issuecomment-388984890 Thanks @anirudh2290 , I have been distracted by other projects, will try to prioritize moving to smaller models this week.

[GitHub] szha commented on issue #10940: Fused LSTM Cell

szha commented on issue #10940: Fused LSTM Cell URL: https://github.com/apache/incubator-mxnet/pull/10940#issuecomment-388979845 @piiswrong Given that RNN operator is actively being changed, I'd prefer to make the backend changes once all CPU kernel changes are merged if that's OK.

[GitHub] reminisce commented on a change in pull request #10940: Fused LSTM Cell

reminisce commented on a change in pull request #10940: Fused LSTM Cell URL: https://github.com/apache/incubator-mxnet/pull/10940#discussion_r188113623 ## File path: src/operator/tensor/matrix_op-inl.h ## @@ -152,18 +152,31 @@ inline TShape InferReshapeShape(const

[GitHub] reminisce commented on a change in pull request #10940: Fused LSTM Cell

reminisce commented on a change in pull request #10940: Fused LSTM Cell URL: https://github.com/apache/incubator-mxnet/pull/10940#discussion_r188114250 ## File path: src/operator/tensor/matrix_op-inl.h ## @@ -152,18 +152,31 @@ inline TShape InferReshapeShape(const

[GitHub] reminisce commented on a change in pull request #10940: Fused LSTM Cell

reminisce commented on a change in pull request #10940: Fused LSTM Cell URL: https://github.com/apache/incubator-mxnet/pull/10940#discussion_r188113913 ## File path: src/operator/tensor/matrix_op-inl.h ## @@ -189,7 +202,7 @@ inline bool ReshapeShape(const nnvm::NodeAttrs&

[GitHub] reminisce commented on a change in pull request #10940: Fused LSTM Cell

reminisce commented on a change in pull request #10940: Fused LSTM Cell URL: https://github.com/apache/incubator-mxnet/pull/10940#discussion_r188113433 ## File path: src/operator/nn/concat.cc ## @@ -39,14 +39,16 @@ static bool ConcatShape(const nnvm::NodeAttrs& attrs,

[GitHub] piiswrong commented on a change in pull request #10940: Fused LSTM Cell

piiswrong commented on a change in pull request #10940: Fused LSTM Cell URL: https://github.com/apache/incubator-mxnet/pull/10940#discussion_r188113083 ## File path: python/mxnet/gluon/rnn/rnn_cell.py ## @@ -297,6 +297,63 @@ def __init__(self, prefix=None, params=None):

[GitHub] piiswrong commented on a change in pull request #10940: Fused LSTM Cell

piiswrong commented on a change in pull request #10940: Fused LSTM Cell URL: https://github.com/apache/incubator-mxnet/pull/10940#discussion_r188112698 ## File path: python/mxnet/gluon/rnn/rnn_cell.py ## @@ -297,6 +297,63 @@ def __init__(self, prefix=None, params=None):

[GitHub] nswamy opened a new pull request #10941: add apache-release as parent to pom file

nswamy opened a new pull request #10941: add apache-release as parent to pom file URL: https://github.com/apache/incubator-mxnet/pull/10941 ## Description ## (Brief description on what this PR is about) ## Checklist ## ### Essentials ### Please feel free to remove

[GitHub] sandeep-krishnamurthy closed issue #9949: Mac build from source script validation fails

sandeep-krishnamurthy closed issue #9949: Mac build from source script validation fails URL: https://github.com/apache/incubator-mxnet/issues/9949 This is an automated message from the Apache Git Service. To respond to the

[GitHub] sandeep-krishnamurthy closed issue #8155: Error when compiling MXNet from source with VS2013 / Windows 10

sandeep-krishnamurthy closed issue #8155: Error when compiling MXNet from source with VS2013 / Windows 10 URL: https://github.com/apache/incubator-mxnet/issues/8155 This is an automated message from the Apache Git Service.

[GitHub] sandeep-krishnamurthy closed issue #8016: mxnet installation GPU validation issue/error

sandeep-krishnamurthy closed issue #8016: mxnet installation GPU validation issue/error URL: https://github.com/apache/incubator-mxnet/issues/8016 This is an automated message from the Apache Git Service. To respond to the

[GitHub] szha opened a new pull request #10940: Fused LSTM Cell

szha opened a new pull request #10940: Fused LSTM Cell URL: https://github.com/apache/incubator-mxnet/pull/10940 ## Description ## Use fused operator for Gluon LSTM cell. ## Checklist ## ### Essentials ### Please feel free to remove inapplicable items for your PR. - [x]

[GitHub] haojin2 opened a new pull request #10939: [MXNET-420] broadcast_mul/div between csr and 1D dense on GPU

haojin2 opened a new pull request #10939: [MXNET-420] broadcast_mul/div between csr and 1D dense on GPU URL: https://github.com/apache/incubator-mxnet/pull/10939 ## Description ## As title ## Checklist ## ### Essentials ### - [x] Changes are complete (i.e. I finished coding

[GitHub] conradwt commented on issue #10185: [Documentation][Installation][macOS] Building From Source Has Incomplete Instructions For macOS 10.13.3

conradwt commented on issue #10185: [Documentation][Installation][macOS] Building From Source Has Incomplete Instructions For macOS 10.13.3 URL: https://github.com/apache/incubator-mxnet/issues/10185#issuecomment-388966282 @bhavinthaker The library which needs the libmxnet.dylib is the

[GitHub] reminisce commented on a change in pull request #10882: move exec.reshape to backend

reminisce commented on a change in pull request #10882: move exec.reshape to backend URL: https://github.com/apache/incubator-mxnet/pull/10882#discussion_r188031230 ## File path: src/executor/graph_executor.cc ## @@ -1043,6 +1043,108 @@ void

[GitHub] reminisce commented on a change in pull request #10882: move exec.reshape to backend

reminisce commented on a change in pull request #10882: move exec.reshape to backend URL: https://github.com/apache/incubator-mxnet/pull/10882#discussion_r188031132 ## File path: src/executor/graph_executor.cc ## @@ -1043,6 +1043,108 @@ void

[GitHub] reminisce commented on a change in pull request #10882: move exec.reshape to backend

reminisce commented on a change in pull request #10882: move exec.reshape to backend URL: https://github.com/apache/incubator-mxnet/pull/10882#discussion_r188030832 ## File path: src/executor/graph_executor.cc ## @@ -1043,6 +1043,108 @@ void

[GitHub] reminisce commented on a change in pull request #10882: move exec.reshape to backend

reminisce commented on a change in pull request #10882: move exec.reshape to backend URL: https://github.com/apache/incubator-mxnet/pull/10882#discussion_r188032054 ## File path: src/c_api/c_api_executor.cc ## @@ -510,6 +510,93 @@ int MXExecutorSimpleBind(SymbolHandle

[GitHub] reminisce commented on a change in pull request #10882: move exec.reshape to backend

reminisce commented on a change in pull request #10882: move exec.reshape to backend URL: https://github.com/apache/incubator-mxnet/pull/10882#discussion_r188032349 ## File path: src/c_api/c_api_executor.cc ## @@ -510,6 +510,93 @@ int MXExecutorSimpleBind(SymbolHandle

[GitHub] anirudh2290 commented on issue #10846: flaky tests in tutorial CI

anirudh2290 commented on issue #10846: flaky tests in tutorial CI URL: https://github.com/apache/incubator-mxnet/issues/10846#issuecomment-388962817 Failed again here: http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-10938/1/pipeline/

[GitHub] bhavinthaker commented on issue #8016: mxnet installation GPU validation issue/error

bhavinthaker commented on issue #8016: mxnet installation GPU validation issue/error URL: https://github.com/apache/incubator-mxnet/issues/8016#issuecomment-388962407 @sandeep-krishnamurthy @eric-haibin-lin Could one of you please close this issue?

[GitHub] bhavinthaker commented on issue #8155: Error when compiling MXNet from source with VS2013 / Windows 10

bhavinthaker commented on issue #8155: Error when compiling MXNet from source with VS2013 / Windows 10 URL: https://github.com/apache/incubator-mxnet/issues/8155#issuecomment-388961823 @sandeep-krishnamurthy @eric-haibin-lin Could one of you please close this issue?

[GitHub] bhavinthaker commented on issue #9949: Mac build from source script validation fails

bhavinthaker commented on issue #9949: Mac build from source script validation fails URL: https://github.com/apache/incubator-mxnet/issues/9949#issuecomment-388961783 @sandeep-krishnamurthy @eric-haibin-lin Could one of you please close this issue?

[GitHub] eric-haibin-lin commented on a change in pull request #10913: Fix _backward_norm op registration

eric-haibin-lin commented on a change in pull request #10913: Fix _backward_norm op registration URL: https://github.com/apache/incubator-mxnet/pull/10913#discussion_r188091468 ## File path: src/operator/tensor/broadcast_reduce_op.h ## @@ -880,17 +878,30 @@ inline bool

  1   2   >