[GitHub] szha commented on issue #8291: Import error in SSD example

szha commented on issue #8291: Import error in SSD example URL: https://github.com/apache/incubator-mxnet/issues/8291#issuecomment-358631746 @apache/mxnet-committers: This issue has been inactive for the past 90 days. It has no label and needs triage. For general "how-to"

[GitHub] some-guy1 closed issue #9453: [R] CNN Memory Leak - Needs to somehow call Garbage Collector

some-guy1 closed issue #9453: [R] CNN Memory Leak - Needs to somehow call Garbage Collector URL: https://github.com/apache/incubator-mxnet/issues/9453 This is an automated message from the Apache Git Service. To respond to

[GitHub] ilibx opened a new issue #9477: distributed trainning with mxnet, get error "ImportError: No module named numpy"

ilibx opened a new issue #9477: distributed trainning with mxnet, get error "ImportError: No module named numpy" URL: https://github.com/apache/incubator-mxnet/issues/9477 ## Description Following by the install guide i built mxnet in the docker, and started multi-container then i run

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

TaoLv commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162370376 ## File path: src/operator/nn/mkldnn/mkldnn_pooling-inl.h ## @@ -0,0 +1,393 @@ +/* + * Licensed to the Apache

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

TaoLv commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162370497 ## File path: src/operator/nn/pooling.cc ## @@ -32,67 +33,304 @@ #if MXNET_USE_NNPACK == 1 #include

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

TaoLv commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162370448 ## File path: src/operator/nn/mkldnn/mkldnn_pooling-inl.h ## @@ -0,0 +1,393 @@ +/* + * Licensed to the Apache

[GitHub] larroy commented on a change in pull request #9397: Remove incorrect link flags when using OPENCV. Exclude binaries from lint.

larroy commented on a change in pull request #9397: Remove incorrect link flags when using OPENCV. Exclude binaries from lint. URL: https://github.com/apache/incubator-mxnet/pull/9397#discussion_r162336205 ## File path: python/mxnet/libinfo.py ## @@ -32,7 +32,11 @@ def

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

TaoLv commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162353890 ## File path: src/operator/nn/mkldnn/mkldnn_lrn-inl.h ## @@ -0,0 +1,139 @@ +/* + * Licensed to the Apache Software

[GitHub] some-guy1 commented on issue #9453: [R] CNN Memory Leak - Needs to somehow call Garbage Collector

some-guy1 commented on issue #9453: [R] CNN Memory Leak - Needs to somehow call Garbage Collector URL: https://github.com/apache/incubator-mxnet/issues/9453#issuecomment-358619167 Oh!! I had no idea you could do that. Ok well I tried what you suggested and **problem solved!**

[GitHub] szha commented on issue #7771: Error when loading JSON network definition with c++ api

szha commented on issue #7771: Error when loading JSON network definition with c++ api URL: https://github.com/apache/incubator-mxnet/issues/7771#issuecomment-358631739 @apache/mxnet-committers: This issue has been inactive for the past 90 days. It has no label and needs triage.

[GitHub] szha commented on issue #6846: Check failed: err == cudaSuccess (8 vs. 0) Name: MapPlanKernel ErrStr:invalid device function

szha commented on issue #6846: Check failed: err == cudaSuccess (8 vs. 0) Name: MapPlanKernel ErrStr:invalid device function URL: https://github.com/apache/incubator-mxnet/issues/6846#issuecomment-358631735 @apache/mxnet-committers: This issue has been inactive for the past 90 days. It

[GitHub] ilibx closed issue #9477: distributed trainning with mxnet, get error "ImportError: No module named numpy"

ilibx closed issue #9477: distributed trainning with mxnet, get error "ImportError: No module named numpy" URL: https://github.com/apache/incubator-mxnet/issues/9477 This is an automated message from the Apache Git Service.

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

TaoLv commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162354585 ## File path: src/operator/nn/mkldnn/mkldnn_pooling-inl.h ## @@ -0,0 +1,393 @@ +/* + * Licensed to the Apache

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

TaoLv commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162370932 ## File path: src/operator/nn/pooling.cc ## @@ -32,67 +33,304 @@ #if MXNET_USE_NNPACK == 1 #include

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

TaoLv commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162370719 ## File path: src/operator/nn/pooling.cc ## @@ -131,8 +369,57 @@ For 3-D pooling, an additional *depth* dimension

[GitHub] larroy commented on issue #9476: libmxnet.so: undefined symbol: dorgqr_?????????????????

larroy commented on issue #9476: libmxnet.so: undefined symbol: dorgqr_? URL: https://github.com/apache/incubator-mxnet/issues/9476#issuecomment-358723838 please provide more info on how you compiled, platform etc.

[GitHub] larroy commented on a change in pull request #9397: Remove incorrect link flags when using OPENCV. Exclude binaries from lint.

larroy commented on a change in pull request #9397: Remove incorrect link flags when using OPENCV. Exclude binaries from lint. URL: https://github.com/apache/incubator-mxnet/pull/9397#discussion_r162398696 ## File path: CMakeLists.txt ## @@ -584,6 +577,12 @@ if(MSVC AND

[GitHub] meissnereric commented on issue #9295: test_operator.test_laop_3 hangs

meissnereric commented on issue #9295: test_operator.test_laop_3 hangs URL: https://github.com/apache/incubator-mxnet/issues/9295#issuecomment-358732072 Ok new update to this story. I synced to the latest changes from master branch for each of them, and now I can't reproduce the

[GitHub] mseeger commented on issue #9295: test_operator.test_laop_3 hangs

mseeger commented on issue #9295: test_operator.test_laop_3 hangs URL: https://github.com/apache/incubator-mxnet/issues/9295#issuecomment-358735928 Are we sure the test hangs due to the issue you saw? Well can watch for a while, see whether issues that Marco raised, are still present

[GitHub] cjolivier01 commented on a change in pull request #9464: refactor logging in infer storage pass

cjolivier01 commented on a change in pull request #9464: refactor logging in infer storage pass URL: https://github.com/apache/incubator-mxnet/pull/9464#discussion_r162443172 ## File path: src/common/utils.h ## @@ -364,6 +364,50 @@ inline std::string dev_type_string(const

[GitHub] cjolivier01 commented on a change in pull request #9464: refactor logging in infer storage pass

cjolivier01 commented on a change in pull request #9464: refactor logging in infer storage pass URL: https://github.com/apache/incubator-mxnet/pull/9464#discussion_r162442133 ## File path: src/common/utils.h ## @@ -426,13 +428,13 @@ inline void LogStorageFallback(const

[GitHub] cjolivier01 commented on a change in pull request #9464: refactor logging in infer storage pass

cjolivier01 commented on a change in pull request #9464: refactor logging in infer storage pass URL: https://github.com/apache/incubator-mxnet/pull/9464#discussion_r162443617 ## File path: src/operator/operator_common.h ## @@ -350,11 +350,12 @@ inline bool

[GitHub] sxjscience commented on issue #8337: mx.autograd.grad works or fails depending on use of slices

sxjscience commented on issue #8337: mx.autograd.grad works or fails depending on use of slices URL: https://github.com/apache/incubator-mxnet/issues/8337#issuecomment-358747950 @leezu I find that `dist[0, 0]` will always be 0 and the `sqrt` function is non-differentiable at 0. The

[GitHub] larroy opened a new pull request #9480: Prep script and Fixes for single shot detector (SSD)

larroy opened a new pull request #9480: Prep script and Fixes for single shot detector (SSD) URL: https://github.com/apache/incubator-mxnet/pull/9480 ## Description ## Fix multibox detector kernel resource limit on launch on volta using __launch_bounds__ ## Checklist ## ###

[GitHub] larroy commented on issue #9480: Prep script and Fixes for single shot detector (SSD)

larroy commented on issue #9480: Prep script and Fixes for single shot detector (SSD) URL: https://github.com/apache/incubator-mxnet/pull/9480#issuecomment-358753666 @KellenSunderland This is an automated message from the

[GitHub] Bestehorn opened a new issue #9478: Loss becomes NAN with larger, more layered network

Bestehorn opened a new issue #9478: Loss becomes NAN with larger, more layered network URL: https://github.com/apache/incubator-mxnet/issues/9478 ## Description When the number of layers/nodes in gluon.Sequential structures increases, the SoftMax cost function will return NaN values

[GitHub] sxjscience commented on issue #8337: mx.autograd.grad works or fails depending on use of slices

sxjscience commented on issue #8337: mx.autograd.grad works or fails depending on use of slices URL: https://github.com/apache/incubator-mxnet/issues/8337#issuecomment-358747950 @leezu I find that `dist[0, 0]` will always be 0 and the `sqrt` function is non-differentiable at 0. The

[GitHub] eric-haibin-lin opened a new pull request #9479: add norm operator for sparse ndarray

eric-haibin-lin opened a new pull request #9479: add norm operator for sparse ndarray URL: https://github.com/apache/incubator-mxnet/pull/9479 ## Description ## support sparse.norm() operator https://discuss.mxnet.io/t/norm-of-sparse-matrix/460/3 @szha @ZiyueHuang ##

[GitHub] sxjscience commented on issue #9478: Loss becomes NAN with larger, more layered network

sxjscience commented on issue #9478: Loss becomes NAN with larger, more layered network URL: https://github.com/apache/incubator-mxnet/issues/9478#issuecomment-358750415 The initializer may be fine for a small network, but the output of the layers may become larger as the depth increases

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

zheng-da commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162432577 ## File path: src/operator/nn/pooling.cc ## @@ -32,67 +33,304 @@ #if MXNET_USE_NNPACK == 1 #include

[GitHub] szha commented on issue #9295: test_operator.test_laop_3 hangs

szha commented on issue #9295: test_operator.test_laop_3 hangs URL: https://github.com/apache/incubator-mxnet/issues/9295#issuecomment-358747194 MKLML is a dependency of the MKL2017 option and was updated recently in #9218

[GitHub] szha commented on issue #9295: test_operator.test_laop_3 hangs

szha commented on issue #9295: test_operator.test_laop_3 hangs URL: https://github.com/apache/incubator-mxnet/issues/9295#issuecomment-358747194 MKLML is a dependency of the MKL2017 option and was updated recently in #9218. Might be related. It would be easy to verify if you revert to the

[GitHub] reminisce commented on issue #9455: An error occurred while calculating the square of ndarray by using gpu context

reminisce commented on issue #9455: An error occurred while calculating the square of ndarray by using gpu context URL: https://github.com/apache/incubator-mxnet/issues/9455#issuecomment-358738026 Since running operators in Python is async with backend, you need to wait until the

[GitHub] sxjscience commented on issue #9478: Loss becomes NAN with larger, more layered network

sxjscience commented on issue #9478: Loss becomes NAN with larger, more layered network URL: https://github.com/apache/incubator-mxnet/issues/9478#issuecomment-358741994 Would you also show the initializer and optimizer used to train the network? A good initialization is also essential

[GitHub] Bestehorn commented on issue #9478: Loss becomes NAN with larger, more layered network

Bestehorn commented on issue #9478: Loss becomes NAN with larger, more layered network URL: https://github.com/apache/incubator-mxnet/issues/9478#issuecomment-358745629 Initialization and other stuff has not been changed compared to the tutorial on MXNET

[GitHub] Bestehorn commented on issue #9478: Loss becomes NAN with larger, more layered network

Bestehorn commented on issue #9478: Loss becomes NAN with larger, more layered network URL: https://github.com/apache/incubator-mxnet/issues/9478#issuecomment-358745629 Initialization and other stuff has not been changed compared to the tutorial on MXNET

[GitHub] eric-haibin-lin commented on issue #9380: nd.stack causes abort for differently sized arrays

eric-haibin-lin commented on issue #9380: nd.stack causes abort for differently sized arrays URL: https://github.com/apache/incubator-mxnet/issues/9380#issuecomment-358745656 @piiswrong in this case is the expected result with shape (10,3)?

[GitHub] eric-haibin-lin commented on issue #9464: refactor logging in infer storage pass

eric-haibin-lin commented on issue #9464: refactor logging in infer storage pass URL: https://github.com/apache/incubator-mxnet/pull/9464#issuecomment-358746155 Updated based on reviews This is an automated message from the

[GitHub] cjolivier01 commented on a change in pull request #9479: add norm operator for sparse ndarray

cjolivier01 commented on a change in pull request #9479: add norm operator for sparse ndarray URL: https://github.com/apache/incubator-mxnet/pull/9479#discussion_r162447097 ## File path: src/operator/tensor/broadcast_reduce_op.h ## @@ -821,20 +821,93 @@ struct ReduceGrad

[GitHub] cjolivier01 commented on a change in pull request #9479: add norm operator for sparse ndarray

cjolivier01 commented on a change in pull request #9479: add norm operator for sparse ndarray URL: https://github.com/apache/incubator-mxnet/pull/9479#discussion_r162447361 ## File path: src/operator/tensor/broadcast_reduce_op.h ## @@ -821,20 +821,93 @@ struct ReduceGrad

[GitHub] larroy commented on issue #9397: [WIP] Remove incorrect link flags when using OPENCV. Exclude binaries from lint.

larroy commented on issue #9397: [WIP] Remove incorrect link flags when using OPENCV. Exclude binaries from lint. URL: https://github.com/apache/incubator-mxnet/pull/9397#issuecomment-358777993 @cjolivier01 can we merge this?

[GitHub] aaronmarkham commented on a change in pull request #9353: added mac gpu install; refactored windows install

aaronmarkham commented on a change in pull request #9353: added mac gpu install; refactored windows install URL: https://github.com/apache/incubator-mxnet/pull/9353#discussion_r162447765 ## File path: docs/install/index.md ## @@ -709,28 +732,32 @@ Alternatively, you may

[GitHub] aidan-plenert-macdonald commented on issue #9380: nd.stack causes abort for differently sized arrays

aidan-plenert-macdonald commented on issue #9380: nd.stack causes abort for differently sized arrays URL: https://github.com/apache/incubator-mxnet/issues/9380#issuecomment-358758416 @eric-haibin-lin So this wasn't so much about it doing the right thing as simply not killing Python. This

[GitHub] Bestehorn commented on issue #9478: Loss becomes NAN with larger, more layered network

Bestehorn commented on issue #9478: Loss becomes NAN with larger, more layered network URL: https://github.com/apache/incubator-mxnet/issues/9478#issuecomment-358764800 First of all: Thanks for your prompt responses and effort to look into this. While changing sigma to 0.01 has the

[GitHub] aidan-plenert-macdonald commented on issue #9380: nd.stack causes abort for differently sized arrays

aidan-plenert-macdonald commented on issue #9380: nd.stack causes abort for differently sized arrays URL: https://github.com/apache/incubator-mxnet/issues/9380#issuecomment-358758416 @eric-haibin-lin So this wasn't so much about it doing the right thing as simply not killing Python. This

[GitHub] eric-haibin-lin commented on issue #9380: nd.stack causes abort for differently sized arrays

eric-haibin-lin commented on issue #9380: nd.stack causes abort for differently sized arrays URL: https://github.com/apache/incubator-mxnet/issues/9380#issuecomment-358759295 I see your point. @anirudh2290 is actively working on better support for exception handling in MXNet. He will

[GitHub] sxjscience commented on issue #9478: Loss becomes NAN with larger, more layered network

sxjscience commented on issue #9478: Loss becomes NAN with larger, more layered network URL: https://github.com/apache/incubator-mxnet/issues/9478#issuecomment-358773759 @Bestehorn I think it's related to the optimizer. You can try to use different learning rate/optimization methods to

[GitHub] reminisce commented on issue #9452: A new deep learning visualization tool from baidu support MXNet!

reminisce commented on issue #9452: A new deep learning visualization tool from baidu support MXNet! URL: https://github.com/apache/incubator-mxnet/issues/9452#issuecomment-358775200 @chinakook It looks easy-to-use, but still lacks many features supported in Tensorboard. Do you know how

[GitHub] eric-haibin-lin commented on a change in pull request #9479: add norm operator for sparse ndarray

eric-haibin-lin commented on a change in pull request #9479: add norm operator for sparse ndarray URL: https://github.com/apache/incubator-mxnet/pull/9479#discussion_r162449862 ## File path: src/operator/tensor/broadcast_reduce_op.h ## @@ -821,20 +821,93 @@ struct

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

zheng-da commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162501336 ## File path: src/operator/tensor/elemwise_unary_op_basic.cc ## @@ -108,12 +109,66 @@

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

zheng-da commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162502091 ## File path: src/operator/nn/mkldnn/mkldnn_sum.cc ## @@ -0,0 +1,75 @@ +/* + * Licensed to the Apache Software

[GitHub] cjolivier01 commented on issue #9397: [WIP] Remove incorrect link flags when using OPENCV. Exclude binaries from lint.

cjolivier01 commented on issue #9397: [WIP] Remove incorrect link flags when using OPENCV. Exclude binaries from lint. URL: https://github.com/apache/incubator-mxnet/pull/9397#issuecomment-358818333 What is the "incorrect link flag"?

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

zheng-da commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162502678 ## File path: src/ndarray/ndarray.cc ## @@ -181,6 +322,408 @@ void NDArray::set_fresh_out_grad(bool state)

[GitHub] cjolivier01 commented on a change in pull request #9479: add norm operator for sparse ndarray

cjolivier01 commented on a change in pull request #9479: add norm operator for sparse ndarray URL: https://github.com/apache/incubator-mxnet/pull/9479#discussion_r162502860 ## File path: src/operator/tensor/broadcast_reduce_op.h ## @@ -821,20 +821,93 @@ struct ReduceGrad

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

TaoLv commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162370894 ## File path: src/operator/nn/pooling.cc ## @@ -32,67 +33,304 @@ #if MXNET_USE_NNPACK == 1 #include

[GitHub] eric-haibin-lin opened a new pull request #9481: add where op with sparse condition

eric-haibin-lin opened a new pull request #9481: add where op with sparse condition URL: https://github.com/apache/incubator-mxnet/pull/9481 ## Description ## support sparse.where with condition being csr ndarray #9464 need to be merged first @reminisce ## Checklist ##

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

CodingCat commented on issue #8972: Profiling enhancements, python API, vtune and chrome tracing objects, etc. URL: https://github.com/apache/incubator-mxnet/pull/8972#issuecomment-358800756 LGTM for the changes, but is there a good way to track the breaking changes in the current release

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162480809 ## File path: src/executor/graph_executor.cc ## @@ -828,7 +828,7 @@ void

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162485031 ## File path: src/operator/nn/mkldnn/mkldnn_sum.cc ## @@ -0,0 +1,75 @@ +/* + * Licensed to the Apache

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162484866 ## File path: src/operator/nn/mkldnn/mkldnn_sum.cc ## @@ -0,0 +1,75 @@ +/* + * Licensed to the Apache

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162484441 ## File path: src/operator/nn/mkldnn/mkldnn_sum.cc ## @@ -0,0 +1,75 @@ +/* + * Licensed to the Apache

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162484098 ## File path: src/operator/nn/mkldnn/mkldnn_sum.cc ## @@ -0,0 +1,75 @@ +/* + * Licensed to the Apache

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162482353 ## File path: src/ndarray/ndarray.cc ## @@ -181,6 +322,408 @@ void NDArray::set_fresh_out_grad(bool

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162478606 ## File path: include/mxnet/ndarray.h ## @@ -611,6 +562,83 @@ class NDArray { <<

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162482725 ## File path: src/ndarray/ndarray.cc ## @@ -181,6 +322,408 @@ void NDArray::set_fresh_out_grad(bool

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162486304 ## File path: src/operator/tensor/elemwise_unary_op_basic.cc ## @@ -108,12 +109,66 @@

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162478052 ## File path: include/mxnet/ndarray.h ## @@ -184,11 +135,17 @@ class NDArray { const TBlob ,

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162487388 ## File path: src/operator/tensor/elemwise_unary_op_basic.cc ## @@ -108,12 +109,66 @@

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162479632 ## File path: prepare_mkldnn.sh ## @@ -0,0 +1,118 @@ +#!/bin/bash + +# Licensed to the Apache Software

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162486857 ## File path: src/operator/tensor/matrix_op.cc ## @@ -180,6 +182,51 @@ If the argument `reverse` is set

[GitHub] eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN

eric-haibin-lin commented on a change in pull request #8302: Refactor operators & MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r162478815 ## File path: include/mxnet/ndarray.h ## @@ -611,6 +562,83 @@ class NDArray { <<

[GitHub] JulianSlzr commented on issue #9353: added mac gpu install; refactored windows install

JulianSlzr commented on issue #9353: added mac gpu install; refactored windows install URL: https://github.com/apache/incubator-mxnet/pull/9353#issuecomment-358789039 Thanks for the changes, looks good to me. This is an

[incubator-mxnet] branch master updated: added mac gpu install; refactored windows install (#9353)

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 55ce97c added mac gpu install;

[GitHub] eric-haibin-lin closed pull request #9353: added mac gpu install; refactored windows install

eric-haibin-lin closed pull request #9353: added mac gpu install; refactored windows install URL: https://github.com/apache/incubator-mxnet/pull/9353 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] eric-haibin-lin closed issue #9146: Install instructions - selector broken

eric-haibin-lin closed issue #9146: Install instructions - selector broken URL: https://github.com/apache/incubator-mxnet/issues/9146 This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] larroy commented on issue #9480: Prep script and Fixes for single shot detector (SSD)

larroy commented on issue #9480: Prep script and Fixes for single shot detector (SSD) URL: https://github.com/apache/incubator-mxnet/pull/9480#issuecomment-358793239 @eric-haibin-lin This is an automated message from the

[GitHub] mbaijal opened a new pull request #9482: [WIP][ReviewRequired] More License Fixes based on Apache RAT Feedback

mbaijal opened a new pull request #9482: [WIP][ReviewRequired] More License Fixes based on Apache RAT Feedback URL: https://github.com/apache/incubator-mxnet/pull/9482 ## Fixing Apache RAT failures - Part 2 ## PLEASE HELP REVIEW! @bhavinthaker @gautamkmr Fixing Licenses based on

[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-358801223 I am not aware of any. This is

[GitHub] gramce opened a new issue #9491: is_train in model.backward will change the bias parameter?

gramce opened a new issue #9491: is_train in model.backward will change the bias parameter? URL: https://github.com/apache/incubator-mxnet/issues/9491 I ran the example adversary_generation. Print the parameter bias before and after the model.backward([out_grad]) and find that the

[GitHub] szha commented on issue #9412: Flaky Tests Tracking Issue

szha commented on issue #9412: Flaky Tests Tracking Issue URL: https://github.com/apache/incubator-mxnet/issues/9412#issuecomment-358878849 Working on #8283 This is an automated message from the Apache Git Service. To

[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-358885449 FYI, we have tested the **inference** accuracy using **pre-trained models** and compared with official and MKLML MXNET. From the

[GitHub] szha commented on issue #8283: Caffe integration test fails because it is not able to download files from caffe server

szha commented on issue #8283: Caffe integration test fails because it is not able to download files from caffe server URL: https://github.com/apache/incubator-mxnet/issues/8283#issuecomment-358879007 Reviewed the license, and it doesn't pose any restriction on use. I will move its

[GitHub] chinakook opened a new pull request #9492: fix print_summary bug and add groups of convolution

chinakook opened a new pull request #9492: fix print_summary bug and add groups of convolution URL: https://github.com/apache/incubator-mxnet/pull/9492 ## Description ## 1. fix "int(node["attrs"]["no_bias"])" bug 2. add groups of convolution param calculation ## Checklist ##

[GitHub] reddysree007 commented on issue #9094: Cannot continue with MXNet Tutorial: Real-time Object Detection with MXNet On The Raspberry Pi

reddysree007 commented on issue #9094: Cannot continue with MXNet Tutorial: Real-time Object Detection with MXNet On The Raspberry Pi URL: https://github.com/apache/incubator-mxnet/issues/9094#issuecomment-358887980 @yimingliu0216 Could you tell me how you downgraded version?

[GitHub] gramce commented on issue #9491: is_train in model.backward will change the bias parameter?

gramce commented on issue #9491: is_train in model.backward will change the bias parameter? URL: https://github.com/apache/incubator-mxnet/issues/9491#issuecomment-358876599 a mistake This is an automated message from the

[GitHub] reddysree007 commented on issue #9094: Cannot continue with MXNet Tutorial: Real-time Object Detection with MXNet On The Raspberry Pi

reddysree007 commented on issue #9094: Cannot continue with MXNet Tutorial: Real-time Object Detection with MXNet On The Raspberry Pi URL: https://github.com/apache/incubator-mxnet/issues/9094#issuecomment-358877955 Can you @wildhunter-66 tell me how to downgrade the mxnet version form

[GitHub] reddysree007 commented on issue #9094: Cannot continue with MXNet Tutorial: Real-time Object Detection with MXNet On The Raspberry Pi

reddysree007 commented on issue #9094: Cannot continue with MXNet Tutorial: Real-time Object Detection with MXNet On The Raspberry Pi URL: https://github.com/apache/incubator-mxnet/issues/9094#issuecomment-358877955 Can anyone tell me how to downgrade the mxnet version form 1.0.1 to

[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-358885449 FYI, we have tested the inference accuracy using **pre-trained models** and compared with official and MKLML MXNET. From the below

[GitHub] safrooze commented on issue #9213: Crash when MXNet API called before spawning multiprocess

safrooze commented on issue #9213: Crash when MXNet API called before spawning multiprocess URL: https://github.com/apache/incubator-mxnet/issues/9213#issuecomment-358880961 I observed a very similar crash on a p3.2x instance with gluon.data.DataLoader() when num_workers was set to

[GitHub] zhangguotai opened a new issue #9476: libmxnet.so: undefined symbol: dorgqr_?????????????????

zhangguotai opened a new issue #9476: libmxnet.so: undefined symbol: dorgqr_? URL: https://github.com/apache/incubator-mxnet/issues/9476 I have compiled mxnet successfully,why this fault happened? This is an

[GitHub] nicklhy closed issue #9354: resize image in GPU for better performance?

nicklhy closed issue #9354: resize image in GPU for better performance? URL: https://github.com/apache/incubator-mxnet/issues/9354 This is an automated message from the Apache Git Service. To respond to the message, please

[GitHub] TaoLv commented on issue #9422: (Please don't merge) Debug BatchNorm with MKLDNN

TaoLv commented on issue #9422: (Please don't merge) Debug BatchNorm with MKLDNN URL: https://github.com/apache/incubator-mxnet/pull/9422#issuecomment-358595311 @eric-haibin-lin @marcoabreu Many thanks. Seems the problem is root caused and fixed by Da. I will close this PR now.

[GitHub] nicklhy commented on issue #9354: resize image in GPU for better performance?

nicklhy commented on issue #9354: resize image in GPU for better performance? URL: https://github.com/apache/incubator-mxnet/issues/9354#issuecomment-358592861 Thank you. BilinearSampler is totally what I am looking for!

[GitHub] xuzhenqi opened a new issue #9475: OpenCV Error: Assertion failed (dst.cols < SHRT_MAX && dst.rows < SHRT_MAX && src.cols < SHRT_MAX && src.rows < SHRT_MAX) in remap, file /home/travis/build/

xuzhenqi opened a new issue #9475: OpenCV Error: Assertion failed (dst.cols < SHRT_MAX && dst.rows < SHRT_MAX && src.cols < SHRT_MAX && src.rows < SHRT_MAX) in remap, file /home/travis/build/dmlc/mxnet-distro/deps/opencv-3.3.0/modules/imgproc/src/imgwarp.cpp, line 4944 terminate called after

[GitHub] ilibx opened a new issue #9489: terminate called after throwing an instance of 'dmlc::Error'

ilibx opened a new issue #9489: terminate called after throwing an instance of 'dmlc::Error' URL: https://github.com/apache/incubator-mxnet/issues/9489 Clone code from this repo, build with USE_DIST_KVSTORE=1 in ubuntu:16.04 container, instance multi container, then run cmd bellow:

[GitHub] mbaijal commented on issue #9472: [Review Required] Adding more file formats to license_header.py (.m, .mk, .R, .cfg)

mbaijal commented on issue #9472: [Review Required] Adding more file formats to license_header.py (.m, .mk, .R, .cfg) URL: https://github.com/apache/incubator-mxnet/pull/9472#issuecomment-358865827 @marcoabreu I agree with Bhavin. Since the license_header.py script is performing this

[GitHub] pabramowitsch opened a new issue #9490: Unable to install mxnet on MacOS .. no version seemingly compatible with Python3

pabramowitsch opened a new issue #9490: Unable to install mxnet on MacOS .. no version seemingly compatible with Python3 URL: https://github.com/apache/incubator-mxnet/issues/9490 I just want to play with mxnet using prebuilt code on MacOS but the pip install fails. Following the

[GitHub] szha commented on issue #9490: Unable to install mxnet on MacOS .. no version seemingly compatible with Python3

szha commented on issue #9490: Unable to install mxnet on MacOS .. no version seemingly compatible with Python3 URL: https://github.com/apache/incubator-mxnet/issues/9490#issuecomment-358866593 Do you need to use 0.11.0 specifically, or would you be fine with trying more recent versions?

[GitHub] reminisce commented on a change in pull request #9481: add where op with sparse condition

reminisce commented on a change in pull request #9481: add where op with sparse condition URL: https://github.com/apache/incubator-mxnet/pull/9481#discussion_r162537564 ## File path: src/operator/tensor/control_flow_op.h ## @@ -185,6 +303,62 @@ void WhereOpForward(const

[GitHub] reminisce commented on a change in pull request #9481: add where op with sparse condition

reminisce commented on a change in pull request #9481: add where op with sparse condition URL: https://github.com/apache/incubator-mxnet/pull/9481#discussion_r162537283 ## File path: src/operator/tensor/control_flow_op.h ## @@ -51,6 +52,35 @@ struct where { } };

  1   2   >