[GitHub] ZhennanQin commented on issue #14053: in-place reshape ops

ZhennanQin commented on issue #14053: in-place reshape ops URL: https://github.com/apache/incubator-mxnet/pull/14053#issuecomment-459949587 Have you tried this with mkldnn enabled? E.g reshape the output of mkldnn convolution?

[GitHub] arcadiaphy opened a new issue #14057: validation stucks when training gluoncv ssd model

arcadiaphy opened a new issue #14057: validation stucks when training gluoncv ssd model URL: https://github.com/apache/incubator-mxnet/issues/14057 ## Description When training gluoncv ssd model, validation sometimes takes way more longer time than the training epoch. After debugging,

[GitHub] arcadiaphy opened a new pull request #14058: add backgroud class in box_nms

arcadiaphy opened a new pull request #14058: add backgroud class in box_nms URL: https://github.com/apache/incubator-mxnet/pull/14058 ## Description ## This PR is mentioned in #14057 What I have done in box_nms operator: 1. add background_id argument 2. filter out

[GitHub] bputrycz commented on issue #10715: Wrong params file is not reported as error in C API

bputrycz commented on issue #10715: Wrong params file is not reported as error in C API URL: https://github.com/apache/incubator-mxnet/issues/10715#issuecomment-459955392 I wasn't blocked by this issue, just noticed it and reported. I haven't invested more time into it. I haven't

[GitHub] rongzha1 commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU

rongzha1 commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU URL: https://github.com/apache/incubator-mxnet/pull/14056#discussion_r253264609 ## File path: src/operator/channel_op_common.h ## @@ -101,6 +101,42 @@ void

[GitHub] rongzha1 commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU

rongzha1 commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU URL: https://github.com/apache/incubator-mxnet/pull/14056#discussion_r253264597 ## File path: src/operator/channel_op_common.h ## @@ -101,6 +101,42 @@ void

[GitHub] renganxu edited a comment on issue #14047: mxnet.base.MXNetError: Cannot find argument 'cudnn_algo_verbose'

renganxu edited a comment on issue #14047: mxnet.base.MXNetError: Cannot find argument 'cudnn_algo_verbose' URL: https://github.com/apache/incubator-mxnet/issues/14047#issuecomment-459983879 @ptrendx I know NGC container works, but I just want to run without container. Now I figured out

[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 ba27f54 Bump the publish

[GitHub] gigasquid commented on a change in pull request #13993: [Clojure] Add resource scope to clojure package

gigasquid commented on a change in pull request #13993: [Clojure] Add resource scope to clojure package URL: https://github.com/apache/incubator-mxnet/pull/13993#discussion_r253269610 ## File path: contrib/clojure-package/test/org/apache/clojure_mxnet/resource_scope_test.clj

[GitHub] kedarbellare commented on issue #13993: [Clojure] Add resource scope to clojure package

kedarbellare commented on issue #13993: [Clojure] Add resource scope to clojure package URL: https://github.com/apache/incubator-mxnet/pull/13993#issuecomment-459977844 can't wait to use this  This is an automated message

[GitHub] kedarbellare commented on a change in pull request #13993: [Clojure] Add resource scope to clojure package

kedarbellare commented on a change in pull request #13993: [Clojure] Add resource scope to clojure package URL: https://github.com/apache/incubator-mxnet/pull/13993#discussion_r253270291 ## File path: contrib/clojure-package/test/org/apache/clojure_mxnet/resource_scope_test.clj

[GitHub] ZhennanQin commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU

ZhennanQin commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU URL: https://github.com/apache/incubator-mxnet/pull/14056#discussion_r253262164 ## File path: src/operator/channel_op_common.h ## @@ -101,6 +101,42 @@ void

[GitHub] ZhennanQin commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU

ZhennanQin commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU URL: https://github.com/apache/incubator-mxnet/pull/14056#discussion_r253262215 ## File path: src/operator/channel_op_common.h ## @@ -101,6 +101,42 @@ void

[GitHub] ZhennanQin commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU

ZhennanQin commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU URL: https://github.com/apache/incubator-mxnet/pull/14056#discussion_r253262126 ## File path: src/operator/channel_op_common.h ## @@ -101,6 +101,42 @@ void

[GitHub] kedarbellare commented on a change in pull request #13993: [Clojure] Add resource scope to clojure package

kedarbellare commented on a change in pull request #13993: [Clojure] Add resource scope to clojure package URL: https://github.com/apache/incubator-mxnet/pull/13993#discussion_r253268961 ## File path:

[GitHub] gigasquid commented on a change in pull request #13993: [Clojure] Add resource scope to clojure package

gigasquid commented on a change in pull request #13993: [Clojure] Add resource scope to clojure package URL: https://github.com/apache/incubator-mxnet/pull/13993#discussion_r253270361 ## File path: contrib/clojure-package/test/org/apache/clojure_mxnet/resource_scope_test.clj

[GitHub] szha commented on issue #14053: in-place reshape ops

szha commented on issue #14053: in-place reshape ops URL: https://github.com/apache/incubator-mxnet/pull/14053#issuecomment-459990251 @ZhennanQin I didn't test that specifically based on the assumption that the backend implementation should not cause existing APIs such as `NDArray.reshape`

[GitHub] mahmoodn closed issue #14002: error: ‘__cpuid’ was not declared in this scope

mahmoodn closed issue #14002: error: ‘__cpuid’ was not declared in this scope URL: https://github.com/apache/incubator-mxnet/issues/14002 This is an automated message from the Apache Git Service. To respond to the message,

[GitHub] larroy closed pull request #13834: Reduce maven verbosity which is filling up the logs of the builds

larroy closed pull request #13834: Reduce maven verbosity which is filling up the logs of the builds URL: https://github.com/apache/incubator-mxnet/pull/13834 This is an automated message from the Apache Git Service. To

[GitHub] TaoLv commented on issue #13749: Add NHWC layout support to Pooling (cpu, gpu cuda, gpu cuDNN)

TaoLv commented on issue #13749: Add NHWC layout support to Pooling (cpu, gpu cuda, gpu cuDNN) URL: https://github.com/apache/incubator-mxnet/pull/13749#issuecomment-459970152 Maybe it's out of the scope of this PR, but I would suggest to remove layout parameters from all operator APIs in

[GitHub] gigasquid commented on a change in pull request #13993: [Clojure] Add resource scope to clojure package

gigasquid commented on a change in pull request #13993: [Clojure] Add resource scope to clojure package URL: https://github.com/apache/incubator-mxnet/pull/13993#discussion_r253269849 ## File path:

[GitHub] mahmoodn opened a new issue #14059: ModuleNotFoundError: No module named 'mxnet'

mahmoodn opened a new issue #14059: ModuleNotFoundError: No module named 'mxnet' URL: https://github.com/apache/incubator-mxnet/issues/14059 Hi, Although I have built mxnet with `make -j8 USE_OPENCV=1 USE_BLAS=openblas USE_CUDA=1 USE_CUDA_PATH=/usr/local/cuda USE_CUDNN=1`, I am not able

[GitHub] mxnet-label-bot commented on issue #14059: ModuleNotFoundError: No module named 'mxnet'

mxnet-label-bot commented on issue #14059: ModuleNotFoundError: No module named 'mxnet' URL: https://github.com/apache/incubator-mxnet/issues/14059#issuecomment-459979035 Hey, this is the MXNet Label Bot. Thank you for submitting the issue! I will try and suggest some labels so that

[GitHub] renganxu commented on issue #14047: mxnet.base.MXNetError: Cannot find argument 'cudnn_algo_verbose'

renganxu commented on issue #14047: mxnet.base.MXNetError: Cannot find argument 'cudnn_algo_verbose' URL: https://github.com/apache/incubator-mxnet/issues/14047#issuecomment-459983879 @ptrendx I know NGC container works, but I just want to run without container. Now I figured out NGC

[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 0755472 Bump the publish

[GitHub] TaoLv commented on a change in pull request #13749: Add NHWC layout support to Pooling (cpu, gpu cuda, gpu cuDNN)

TaoLv commented on a change in pull request #13749: Add NHWC layout support to Pooling (cpu, gpu cuda, gpu cuDNN) URL: https://github.com/apache/incubator-mxnet/pull/13749#discussion_r253266821 ## File path: src/operator/nn/pooling.cc ## @@ -421,11 +463,16 @@

[GitHub] TaoLv commented on a change in pull request #13749: Add NHWC layout support to Pooling (cpu, gpu cuda, gpu cuDNN)

TaoLv commented on a change in pull request #13749: Add NHWC layout support to Pooling (cpu, gpu cuda, gpu cuDNN) URL: https://github.com/apache/incubator-mxnet/pull/13749#discussion_r253267148 ## File path: src/operator/nn/mkldnn/mkldnn_pooling-inl.h ## @@ -104,7 +104,8

[GitHub] TaoLv commented on a change in pull request #13749: Add NHWC layout support to Pooling (cpu, gpu cuda, gpu cuDNN)

TaoLv commented on a change in pull request #13749: Add NHWC layout support to Pooling (cpu, gpu cuda, gpu cuDNN) URL: https://github.com/apache/incubator-mxnet/pull/13749#discussion_r253266876 ## File path: tests/python/gpu/test_operator_gpu.py ## @@ -608,6 +608,72 @@

[GitHub] kedarbellare commented on a change in pull request #13993: [Clojure] Add resource scope to clojure package

kedarbellare commented on a change in pull request #13993: [Clojure] Add resource scope to clojure package URL: https://github.com/apache/incubator-mxnet/pull/13993#discussion_r253268889 ## File path: contrib/clojure-package/test/org/apache/clojure_mxnet/resource_scope_test.clj

[GitHub] mahmoodn commented on issue #14002: error: ‘__cpuid’ was not declared in this scope

mahmoodn commented on issue #14002: error: ‘__cpuid’ was not declared in this scope URL: https://github.com/apache/incubator-mxnet/issues/14002#issuecomment-459975490 It seems that I was able to compile the latest git version. I will close the issue. The last thing in the build output is

[GitHub] rajeshii opened a new pull request #14060: Exclude concat layer for gpu quantization

rajeshii opened a new pull request #14060: Exclude concat layer for gpu quantization URL: https://github.com/apache/incubator-mxnet/pull/14060 ## Description ## Exclude concat layer for gpu quantization since #13297 enable quantizaed_concat op for cpu. below is the error log before

[GitHub] KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API

KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API URL: https://github.com/apache/incubator-mxnet/pull/14040#discussion_r253288021 ## File path: include/mxnet/c_api.h ## @@ -1835,12 +1835,6 @@ MXNET_DLL int

[GitHub] KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API

KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API URL: https://github.com/apache/incubator-mxnet/pull/14040#discussion_r253288021 ## File path: include/mxnet/c_api.h ## @@ -1835,12 +1835,6 @@ MXNET_DLL int

[GitHub] KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API

KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API URL: https://github.com/apache/incubator-mxnet/pull/14040#discussion_r253288413 ## File path: src/operator/subgraph/tensorrt/tensorrt-inl.h ## @@ -0,0 +1,217 @@

[GitHub] KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API

KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API URL: https://github.com/apache/incubator-mxnet/pull/14040#discussion_r253288427 ## File path: src/operator/subgraph/tensorrt/tensorrt-inl.h ## @@ -0,0 +1,217 @@

[GitHub] KellenSunderland commented on issue #14040: Reformat of TensorRT to use subgraph API

KellenSunderland commented on issue #14040: Reformat of TensorRT to use subgraph API URL: https://github.com/apache/incubator-mxnet/pull/14040#issuecomment-460027375 ONNX-Tensorrt might fail to build as it adds a new default target that requires a header not in a standard search path.

[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 1c2245b Bump the publish

[GitHub] KellenSunderland commented on issue #13906: [MXNET-703] Update onnx-tensorrt for int8/fp16 support

KellenSunderland commented on issue #13906: [MXNET-703] Update onnx-tensorrt for int8/fp16 support URL: https://github.com/apache/incubator-mxnet/pull/13906#issuecomment-460020295 18 days, no review, closing. This is an

[GitHub] KellenSunderland closed pull request #13906: [MXNET-703] Update onnx-tensorrt for int8/fp16 support

KellenSunderland closed pull request #13906: [MXNET-703] Update onnx-tensorrt for int8/fp16 support URL: https://github.com/apache/incubator-mxnet/pull/13906 This is an automated message from the Apache Git Service. To

[GitHub] Caenorst commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API

Caenorst commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API URL: https://github.com/apache/incubator-mxnet/pull/14040#discussion_r253287088 ## File path: include/mxnet/c_api.h ## @@ -1835,12 +1835,6 @@ MXNET_DLL int

[GitHub] KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API

KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API URL: https://github.com/apache/incubator-mxnet/pull/14040#discussion_r253288158 ## File path: src/operator/subgraph/tensorrt/nnvm_to_onnx.cc ## @@ -411,6 +383,49 @@

[GitHub] KellenSunderland removed a comment on issue #13906: [MXNET-703] Update onnx-tensorrt for int8/fp16 support

KellenSunderland removed a comment on issue #13906: [MXNET-703] Update onnx-tensorrt for int8/fp16 support URL: https://github.com/apache/incubator-mxnet/pull/13906#issuecomment-460020295 18 days, no review, closing. This is

[GitHub] KellenSunderland opened a new pull request #13906: [MXNET-703] Update onnx-tensorrt for int8/fp16 support

KellenSunderland opened a new pull request #13906: [MXNET-703] Update onnx-tensorrt for int8/fp16 support URL: https://github.com/apache/incubator-mxnet/pull/13906 ## Description ## Update onnx-tensorrt with new support for different data types, including (importantly for v100) fp16.

[GitHub] zhreshold commented on a change in pull request #14031: Fix transposed convolution in CPU w/o MKLDNN.

zhreshold commented on a change in pull request #14031: Fix transposed convolution in CPU w/o MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/14031#discussion_r253289470 ## File path: tests/python/unittest/test_gluon.py ## @@ -503,6 +503,40 @@ def

[GitHub] KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API

KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API URL: https://github.com/apache/incubator-mxnet/pull/14040#discussion_r253288274 ## File path: src/operator/subgraph/tensorrt/onnx_to_tensorrt.cc ## @@ -127,16 +128,15

[GitHub] KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API

KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API URL: https://github.com/apache/incubator-mxnet/pull/14040#discussion_r253288287 ## File path: src/operator/subgraph/tensorrt/onnx_to_tensorrt.cc ## @@ -127,16 +128,15

[GitHub] KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API

KellenSunderland commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API URL: https://github.com/apache/incubator-mxnet/pull/14040#discussion_r253288751 ## File path: src/operator/subgraph/tensorrt/tensorrt-inl.h ## @@ -0,0 +1,217 @@

[GitHub] ZhennanQin commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU

ZhennanQin commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU URL: https://github.com/apache/incubator-mxnet/pull/14056#discussion_r253262094 ## File path: src/operator/channel_op_common.h ## @@ -101,6 +101,42 @@ void

[GitHub] ZhennanQin commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU

ZhennanQin commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU URL: https://github.com/apache/incubator-mxnet/pull/14056#discussion_r253262059 ## File path: src/operator/channel_op_common.h ## @@ -101,6 +101,42 @@ void

[GitHub] ZhennanQin commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU

ZhennanQin commented on a change in pull request #14056: A better split-2D(SliceChannel) op forward kernel for CPU URL: https://github.com/apache/incubator-mxnet/pull/14056#discussion_r253262122 ## File path: src/operator/channel_op_common.h ## @@ -101,6 +101,42 @@ void

[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 67dc46d Bump the publish

[GitHub] stephenrawls commented on issue #13680: [MXNET-1121] Example to demonstrate the inference workflow using RNN

stephenrawls commented on issue #13680: [MXNET-1121] Example to demonstrate the inference workflow using RNN URL: https://github.com/apache/incubator-mxnet/pull/13680#issuecomment-46730 Looks good to me, thanks for the example. Made a few small comments in the code about sharing

[GitHub] stephenrawls commented on issue #14053: in-place reshape ops

stephenrawls commented on issue #14053: in-place reshape ops URL: https://github.com/apache/incubator-mxnet/pull/14053#issuecomment-460005020 @szha -- Thanks for putting in this patch! I have a couple questions about the operator `expand_dims`, which I understand you are actually

[GitHub] pengzhao-intel commented on issue #13697: [MKLDNN] Enable signed int8 support for convolution.

pengzhao-intel commented on issue #13697: [MKLDNN] Enable signed int8 support for convolution. URL: https://github.com/apache/incubator-mxnet/pull/13697#issuecomment-460013352 @xinyu-intel could you help verify the GPU accuracy with this PR?

[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 c4de8b2 Bump the publish

[GitHub] pengzhao-intel commented on issue #13668: Update MKL-DNN to fix the Dense layer issue

pengzhao-intel commented on issue #13668: Update MKL-DNN to fix the Dense layer issue URL: https://github.com/apache/incubator-mxnet/pull/13668#issuecomment-460016722 @zheng-da please help verify the fix. This is an

[GitHub] pengzhao-intel commented on issue #13668: Update MKL-DNN to fix the Dense layer issue

pengzhao-intel commented on issue #13668: Update MKL-DNN to fix the Dense layer issue URL: https://github.com/apache/incubator-mxnet/pull/13668#issuecomment-460016667 We will update the latest MKLDNN CI before the merge.

[GitHub] zheng-da commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API

zheng-da commented on a change in pull request #14040: Reformat of TensorRT to use subgraph API URL: https://github.com/apache/incubator-mxnet/pull/14040#discussion_r253284118 ## File path: include/mxnet/c_api.h ## @@ -1835,12 +1835,6 @@ MXNET_DLL int

[GitHub] stephenrawls commented on a change in pull request #13680: [MXNET-1121] Example to demonstrate the inference workflow using RNN

stephenrawls commented on a change in pull request #13680: [MXNET-1121] Example to demonstrate the inference workflow using RNN URL: https://github.com/apache/incubator-mxnet/pull/13680#discussion_r253278383 ## File path: cpp-package/example/inference/sentiment_analysis_rnn.cpp

[GitHub] stephenrawls commented on a change in pull request #13680: [MXNET-1121] Example to demonstrate the inference workflow using RNN

stephenrawls commented on a change in pull request #13680: [MXNET-1121] Example to demonstrate the inference workflow using RNN URL: https://github.com/apache/incubator-mxnet/pull/13680#discussion_r253278924 ## File path: cpp-package/example/inference/sentiment_analysis_rnn.cpp

[GitHub] huangzhiyuan commented on issue #14056: A better split-2D(SliceChannel) op forward kernel for CPU

huangzhiyuan commented on issue #14056: A better split-2D(SliceChannel) op forward kernel for CPU URL: https://github.com/apache/incubator-mxnet/pull/14056#issuecomment-460014063 Seems the PR submitted recently block in the same test case, I will rebase my code after CNY, ths!

[GitHub] ZhennanQin commented on issue #14052: flaky test: test_operator.test_depthwise_convolution

ZhennanQin commented on issue #14052: flaky test: test_operator.test_depthwise_convolution URL: https://github.com/apache/incubator-mxnet/issues/14052#issuecomment-460016527 @mseth10, Thanks for reporting this. Confirmed it's caused by mkldnn bug. The fix is already in mkldnn master, but