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?
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,
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
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
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
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
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
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
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
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
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
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
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
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
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:
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
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`
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,
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
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
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:
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
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
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
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
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 @@
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
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 @@
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
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
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
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
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
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 @@
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 @@
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.
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
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
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
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
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 @@
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
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.
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
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
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
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 @@
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
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
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
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
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
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
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?
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
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
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.
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
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
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
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!
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
62 matches
Mail list logo