[incubator-mxnet] branch v0.11.0 updated: Remove finalizers from Scala API (#8887) (#9568)

This is an automated email from the ASF dual-hosted git repository. liuyizhi pushed a commit to branch v0.11.0 in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git The following commit(s) were added to refs/heads/v0.11.0 by this push: new f8f0b8a Remove finalizers from

[GitHub] yzhliu closed pull request #9568: Cherry-picked from master branch(Remove finalizers from Scala API (#8887))

yzhliu closed pull request #9568: Cherry-picked from master branch(Remove finalizers from Scala API (#8887)) URL: https://github.com/apache/incubator-mxnet/pull/9568 This is a PR merged from a forked repository. As GitHub hides the original diff on merge, it is displayed below for the

[GitHub] zackchase opened a new issue #9571: ACCURACY IS USING NUMPY, URGENT FIX

zackchase opened a new issue #9571: ACCURACY IS USING NUMPY, URGENT FIX URL: https://github.com/apache/incubator-mxnet/issues/9571 Accuracy is the bottleneck for even a 4000 node neural network. This is a practical job that many people will run and which a GPU ***CAN accelerate***.

[GitHub] anirudh2290 opened a new pull request #9570: Add checksum for file download in test_consistency

anirudh2290 opened a new pull request #9570: Add checksum for file download in test_consistency URL: https://github.com/apache/incubator-mxnet/pull/9570 ## Description ## Adding checksum test to check if file is downloaded fully otherwise raise exception @eric-haibin-lin ##

[GitHub] cjolivier01 commented on issue #9545: Profiling discussion

cjolivier01 commented on issue #9545: Profiling discussion URL: https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360684117 so far i?ve tried to make the profiling have as close to zero impact as possible on the runtime performance. i?ll need to think about how to get

[GitHub] pengzhao-intel commented on issue #9545: Profiling discussion

pengzhao-intel commented on issue #9545: Profiling discussion URL: https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360680765 @cjolivier01 yes, looks good :) One minor suggestion is to show the different instance of the Class/OP because their performance will be very

[GitHub] pengzhao-intel commented on issue #9545: Profiling discussion

pengzhao-intel commented on issue #9545: Profiling discussion URL: https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360680765 @cjolivier01 yes, looks good :) One minor suggestion is to show the different instance of the Class/OP because their performance will be very

[GitHub] cjolivier01 commented on issue #9545: Profiling discussion

cjolivier01 commented on issue #9545: Profiling discussion URL: https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360680487 Question: _"Operator tuning: I saw you have contributed to the operator tuning of mxnet. Seems it will create many omp threads before

[GitHub] cjolivier01 commented on issue #9545: Profiling discussion

cjolivier01 commented on issue #9545: Profiling discussion URL: https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360680487 Question: _"Operator tuning: I saw you have contributed to the operator tuning of mxnet. Seems it will create many omp threads before

[GitHub] cjolivier01 commented on issue #9545: Profiling discussion

cjolivier01 commented on issue #9545: Profiling discussion URL: https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360680487 Question: Operator tuning: I saw you have contributed to the operator tuning of mxnet. Seems it will create many omp threads before executing

[GitHub] pengzhao-intel commented on issue #8302: Refactor operators & MKLDNN

pengzhao-intel commented on issue #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#issuecomment-360659821 @eric-haibin-lin @marcoabreu @cjolivier01 @szha @reminisce @zheng-da Do you have the time schedule of the merging for this PR? We

[GitHub] Soonhwan-Kwon opened a new pull request #9569: fix CreateOp of altering shape operations to accept dtypes of uint8 and int32

Soonhwan-Kwon opened a new pull request #9569: fix CreateOp of altering shape operations to accept dtypes of uint8 and int32 URL: https://github.com/apache/incubator-mxnet/pull/9569 ## Description ## Same context as https://github.com/apache/incubator-mxnet/pull/9561, to deal with

[GitHub] nswamy opened a new pull request #9568: Cherry-picked from master branch(Remove finalizers from Scala API (#8887))

nswamy opened a new pull request #9568: Cherry-picked from master branch(Remove finalizers from Scala API (#8887)) URL: https://github.com/apache/incubator-mxnet/pull/9568 * [scala] Remove finalizers for leakable resources Finalizers always run in a separate thread which is not

[GitHub] ZiyueHuang commented on a change in pull request #9540: refactor regression ops to nnvm interface

ZiyueHuang commented on a change in pull request #9540: refactor regression ops to nnvm interface URL: https://github.com/apache/incubator-mxnet/pull/9540#discussion_r164027751 ## File path: src/operator/regression_output.cc ## @@ -18,86 +18,96 @@ */ /*! - *

[GitHub] szha commented on issue #9514: Language Modeling Datasets and Sampler

szha commented on issue #9514: Language Modeling Datasets and Sampler URL: https://github.com/apache/incubator-mxnet/pull/9514#issuecomment-360668139 To address the concern of merging datasets based on frequencies, I made the frequencies (word-counts) a property of the dataset too. This

[GitHub] szha commented on issue #9514: Language Modeling Datasets and Sampler

szha commented on issue #9514: Language Modeling Datasets and Sampler URL: https://github.com/apache/incubator-mxnet/pull/9514#issuecomment-360668139 To address the concern of merging datasets based on frequencies, I made the frequencies counts a property of the dataset too. This way, user

[GitHub] anirudh2290 commented on a change in pull request #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model Quantization with Calibration

anirudh2290 commented on a change in pull request #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model Quantization with Calibration URL: https://github.com/apache/incubator-mxnet/pull/9552#discussion_r164022315 ## File path: python/mxnet/quantization.py ## @@ -0,0 +1,467 @@

[GitHub] anirudh2290 commented on a change in pull request #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model Quantization with Calibration

anirudh2290 commented on a change in pull request #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model Quantization with Calibration URL: https://github.com/apache/incubator-mxnet/pull/9552#discussion_r164022315 ## File path: python/mxnet/quantization.py ## @@ -0,0 +1,467 @@

[GitHub] mbaijal commented on a change in pull request #9554: [Review Required] Adding Apache License header to the perl-package

mbaijal commented on a change in pull request #9554: [Review Required] Adding Apache License header to the perl-package URL: https://github.com/apache/incubator-mxnet/pull/9554#discussion_r164020942 ## File path: tools/license_header.py ## @@ -83,7 +83,7 @@ _LANGS =

[GitHub] eric-haibin-lin closed pull request #9554: [Review Required] Adding Apache License header to the perl-package

eric-haibin-lin closed pull request #9554: [Review Required] Adding Apache License header to the perl-package URL: https://github.com/apache/incubator-mxnet/pull/9554 This is a PR merged from a forked repository. As GitHub hides the original diff on merge, it is displayed below for the

[incubator-mxnet] branch master updated: [Review Required] Adding Apache License header to the perl-package (#9554)

This is an automated email from the ASF dual-hosted git repository. haibin 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 32463f8 [Review Required] Adding

[GitHub] eric-haibin-lin commented on a change in pull request #9554: [Review Required] Adding Apache License header to the perl-package

eric-haibin-lin commented on a change in pull request #9554: [Review Required] Adding Apache License header to the perl-package URL: https://github.com/apache/incubator-mxnet/pull/9554#discussion_r164020859 ## File path: tools/license_header.py ## @@ -83,7 +83,7 @@

[incubator-mxnet] branch master updated: Adding License Header to docs folder (#9556)

This is an automated email from the ASF dual-hosted git repository. haibin 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 b8a832e Adding License Header to docs

[GitHub] eric-haibin-lin closed pull request #9556: [Review Required] Fixing Licenses in the Docs folder wherever possible

eric-haibin-lin closed pull request #9556: [Review Required] Fixing Licenses in the Docs folder wherever possible URL: https://github.com/apache/incubator-mxnet/pull/9556 This is a PR merged from a forked repository. As GitHub hides the original diff on merge, it is displayed below for

[GitHub] eric-haibin-lin commented on a change in pull request #9554: [Review Required] Adding Apache License header to the perl-package

eric-haibin-lin commented on a change in pull request #9554: [Review Required] Adding Apache License header to the perl-package URL: https://github.com/apache/incubator-mxnet/pull/9554#discussion_r164020300 ## File path: tools/license_header.py ## @@ -83,7 +83,7 @@

[GitHub] zheng-da commented on issue #8302: Refactor operators & MKLDNN

zheng-da commented on issue #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#issuecomment-353705559 Remaining MKLDNN-related tasks before merging the PR. - [x] Speedup inference with batch size 1. The goal is to reach the speed of old.mkldnn.

[GitHub] zheng-da commented on issue #8302: Refactor operators & MKLDNN

zheng-da commented on issue #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#issuecomment-353705559 Remaining MKLDNN-related tasks before merging the PR. - [x] Speedup inference with batch size 1. The goal is to reach the speed of old.mkldnn.

[GitHub] pengzhao-intel commented on a change in pull request #8302: Refactor operators & MKLDNN

pengzhao-intel commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r164019210 ## File path: .gitmodules ## @@ -22,3 +22,7 @@ [submodule "3rdparty/googletest"] path =

[GitHub] eric-haibin-lin closed pull request #9559: [Review Required] Fixing Licenses - Some more

eric-haibin-lin closed pull request #9559: [Review Required] Fixing Licenses - Some more URL: https://github.com/apache/incubator-mxnet/pull/9559 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:

[incubator-mxnet] branch master updated: License Adds - some more (#9559)

This is an automated email from the ASF dual-hosted git repository. haibin 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 7ee8a1f License Adds - some more

[GitHub] pengzhao-intel commented on issue #8302: Refactor operators & MKLDNN

pengzhao-intel commented on issue #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#issuecomment-360659821 @eric-haibin-lin @marcoabreu @cjolivier01 @szha @reminisce @zheng-da Do you have the time schedule of the merging for this PR? We

[GitHub] szha closed pull request #9566: Fixed the 'rm -rfv' command that was not cleaning the versioned master folder

szha closed pull request #9566: Fixed the 'rm -rfv' command that was not cleaning the versioned master folder URL: https://github.com/apache/incubator-mxnet/pull/9566 This is a PR merged from a forked repository. As GitHub hides the original diff on merge, it is displayed below for the

[incubator-mxnet] branch master updated: fixed the rm command that was not cleaning the versioned master folder (#9566)

This is an automated email from the ASF dual-hosted git repository. zhasheng pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git The following commit(s) were added to refs/heads/master by this push: new b647248 fixed the rm command that

[GitHub] rahul003 commented on a change in pull request #8302: Refactor operators & MKLDNN

rahul003 commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r164015700 ## File path: src/kvstore/kvstore_dist.h ## @@ -32,11 +32,6 @@ #include "mxnet/engine.h" #include "ps/ps.h"

[GitHub] rahul003 commented on a change in pull request #8302: Refactor operators & MKLDNN

rahul003 commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r164015700 ## File path: src/kvstore/kvstore_dist.h ## @@ -32,11 +32,6 @@ #include "mxnet/engine.h" #include "ps/ps.h"

[GitHub] kaonashi-tyc opened a new issue #9567: nd.argmax cause "Kernel Died"

kaonashi-tyc opened a new issue #9567: nd.argmax cause "Kernel Died" URL: https://github.com/apache/incubator-mxnet/issues/9567 ## Description Running the following code is causing my "Kernel Died" error on my Jupyter notebook: ```python from mxnet import nd as nd

[GitHub] rahul003 commented on issue #9558: Signum with grad compression

rahul003 commented on issue #9558: Signum with grad compression URL: https://github.com/apache/incubator-mxnet/pull/9558#issuecomment-360647400 It's fairly easy to change the compression to use 1bit. You need to change the bitmasks in the struct quantize_signum and dequantize_signum, and

[GitHub] rahul003 commented on issue #9558: Signum with grad compression

rahul003 commented on issue #9558: Signum with grad compression URL: https://github.com/apache/incubator-mxnet/pull/9558#issuecomment-360647400 It's fairly easy to change the compression to use 1bit. You need to change the bitmasks in the struct quantize_signum and dequantize_signum, and

[GitHub] marcoabreu commented on issue #9564: Fix non-MSVC build problem

marcoabreu commented on issue #9564: Fix non-MSVC build problem URL: https://github.com/apache/incubator-mxnet/pull/9564#issuecomment-360647292 Thanks! @KellenSunderland do you know which Cmake version we are using? In the dockerfile build_cuda, we're not explicitly specifying the

[GitHub] cjolivier01 commented on issue #9564: Fix non-MSVC build problem

cjolivier01 commented on issue #9564: Fix non-MSVC build problem URL: https://github.com/apache/incubator-mxnet/pull/9564#issuecomment-360646629 broken for linux because those are windows compile flags On Thu, Jan 25, 2018 at 4:14 PM Marco de Abreu

[GitHub] szha commented on issue #8439: Scaling between tensors

szha commented on issue #8439: Scaling between tensors URL: https://github.com/apache/incubator-mxnet/issues/8439#issuecomment-360646319 @apache/mxnet-committers: This issue has been inactive for the past 90 days. It has no label and needs triage. For general "how-to" questions,

[GitHub] szha commented on issue #8440: MXNet docker image mxnet/python has worse performance than outside docker environment.

szha commented on issue #8440: MXNet docker image mxnet/python has worse performance than outside docker environment. URL: https://github.com/apache/incubator-mxnet/issues/8440#issuecomment-360646326 @apache/mxnet-committers: This issue has been inactive for the past 90 days. It has no

[GitHub] cjolivier01 commented on issue #9545: Profiling discussion

cjolivier01 commented on issue #9545: Profiling discussion URL: https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360645754 I am not yet sure how to incorporate the other MKL output or where I would get it at runtime?

[GitHub] cjolivier01 commented on issue #9545: Profiling discussion

cjolivier01 commented on issue #9545: Profiling discussion URL: https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360644822 @TaoLv Some aggregate output added today (see screenshot below) Not all of that yet...

[GitHub] cjolivier01 commented on issue #9545: Profiling discussion

cjolivier01 commented on issue #9545: Profiling discussion URL: https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360645371 Like this? This is an automated message from the Apache Git Service. To respond to

[GitHub] cjolivier01 commented on issue #9545: Profiling discussion

cjolivier01 commented on issue #9545: Profiling discussion URL: https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360644943 You can dump those stats and reset every iteration, I suppose... This is an

[GitHub] cjolivier01 commented on issue #9545: Profiling discussion

cjolivier01 commented on issue #9545: Profiling discussion URL: https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360645274 

[GitHub] cjolivier01 commented on issue #9545: Profiling discussion

cjolivier01 commented on issue #9545: Profiling discussion URL: https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360644943 You can dump those stats and reset every iteration, I suppose... This is an

[GitHub] cjolivier01 commented on issue #9545: Profiling discussion

cjolivier01 commented on issue #9545: Profiling discussion URL: https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360644822 @TaoLv Some aggregate output added today (see screenshot in bottom of PR) Not all of that yet...

[GitHub] cjolivier01 commented on issue #8972: Profiling enhancements, python API, vtune and chrome tracing objects, etc.

cjolivier01 commented on issue #8972: Profiling enhancements, python API, vtune and chrome tracing objects, etc. URL: https://github.com/apache/incubator-mxnet/pull/8972#issuecomment-360298273 @hyandell Can you check out the Perl changes, in this commit, please?

[GitHub] marcoabreu commented on issue #9564: Fix non-MSVC build problem

marcoabreu commented on issue #9564: Fix non-MSVC build problem URL: https://github.com/apache/incubator-mxnet/pull/9564#issuecomment-360644127 On Windows, right? On Thu, Jan 25, 2018 at 4:08 PM, Chris Olivier wrote: > 3.9.6 > > On Thu, Jan

[GitHub] cjolivier01 commented on issue #9564: Fix non-MSVC build problem

cjolivier01 commented on issue #9564: Fix non-MSVC build problem URL: https://github.com/apache/incubator-mxnet/pull/9564#issuecomment-360642957 3.9.6 On Thu, Jan 25, 2018 at 4:04 PM, Marco de Abreu wrote: > Which version Cmake are you using? >

[GitHub] cjolivier01 closed issue #9501: LSTM doesn't work with Symbols and HybridBlocks

cjolivier01 closed issue #9501: LSTM doesn't work with Symbols and HybridBlocks URL: https://github.com/apache/incubator-mxnet/issues/9501 This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] yuxiangw commented on issue #9558: Signum with grad compression

yuxiangw commented on issue #9558: Signum with grad compression URL: https://github.com/apache/incubator-mxnet/pull/9558#issuecomment-360641078 Just + and -. We do not need 0 in Signum algorithm. If the gradient is 0, we basically can use either + or -. It does not really matter.

[GitHub] marcoabreu commented on issue #9553: test_operator_gpu.test_correlation fails

marcoabreu commented on issue #9553: test_operator_gpu.test_correlation fails URL: https://github.com/apache/incubator-mxnet/issues/9553#issuecomment-360642205 It also failed on the master branch:

[GitHub] marcoabreu commented on issue #9564: Fix non-MSVC build problem

marcoabreu commented on issue #9564: Fix non-MSVC build problem URL: https://github.com/apache/incubator-mxnet/pull/9564#issuecomment-360642083 Which version Cmake are you using? This is an automated message from the Apache

[GitHub] yuxiangw commented on issue #9558: Signum with grad compression

yuxiangw commented on issue #9558: Signum with grad compression URL: https://github.com/apache/incubator-mxnet/pull/9558#issuecomment-360641078 Just + and -. We do not need 0 in Signum algorithm. If the gradient is 0, we basically can use either + or -. It does not really matter.

[GitHub] yuxiangw commented on issue #9558: Signum with grad compression

yuxiangw commented on issue #9558: Signum with grad compression URL: https://github.com/apache/incubator-mxnet/pull/9558#issuecomment-360641078 Just + and -. RE: gradients are compressed when sending from worker to server, but not from server to worker. This is not ideal. For

[GitHub] cjolivier01 commented on issue #8972: Profiling enhancements, python API, vtune and chrome tracing objects, etc.

cjolivier01 commented on issue #8972: Profiling enhancements, python API, vtune and chrome tracing objects, etc. URL: https://github.com/apache/incubator-mxnet/pull/8972#issuecomment-360640413 Added aggregate stats ![screenshot from 2018-01-25

[GitHub] cjolivier01 commented on issue #8972: Profiling enhancements, python API, vtune and chrome tracing objects, etc.

cjolivier01 commented on issue #8972: Profiling enhancements, python API, vtune and chrome tracing objects, etc. URL: https://github.com/apache/incubator-mxnet/pull/8972#issuecomment-360640251 Added aggregate stats and output: Profile Statistics. Note that counter items

[GitHub] cjolivier01 commented on issue #8972: Profiling enhancements, python API, vtune and chrome tracing objects, etc.

cjolivier01 commented on issue #8972: Profiling enhancements, python API, vtune and chrome tracing objects, etc. URL: https://github.com/apache/incubator-mxnet/pull/8972#issuecomment-360640251 Added aggregate stats and output: Profile Statistics. Note that counter items

[GitHub] rahul003 commented on issue #9558: Signum with grad compression

rahul003 commented on issue #9558: Signum with grad compression URL: https://github.com/apache/incubator-mxnet/pull/9558#issuecomment-360640140 How many states do you want the gradient to be stored in? (+,- and 0), or only (+ and -)? Currently your code uses three states, in which case we

[GitHub] thinksanky opened a new pull request #9566: Fixed the 'rm -rfv' command that was not cleaning the versioned master folder

thinksanky opened a new pull request #9566: Fixed the 'rm -rfv' command that was not cleaning the versioned master folder URL: https://github.com/apache/incubator-mxnet/pull/9566 ## Description ## build_doc.sh script was not cleaning up the versioneWeb/versions/master folder due to a

[GitHub] yuxiangw commented on issue #9558: Signum with grad compression

yuxiangw commented on issue #9558: Signum with grad compression URL: https://github.com/apache/incubator-mxnet/pull/9558#issuecomment-360639214 @rahul003 The performance can be improved by a little bit I think (but not from the current implementation). I am thinking of changing the

[GitHub] yuxiangw commented on issue #9558: Signum with grad compression

yuxiangw commented on issue #9558: Signum with grad compression URL: https://github.com/apache/incubator-mxnet/pull/9558#issuecomment-360638860 Thanks @rahul003 for looking into this. This is used for a specific purpose --- running Signum over multiple GPU (and multiple machines) and

[GitHub] rahul003 commented on issue #9558: Signum with grad compression

rahul003 commented on issue #9558: Signum with grad compression URL: https://github.com/apache/incubator-mxnet/pull/9558#issuecomment-360638014 Thanks for contributing to this. The implementation looks fine. Performance should be similar to 2bit compression, right? But have you run

[GitHub] rahul003 commented on issue #8915: NVLink communication pattern updated

rahul003 commented on issue #8915: NVLink communication pattern updated URL: https://github.com/apache/incubator-mxnet/pull/8915#issuecomment-360627429 Could you please do these three things, 1. add some comments on the flow of data for the reduce function? The flow is not easy to

[GitHub] Laurawly commented on a change in pull request #8915: NVLink communication pattern updated

Laurawly commented on a change in pull request #8915: NVLink communication pattern updated URL: https://github.com/apache/incubator-mxnet/pull/8915#discussion_r163997689 ## File path: src/kvstore/comm.h ## @@ -526,101 +541,238 @@ class CommDevice : public Comm { }

[GitHub] anirudh2290 commented on issue #9553: test_operator_gpu.test_correlation fails

anirudh2290 commented on issue #9553: test_operator_gpu.test_correlation fails URL: https://github.com/apache/incubator-mxnet/issues/9553#issuecomment-360628473 are there any other instances of this failure apart from #9560 ? Does it consistently fail or intermittently ?

[GitHub] anirudh2290 commented on issue #9553: test_operator_gpu.test_correlation fails

anirudh2290 commented on issue #9553: test_operator_gpu.test_correlation fails URL: https://github.com/apache/incubator-mxnet/issues/9553#issuecomment-360628473 are there any other instances of this failure apart from #9560 ?

[GitHub] rahul003 commented on issue #8915: NVLink communication pattern updated

rahul003 commented on issue #8915: NVLink communication pattern updated URL: https://github.com/apache/incubator-mxnet/pull/8915#issuecomment-360627429 Could you please do these two things, 1. add some comments on the flow of data for the reduce function? The flow is not easy to follow

[GitHub] cjolivier01 commented on issue #9564: Fix non-MSVC build problem

cjolivier01 commented on issue #9564: Fix non-MSVC build problem URL: https://github.com/apache/incubator-mxnet/pull/9564#issuecomment-360624497 maybe because you have older Cmake that doesn?t know how to compile cuda as a first-class language? On The, Jan 25, 2018 at

[GitHub] sandeep-krishnamurthy closed issue #8911: can't load weight from hdf5 using mxnet as backend

sandeep-krishnamurthy closed issue #8911: can't load weight from hdf5 using mxnet as backend URL: https://github.com/apache/incubator-mxnet/issues/8911 This is an automated message from the Apache Git Service. To respond to

[GitHub] sandeep-krishnamurthy commented on issue #8911: can't load weight from hdf5 using mxnet as backend

sandeep-krishnamurthy commented on issue #8911: can't load weight from hdf5 using mxnet as backend URL: https://github.com/apache/incubator-mxnet/issues/8911#issuecomment-360620795 @R1ncy - Looking into the issue in this issue on the keras repo -

[GitHub] Garglesoap opened a new issue #9565: Are saved mxnet models GPU or system specific?

Garglesoap opened a new issue #9565: Are saved mxnet models GPU or system specific? URL: https://github.com/apache/incubator-mxnet/issues/9565 Model train system: AWS Ubuntu p2.xlarge, R 3.4.0, mxnet_1.0.1. Saved via: mx.model.save(A3.MXmodel, "Action/A3.MXmodel", iteration = 3000)

[GitHub] calendarbase commented on issue #9399: error: unable to load shared object /mxnet/libs/libmxnet.so

calendarbase commented on issue #9399: error: unable to load shared object /mxnet/libs/libmxnet.so URL: https://github.com/apache/incubator-mxnet/issues/9399#issuecomment-358603756 I managed to install OpenCV based on this: https://www.learnopencv.com/install-opencv3-on-ubuntu/

[GitHub] Garglesoap commented on issue #9399: error: unable to load shared object /mxnet/libs/libmxnet.so

Garglesoap commented on issue #9399: error: unable to load shared object /mxnet/libs/libmxnet.so URL: https://github.com/apache/incubator-mxnet/issues/9399#issuecomment-360616809 FWIW I consistently ran into this error on Amazon linux, but switching to Ubuntu and the LD library issue

[GitHub] Laurawly commented on a change in pull request #8915: NVLink communication pattern updated

Laurawly commented on a change in pull request #8915: NVLink communication pattern updated URL: https://github.com/apache/incubator-mxnet/pull/8915#discussion_r163980218 ## File path: src/kvstore/comm.h ## @@ -526,101 +541,238 @@ class CommDevice : public Comm { }

[GitHub] rahul003 commented on a change in pull request #8915: NVLink communication pattern updated

rahul003 commented on a change in pull request #8915: NVLink communication pattern updated URL: https://github.com/apache/incubator-mxnet/pull/8915#discussion_r163978620 ## File path: src/kvstore/comm.h ## @@ -526,101 +541,238 @@ class CommDevice : public Comm { }

[GitHub] zhreshold commented on a change in pull request #9457: Move get_images to test_utils.py, so it can be used from other tests. Fix documentation with recent im2rec changes.

zhreshold commented on a change in pull request #9457: Move get_images to test_utils.py, so it can be used from other tests. Fix documentation with recent im2rec changes. URL: https://github.com/apache/incubator-mxnet/pull/9457#discussion_r163971202 ## File path:

[GitHub] zhreshold commented on a change in pull request #9457: Move get_images to test_utils.py, so it can be used from other tests. Fix documentation with recent im2rec changes.

zhreshold commented on a change in pull request #9457: Move get_images to test_utils.py, so it can be used from other tests. Fix documentation with recent im2rec changes. URL: https://github.com/apache/incubator-mxnet/pull/9457#discussion_r163971039 ## File path:

[GitHub] marcoabreu commented on issue #9553: test_operator_gpu.test_correlation fails

marcoabreu commented on issue #9553: test_operator_gpu.test_correlation fails URL: https://github.com/apache/incubator-mxnet/issues/9553#issuecomment-360603859 http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-9522/1/pipeline

[GitHub] marcoabreu commented on issue #9553: test_operator_gpu.test_correlation fails

marcoabreu commented on issue #9553: test_operator_gpu.test_correlation fails URL: https://github.com/apache/incubator-mxnet/issues/9553#issuecomment-360603112 http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-9560/1/pipeline

[GitHub] Laurawly commented on a change in pull request #8915: NVLink communication pattern updated

Laurawly commented on a change in pull request #8915: NVLink communication pattern updated URL: https://github.com/apache/incubator-mxnet/pull/8915#discussion_r163969344 ## File path: src/kvstore/comm.h ## @@ -526,101 +541,238 @@ class CommDevice : public Comm { }

[GitHub] marcoabreu commented on issue #9564: Fix non-MSVC build problem

marcoabreu commented on issue #9564: Fix non-MSVC build problem URL: https://github.com/apache/incubator-mxnet/pull/9564#issuecomment-360598612 I know... But how was it able to break the build and pass CI? This is an

[GitHub] eric-haibin-lin closed pull request #9562: Broken Link Fixes

eric-haibin-lin closed pull request #9562: Broken Link Fixes URL: https://github.com/apache/incubator-mxnet/pull/9562 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 pull

[incubator-mxnet] branch master updated: Broken Link Fixes (#9562)

This is an automated email from the ASF dual-hosted git repository. haibin 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 1fc46e0 Broken Link Fixes (#9562)

[GitHub] jesterhazy commented on issue #9501: LSTM doesn't work with Symbols and HybridBlocks

jesterhazy commented on issue #9501: LSTM doesn't work with Symbols and HybridBlocks URL: https://github.com/apache/incubator-mxnet/issues/9501#issuecomment-360575548 @aidan-plenert-macdonald you could try: - training your model - saving the weights with `net.save_params(path,

[GitHub] piiswrong commented on issue #9535: Tutorials index page update

piiswrong commented on issue #9535: Tutorials index page update URL: https://github.com/apache/incubator-mxnet/pull/9535#issuecomment-360567663 Some comments: 1. tutorial is not showing on the top bar. 2. Could you show TL;DR in a bubble with `.. TL;DR:`? 3. the Python tab isn't

[GitHub] piiswrong commented on issue #9535: Tutorials index page update

piiswrong commented on issue #9535: Tutorials index page update URL: https://github.com/apache/incubator-mxnet/pull/9535#issuecomment-360567663 Some comments: 1. tutorial is not showing on the top bar. 2. Could you show TL;DR in a bubble with `.. TL;DR:`? 3. the Python tab isn't

[GitHub] piiswrong commented on issue #9535: Tutorials index page update

piiswrong commented on issue #9535: Tutorials index page update URL: https://github.com/apache/incubator-mxnet/pull/9535#issuecomment-360567663 Some comments: 1. tutorial is not showing on the top bar. 2. Could you show TL;DR in a bubble with `.. TL;DR:`? 3. the Python tab isn't

[GitHub] piiswrong closed pull request #9046: Fix example example/reinforcement-learning/a3c

piiswrong closed pull request #9046: Fix example example/reinforcement-learning/a3c URL: https://github.com/apache/incubator-mxnet/pull/9046 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

[incubator-mxnet] branch master updated: Fix example example/reinforcement-learning/a3c (#9046)

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 52219e8 Fix example

[incubator-mxnet] branch master updated: fix CreateOp to accept dtypes of uint8 and int32 (#9561)

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 9aaa29d fix CreateOp to accept dtypes

[GitHub] piiswrong closed pull request #9561: fix CreateOp of slice_channel to accept dtypes of uint8 and int32

piiswrong closed pull request #9561: fix CreateOp of slice_channel to accept dtypes of uint8 and int32 URL: https://github.com/apache/incubator-mxnet/pull/9561 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

[GitHub] nswamy closed pull request #9564: Fix non-MSVC build problem

nswamy closed pull request #9564: Fix non-MSVC build problem URL: https://github.com/apache/incubator-mxnet/pull/9564 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 pull

[incubator-mxnet] branch master updated: Fix non-MSVC build problem (#9564)

This is an automated email from the ASF dual-hosted git repository. nswamy 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 542c3d7 Fix non-MSVC build problem

[GitHub] reminisce commented on a change in pull request #9562: Broken Link Fixes

reminisce commented on a change in pull request #9562: Broken Link Fixes URL: https://github.com/apache/incubator-mxnet/pull/9562#discussion_r163921703 ## File path: docs/faq/security.md ## @@ -0,0 +1,24 @@ +# MXNet Security best practices Review comment: Best

[GitHub] reminisce commented on a change in pull request #9562: Broken Link Fixes

reminisce commented on a change in pull request #9562: Broken Link Fixes URL: https://github.com/apache/incubator-mxnet/pull/9562#discussion_r163922585 ## File path: docs/faq/add_op_in_backend.md ## @@ -0,0 +1,608 @@ +# A Beginner's Guide to Implementing Operators in MXNet

[GitHub] reminisce commented on a change in pull request #9562: Broken Link Fixes

reminisce commented on a change in pull request #9562: Broken Link Fixes URL: https://github.com/apache/incubator-mxnet/pull/9562#discussion_r163922754 ## File path: docs/faq/add_op_in_backend.md ## @@ -0,0 +1,608 @@ +# A Beginner's Guide to Implementing Operators in MXNet

  1   2   >