[GitHub] larroy commented on a change in pull request #11036: [MXNET-472] Add ccache support to CI

larroy commented on a change in pull request #11036: [MXNET-472] Add ccache support to CI URL: https://github.com/apache/incubator-mxnet/pull/11036#discussion_r191077899 ## File path: make/config.mk ## @@ -37,8 +37,8 @@ # choice of compiler #

[incubator-mxnet] branch master updated: bump up version number to 1.3.0 to make nightly build to build with right source (#11070)

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 79bcd9c bump up version number to

[GitHub] sandeep-krishnamurthy commented on issue #10126: Change a block

sandeep-krishnamurthy commented on issue #10126: Change a block URL: https://github.com/apache/incubator-mxnet/issues/10126#issuecomment-392366695 This question is very likely to get answered at - https://discuss.mxnet.io/

[GitHub] sl1pkn07 commented on issue #10558: NNVM build failed in the newst mxnet version

sl1pkn07 commented on issue #10558: NNVM build failed in the newst mxnet version URL: https://github.com/apache/incubator-mxnet/issues/10558#issuecomment-392367383 just for note new Cuda 9.2.88.1 build OK with GCC 7.3.1

[incubator-mxnet] branch master updated: [MXNET-471] Add Builder class for Scala Module and DataBatch to simplify construction (#11045)

This is an automated email from the ASF dual-hosted git repository. liuyizhi 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 5153370 [MXNET-471] Add Builder

[GitHub] yzhliu closed pull request #11045: [MXNET-471] Add Builder class for Scala Module and DataBatch to simplify construction

yzhliu closed pull request #11045: [MXNET-471] Add Builder class for Scala Module and DataBatch to simplify construction URL: https://github.com/apache/incubator-mxnet/pull/11045 This is a PR merged from a forked repository. As GitHub hides the original diff on merge, it is displayed

[GitHub] WalterMa commented on issue #10242: [MXNET-137]fix parameters name inconsistent for Proposal OP and Multi Proposal OP

WalterMa commented on issue #10242: [MXNET-137]fix parameters name inconsistent for Proposal OP and Multi Proposal OP URL: https://github.com/apache/incubator-mxnet/pull/10242#issuecomment-392319146 Same strange issue: symbol and ndarray proposal operator api are inconsistent. @wkcn Do

[GitHub] marcoabreu commented on a change in pull request #11036: [MXNET-472] Add ccache support to CI

marcoabreu commented on a change in pull request #11036: [MXNET-472] Add ccache support to CI URL: https://github.com/apache/incubator-mxnet/pull/11036#discussion_r191072956 ## File path: make/config.mk ## @@ -37,8 +37,8 @@ # choice of compiler #

[GitHub] WalterMa commented on issue #9644: mxnet.base.MXNetError: [19:28:21] src/core/symbolic.cc:72: Symbol.ComposeKeyword argument name cls_score not found.

WalterMa commented on issue #9644: mxnet.base.MXNetError: [19:28:21] src/core/symbolic.cc:72: Symbol.ComposeKeyword argument name cls_score not found. URL: https://github.com/apache/incubator-mxnet/issues/9644#issuecomment-392317531 Same issue here. `mx.sym.contrib.Proposal` will

[GitHub] WalterMa commented on issue #9644: mxnet.base.MXNetError: [19:28:21] src/core/symbolic.cc:72: Symbol.ComposeKeyword argument name cls_score not found.

WalterMa commented on issue #9644: mxnet.base.MXNetError: [19:28:21] src/core/symbolic.cc:72: Symbol.ComposeKeyword argument name cls_score not found. URL: https://github.com/apache/incubator-mxnet/issues/9644#issuecomment-392317531 Same issue here. `mx.sym.contrib.Proposal` will

[GitHub] sandeep-krishnamurthy closed issue #10878: Typo Erro in Optimizer.R

sandeep-krishnamurthy closed issue #10878: Typo Erro in Optimizer.R URL: https://github.com/apache/incubator-mxnet/issues/10878 This is an automated message from the Apache Git Service. To respond to the message, please log

[GitHub] sandeep-krishnamurthy commented on issue #10903: how to set output value but not output possibility

sandeep-krishnamurthy commented on issue #10903: how to set output value but not output possibility URL: https://github.com/apache/incubator-mxnet/issues/10903#issuecomment-392383706 Please ask usage related questions at https://discuss.mxnet.io/ for more people being able to answer,

[GitHub] sandeep-krishnamurthy commented on issue #10905: Why I can't run gpu model on jeston TX2?

sandeep-krishnamurthy commented on issue #10905: Why I can't run gpu model on jeston TX2? URL: https://github.com/apache/incubator-mxnet/issues/10905#issuecomment-392383821 @lovehuanhuan - Please reopen if issue still persists.

[GitHub] sandeep-krishnamurthy closed issue #10903: how to set output value but not output possibility

sandeep-krishnamurthy closed issue #10903: how to set output value but not output possibility URL: https://github.com/apache/incubator-mxnet/issues/10903 This is an automated message from the Apache Git Service. To respond

[GitHub] sandeep-krishnamurthy closed issue #10905: Why I can't run gpu model on jeston TX2?

sandeep-krishnamurthy closed issue #10905: Why I can't run gpu model on jeston TX2? URL: https://github.com/apache/incubator-mxnet/issues/10905 This is an automated message from the Apache Git Service. To respond to the

[GitHub] sandeep-krishnamurthy commented on issue #10967: Is the doc for launch.py parameters outdated？

sandeep-krishnamurthy commented on issue #10967: Is the doc for launch.py parameters outdated？ URL: https://github.com/apache/incubator-mxnet/issues/10967#issuecomment-392384731 @rahul003 - Do we need to update the doc to reflect that --launcher is mapped to --cluster

[GitHub] sandeep-krishnamurthy commented on issue #11015: Check failed: err == CUSOLVER_STATUS_SUCCESS (7 vs. 0) Create cusolver handle failed

sandeep-krishnamurthy commented on issue #11015: Check failed: err == CUSOLVER_STATUS_SUCCESS (7 vs. 0) Create cusolver handle failed URL: https://github.com/apache/incubator-mxnet/issues/11015#issuecomment-392389640 Hello looks like Ubuntu (I am assuming?) version, CUDA version and cuDNN

[GitHub] sandeep-krishnamurthy commented on issue #11022: nd.stack with different input sizes

sandeep-krishnamurthy commented on issue #11022: nd.stack with different input sizes URL: https://github.com/apache/incubator-mxnet/issues/11022#issuecomment-392390447 That kind of stack should not be allowed. This looks like a bug in stack().

[GitHub] sandeep-krishnamurthy commented on issue #11028: Pre-trained Shufflenet model fails during inference on mxnet-mkl==1.2.0

sandeep-krishnamurthy commented on issue #11028: Pre-trained Shufflenet model fails during inference on mxnet-mkl==1.2.0 URL: https://github.com/apache/incubator-mxnet/issues/11028#issuecomment-392390647 This issue is fixed on Master. This fix is not applied on 1.2 branch. CC

[GitHub] marcoabreu commented on issue #11066: migrating docs build and publish job to secure nodes

marcoabreu commented on issue #11066: migrating docs build and publish job to secure nodes URL: https://github.com/apache/incubator-mxnet/pull/11066#issuecomment-392392610 LGTM. Do you need any help? This is an automated

[GitHub] liuzx32 commented on issue #11046: run mxnet on spark throws error: libopenblas.so.0: cannot open shared object file: No such file or directory

liuzx32 commented on issue #11046: run mxnet on spark throws error: libopenblas.so.0: cannot open shared object file: No such file or directory URL: https://github.com/apache/incubator-mxnet/issues/11046#issuecomment-392400717 @sandeep-krishnamurthy Thank you for the answer, I have

[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 e7a7954 Bump the publish

[GitHub] sandeep-krishnamurthy commented on issue #10782: Multiple calls to mx.sym.Variable with same name: Dangerous / unexpected behavior?

sandeep-krishnamurthy commented on issue #10782: Multiple calls to mx.sym.Variable with same name: Dangerous / unexpected behavior? URL: https://github.com/apache/incubator-mxnet/issues/10782#issuecomment-392381803 Resolving the issue, because, with 2 issues referenced, this is an

[GitHub] sandeep-krishnamurthy closed issue #10782: Multiple calls to mx.sym.Variable with same name: Dangerous / unexpected behavior?

sandeep-krishnamurthy closed issue #10782: Multiple calls to mx.sym.Variable with same name: Dangerous / unexpected behavior? URL: https://github.com/apache/incubator-mxnet/issues/10782 This is an automated message from the

[GitHub] sandeep-krishnamurthy commented on issue #10834: Share Numpy Arrays between processes like NDArrays?

sandeep-krishnamurthy commented on issue #10834: Share Numpy Arrays between processes like NDArrays? URL: https://github.com/apache/incubator-mxnet/issues/10834#issuecomment-392382094 Please ask usage related questions at https://discuss.mxnet.io/ for more people being able to answer,

[GitHub] sandeep-krishnamurthy closed issue #10834: Share Numpy Arrays between processes like NDArrays?

sandeep-krishnamurthy closed issue #10834: Share Numpy Arrays between processes like NDArrays? URL: https://github.com/apache/incubator-mxnet/issues/10834 This is an automated message from the Apache Git Service. To respond

[GitHub] sandeep-krishnamurthy commented on issue #10832: custom kernel

sandeep-krishnamurthy commented on issue #10832: custom kernel URL: https://github.com/apache/incubator-mxnet/issues/10832#issuecomment-392382055 Please ask usage related questions at https://discuss.mxnet.io/ for more people being able to answer, search and refer in future. We want to

[GitHub] sandeep-krishnamurthy closed issue #10832: custom kernel

sandeep-krishnamurthy closed issue #10832: custom kernel URL: https://github.com/apache/incubator-mxnet/issues/10832 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub

[GitHub] sandeep-krishnamurthy closed issue #10965: How to show the ctc loss?

sandeep-krishnamurthy closed issue #10965: How to show the ctc loss? URL: https://github.com/apache/incubator-mxnet/issues/10965 This is an automated message from the Apache Git Service. To respond to the message, please log

[GitHub] sandeep-krishnamurthy commented on issue #10965: How to show the ctc loss?

sandeep-krishnamurthy commented on issue #10965: How to show the ctc loss? URL: https://github.com/apache/incubator-mxnet/issues/10965#issuecomment-392384487 Please ask usage related questions at https://discuss.mxnet.io/ for more people being able to answer, search and refer in future.

[GitHub] sandeep-krishnamurthy closed issue #10964: linalg_potri() Error: linalg_impl.h:606: Check failed: ret == 0 (3 vs. 0) spotri failed in lapack on cpu.

sandeep-krishnamurthy closed issue #10964: linalg_potri() Error: linalg_impl.h:606: Check failed: ret == 0 (3 vs. 0) spotri failed in lapack on cpu. URL: https://github.com/apache/incubator-mxnet/issues/10964 This is an

[GitHub] sandeep-krishnamurthy commented on issue #11046: run mxnet on spark throws error: libopenblas.so.0: cannot open shared object file: No such file or directory

sandeep-krishnamurthy commented on issue #11046: run mxnet on spark throws error: libopenblas.so.0: cannot open shared object file: No such file or directory URL: https://github.com/apache/incubator-mxnet/issues/11046#issuecomment-392391483 Hello you need to install Openblas. Please

[GitHub] sandeep-krishnamurthy closed issue #11043: Does convolution layer has weight decay param? just as fc layer "wd_mult"

sandeep-krishnamurthy closed issue #11043: Does convolution layer has weight decay param? just as fc layer "wd_mult" URL: https://github.com/apache/incubator-mxnet/issues/11043 This is an automated message from the Apache

[GitHub] sandeep-krishnamurthy commented on issue #11043: Does convolution layer has weight decay param? just as fc layer "wd_mult"

sandeep-krishnamurthy commented on issue #11043: Does convolution layer has weight decay param? just as fc layer "wd_mult" URL: https://github.com/apache/incubator-mxnet/issues/11043#issuecomment-392391345 Please ask usage related questions at https://discuss.mxnet.io/ for more people

[GitHub] sandeep-krishnamurthy commented on issue #11040: Big acc difference when fine-tune using vgg16 and resnet-50

sandeep-krishnamurthy commented on issue #11040: Big acc difference when fine-tune using vgg16 and resnet-50 URL: https://github.com/apache/incubator-mxnet/issues/11040#issuecomment-392391326 Hello, I would also encourage you to ask your question at https://discuss.mxnet.io/ for more

[GitHub] sandeep-krishnamurthy commented on issue #10866: Scala Module API not resizing properly when returning to original binded size

sandeep-krishnamurthy commented on issue #10866: Scala Module API not resizing properly when returning to original binded size URL: https://github.com/apache/incubator-mxnet/issues/10866#issuecomment-392382344 @lanking520 - Any inputs on this?

[GitHub] sandeep-krishnamurthy closed issue #10916: `mxnet.nd.smooth_l1(mxnet.nd.array([1, 2, 3]))' core dumped

sandeep-krishnamurthy closed issue #10916: `mxnet.nd.smooth_l1(mxnet.nd.array([1, 2, 3]))' core dumped URL: https://github.com/apache/incubator-mxnet/issues/10916 This is an automated message from the Apache Git Service. To

[GitHub] sandeep-krishnamurthy closed issue #11039: Linux-R

sandeep-krishnamurthy closed issue #11039: Linux-R URL: https://github.com/apache/incubator-mxnet/issues/11039 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use

[GitHub] sandeep-krishnamurthy commented on issue #11039: Linux-R

sandeep-krishnamurthy commented on issue #11039: Linux-R URL: https://github.com/apache/incubator-mxnet/issues/11039#issuecomment-392390821 Hello, Please refer installation guide, choose Linux and R from the options: http://mxnet.incubator.apache.org/install/index.html Please

[GitHub] sandeep-krishnamurthy closed issue #11050: How to get the shape of input variable for both nd and sym in hybrid_forward dynamically?

sandeep-krishnamurthy closed issue #11050: How to get the shape of input variable for both nd and sym in hybrid_forward dynamically? URL: https://github.com/apache/incubator-mxnet/issues/11050 This is an automated message

[GitHub] sandeep-krishnamurthy commented on issue #11050: How to get the shape of input variable for both nd and sym in hybrid_forward dynamically?

sandeep-krishnamurthy commented on issue #11050: How to get the shape of input variable for both nd and sym in hybrid_forward dynamically? URL: https://github.com/apache/incubator-mxnet/issues/11050#issuecomment-392391611 Please ask usage related questions at https://discuss.mxnet.io/ for

[GitHub] Paul0M commented on issue #11040: Big acc difference when fine-tune using vgg16 and resnet-50

Paul0M commented on issue #11040: Big acc difference when fine-tune using vgg16 and resnet-50 URL: https://github.com/apache/incubator-mxnet/issues/11040#issuecomment-392398470 @wsokolow GPU This is an automated message

[GitHub] azai91 commented on a change in pull request #11026: Test/mkl dnn act

azai91 commented on a change in pull request #11026: Test/mkl dnn act URL: https://github.com/apache/incubator-mxnet/pull/11026#discussion_r191116270 ## File path: src/operator/nn/mkldnn/mkldnn_sum.cc ## @@ -62,12 +62,9 @@ void MKLDNNSumForward(const nnvm::NodeAttrs&

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

sandeep-krishnamurthy 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-392383502 Please reopen in case if issue still persists.

[GitHub] sandeep-krishnamurthy closed issue #10890: [R] Is it possible to train Mxnet symbols and a GLM pretrained model as loss?

sandeep-krishnamurthy closed 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 This is an automated message from the

[GitHub] sandeep-krishnamurthy commented on issue #11071: GPU memory usage for VGG16 prediction

sandeep-krishnamurthy commented on issue #11071: GPU memory usage for VGG16 prediction URL: https://github.com/apache/incubator-mxnet/issues/11071#issuecomment-392391927 VGG is a dense network with many parameters. GPU memory is for complete model. I think this is normal. What is the

[GitHub] sandeep-krishnamurthy commented on issue #11062: how to manually occupy all gpu memory like tensorflow?

sandeep-krishnamurthy commented on issue #11062: how to manually occupy all gpu memory like tensorflow? URL: https://github.com/apache/incubator-mxnet/issues/11062#issuecomment-392391821 I think MXNet allocates the required memory, there is no option to forcefully occupy full memory.

[GitHub] zheng-da commented on issue #10921: Test cases improvement for MKLDNN on Gluon

zheng-da commented on issue #10921: Test cases improvement for MKLDNN on Gluon URL: https://github.com/apache/incubator-mxnet/pull/10921#issuecomment-392403691 @marcoabreu I think we should remove the tests for generating huge weight matrices, as @juliusshufan is doing right now for Dense.

[GitHub] liuzx32 closed issue #11011: ../../tools/launch.py -n 2 -s 2 --launcher yarn python train_mnist.py --network lenet --kv-store dist_sync

liuzx32 closed issue #11011: ../../tools/launch.py -n 2 -s 2 --launcher yarn python train_mnist.py --network lenet --kv-store dist_sync URL: https://github.com/apache/incubator-mxnet/issues/11011 This is an automated

[GitHub] liuzx32 commented on issue #11011: ../../tools/launch.py -n 2 -s 2 --launcher yarn python train_mnist.py --network lenet --kv-store dist_sync

liuzx32 commented on issue #11011: ../../tools/launch.py -n 2 -s 2 --launcher yarn python train_mnist.py --network lenet --kv-store dist_sync URL: https://github.com/apache/incubator-mxnet/issues/11011#issuecomment-392404579 @sandeep-krishnamurthy Thank you very much!

[GitHub] liuzx32 commented on issue #11046: run mxnet on spark throws error: libopenblas.so.0: cannot open shared object file: No such file or directory

liuzx32 commented on issue #11046: run mxnet on spark throws error: libopenblas.so.0: cannot open shared object file: No such file or directory URL: https://github.com/apache/incubator-mxnet/issues/11046#issuecomment-392405364 @sandeep-krishnamurthy Perhaps mxnet can complie with static

[GitHub] liuzx32 closed issue #11046: run mxnet on spark throws error: libopenblas.so.0: cannot open shared object file: No such file or directory

liuzx32 closed issue #11046: run mxnet on spark throws error: libopenblas.so.0: cannot open shared object file: No such file or directory URL: https://github.com/apache/incubator-mxnet/issues/11046 This is an automated

[GitHub] catonlion commented on issue #3685: Cannot train on multiple nodes, because port blocked by firewall

catonlion commented on issue #3685: Cannot train on multiple nodes, because port blocked by firewall URL: https://github.com/apache/incubator-mxnet/issues/3685#issuecomment-392406332 @mli how to limit the port range

[GitHub] WalterMa commented on issue #10242: [MXNET-137]fix parameters name inconsistent for Proposal OP and Multi Proposal OP

WalterMa commented on issue #10242: [MXNET-137]fix parameters name inconsistent for Proposal OP and Multi Proposal OP URL: https://github.com/apache/incubator-mxnet/pull/10242#issuecomment-392407319 @wkcn Thanks a lot! Just personal opinion: Would it be better to replace

[GitHub] zheng-da commented on a change in pull request #11026: Test/mkl dnn act

zheng-da commented on a change in pull request #11026: Test/mkl dnn act URL: https://github.com/apache/incubator-mxnet/pull/11026#discussion_r191104011 ## File path: src/operator/nn/mkldnn/mkldnn_sum.cc ## @@ -62,12 +62,9 @@ void MKLDNNSumForward(const nnvm::NodeAttrs&

[GitHub] marcoabreu commented on issue #10921: Test cases improvement for MKLDNN on Gluon

marcoabreu commented on issue #10921: Test cases improvement for MKLDNN on Gluon URL: https://github.com/apache/incubator-mxnet/pull/10921#issuecomment-392412662 I think they were originally introduced to test int32 overflow, but I totally agree that these are suboptimal and should be

[GitHub] zheng-da commented on issue #10921: Test cases improvement for MKLDNN on Gluon

zheng-da commented on issue #10921: Test cases improvement for MKLDNN on Gluon URL: https://github.com/apache/incubator-mxnet/pull/10921#issuecomment-392415271 +1 for C++ unit test This is an automated message from the Apache

[GitHub] maybepossible opened a new issue #11073: Data copy between cpu adn gpu

maybepossible opened a new issue #11073: Data copy between cpu adn gpu URL: https://github.com/apache/incubator-mxnet/issues/11073 I use jetson tx1 for inference. It said that there's data copy between CPU and GPU memory. As CPU and GPU share the same memory, Is there a way to not use the

[GitHub] maybepossible opened a new issue #11074: Data copy between cpu and gpu in jetson Tx1

maybepossible opened a new issue #11074: Data copy between cpu and gpu in jetson Tx1 URL: https://github.com/apache/incubator-mxnet/issues/11074 I use jetson tx1 for inference. It said that there's data copy between CPU and GPU memory. As CPU and GPU share the same memory, Is there a way

[GitHub] maybepossible closed issue #11073: Data copy between cpu adn gpu

maybepossible closed issue #11073: Data copy between cpu adn gpu URL: https://github.com/apache/incubator-mxnet/issues/11073 This is an automated message from the Apache Git Service. To respond to the message, please log on

[GitHub] wkcn commented on issue #10242: [MXNET-137]fix parameters name inconsistent for Proposal OP and Multi Proposal OP

wkcn commented on issue #10242: [MXNET-137]fix parameters name inconsistent for Proposal OP and Multi Proposal OP URL: https://github.com/apache/incubator-mxnet/pull/10242#issuecomment-392326302 @WalterMa Yes. If you replace cls_prob with cls_score in [this