wentingj commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r182639741
##
File path: src/operator/quantization/quantize_graph_pass.cc
##
@@ -198,7 +198,7 @@
li-haoran opened a new issue #10610: memory increase when use adam and rmsprop.
URL: https://github.com/apache/incubator-mxnet/issues/10610
recently, i implement an hourglass like model for pose estimation. during
training, the memory increases. and i use the memory_profile to diagnose
zheng-da opened a new issue #10609: Gluon code fails in the normal mode but
succeeds in the hybrid mode.
URL: https://github.com/apache/incubator-mxnet/issues/10609
## Description
```python
from mxnet import gluon
import mxnet as mx
gru = gluon.rnn.GRUCell(5)
ariwaranosai commented on issue #10604: fix strip prefix in HybridBlock with
parameter to control it.
URL: https://github.com/apache/incubator-mxnet/pull/10604#issuecomment-382591897
change strip_prefix be False by default.
ZiyueHuang commented on issue #10509: [Improvement] Distinguish Sparse/Dense
operator for temp resource request
URL:
https://github.com/apache/incubator-mxnet/issues/10509#issuecomment-382590971
Sounds good. I will have a try. Thanks.
Author: haibin
Date: Thu Apr 19 01:46:44 2018
New Revision: 26411
Log:
add mxnet 1.2.0.rc0
Added:
dev/incubator/mxnet/1.2.0.rc0/apache-mxnet-src-1.2.0.rc0-incubating.tar.gz
(with props)
dev/incubator/mxnet/1.2.0.rc0/apache-mxnet-src-1.2.0.rc0-incubating.tar.gz.asc
Author: haibin
Date: Thu Apr 19 01:46:02 2018
New Revision: 26410
Log:
remove
Removed:
dev/incubator/mxnet/1.2.0.rc0/apache-mxnet-src-1.2.0.rc0-incubating.tar.gz
dev/incubator/mxnet/1.2.0.rc0/apache-mxnet-src-1.2.0.rc0-incubating.tar.gz.asc
xinyu-intel commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r182614674
##
File path: ci/docker/runtime_functions.sh
##
@@ -371,7 +371,6 @@
ThomasDelteil opened a new pull request #10608: [MX-292] Add tutorial tests to
the CI
URL: https://github.com/apache/incubator-mxnet/pull/10608
## Description ##
- Adding the tutorials to the CI
- Adding a sanity check to tutorials:
- Every tutorial needs to be
Author: haibin
Date: Thu Apr 19 01:20:42 2018
New Revision: 26409
Log:
Add mxnet-1.2.0.rc0
Added:
dev/incubator/mxnet/1.2.0.rc0/
dev/incubator/mxnet/1.2.0.rc0/apache-mxnet-src-1.2.0.rc0-incubating.tar.gz
(with props)
Author: haibin
Date: Thu Apr 19 01:16:11 2018
New Revision: 26408
Log:
remove
Removed:
dev/incubator/mxnet/1.2.0.rc0/
Author: haibin
Date: Thu Apr 19 01:14:31 2018
New Revision: 26407
Log:
Add mxnet-1.2.0.rc0
Added:
dev/incubator/mxnet/1.2.0.rc0/
dev/incubator/mxnet/1.2.0.rc0/apache-mxnet-src-1.2.0.rc0-incubating.tar.gz
(with props)
rahul003 commented on a change in pull request #10530: Jetson build with cmake
and CUDA
URL: https://github.com/apache/incubator-mxnet/pull/10530#discussion_r182611722
##
File path: ci/docker/runtime_functions.sh
##
@@ -38,9 +38,30 @@ clean_repo() {
build_jetson() {
rahul003 commented on issue #10558: NNVM build failed in the newst mxnet version
URL:
https://github.com/apache/incubator-mxnet/issues/10558#issuecomment-382576446
For the CMake build I checked both that the CPU supports f16c and compiler
supports f16c and then set the flag on. For Make I
ThomasDelteil commented on issue #10607: New tutorial on how to create a new
custom layer in Gluon
URL: https://github.com/apache/incubator-mxnet/pull/10607#issuecomment-382571821
@Ishitori you are missing the download placeholder at the end of your
tutorial and you haven't updated the
Roshrini commented on issue #10504: Can MXNet operator profiling work well
using gluon model?
URL:
https://github.com/apache/incubator-mxnet/issues/10504#issuecomment-382570070
@eric-haibin-lin Can you please add labels-
Python, Gluon, Profiler
Roshrini commented on issue #10505: profiler shoudl collect call durations not
just timestamps
URL:
https://github.com/apache/incubator-mxnet/issues/10505#issuecomment-382569557
@eric-haibin-lin could you please add label-
Profiler
Ishitori opened a new pull request #10607: New tutorial on how to create a new
custom layer in Gluon
URL: https://github.com/apache/incubator-mxnet/pull/10607
## Description ##
This pull request adds a new tutorial explaining how to create a new custom
layer using Gluon API. It doesn't
eric-haibin-lin commented on issue #60: Site refresh
URL:
https://github.com/apache/incubator-mxnet-site/pull/60#issuecomment-382567814
Wow.. big change.. How are ppl supposed to review this..?
This is an automated message
Roshrini commented on issue #10494: stop_gradient fails under adam and wd
URL:
https://github.com/apache/incubator-mxnet/issues/10494#issuecomment-382567521
@eric-haibin-lin could you please add labels-
Python, Optimizer
Roshrini commented on issue #10492: Logging level for C++ lib
URL:
https://github.com/apache/incubator-mxnet/issues/10492#issuecomment-382567329
@cjolivier01 Any idea how to set the logging level?
Also could you please add labels-
C++, Question
Roshrini commented on issue #10549: scala-package 1.1.0 build instruction
Windows VS2015
URL:
https://github.com/apache/incubator-mxnet/issues/10549#issuecomment-382565462
@nswamy Could you please add labels-
Scala, Windows, Installation
Roshrini commented on issue #10532: NDArray failed to allocate CPU memory
URL:
https://github.com/apache/incubator-mxnet/issues/10532#issuecomment-382565304
@nswamy Could you please add labels-
Python, NDarray
This is an
marcoabreu commented on a change in pull request #10354: Expose the number of
GPUs.
URL: https://github.com/apache/incubator-mxnet/pull/10354#discussion_r182601933
##
File path: include/mxnet/base.h
##
@@ -307,6 +312,21 @@ inline Context Context::GPU(int32_t dev_id) {
marcoabreu commented on issue #10354: Expose the number of GPUs.
URL: https://github.com/apache/incubator-mxnet/pull/10354#issuecomment-382564700
Exactly
This is an automated message from the Apache Git Service.
To respond to
marcoabreu commented on a change in pull request #60: Site refresh
URL:
https://github.com/apache/incubator-mxnet-site/pull/60#discussion_r182601395
##
File path: .htaccess
##
@@ -1,5 +0,0 @@
-RewriteEngine on
Review comment:
Is it intended to remove this?
Roshrini commented on issue #10432: [scala-infer]move test files for scala
infer API to s3
URL:
https://github.com/apache/incubator-mxnet/issues/10432#issuecomment-382564261
@lanking520 Can you upload models files for imageclassification and SSD
examples in S3 and change the links?
marcoabreu commented on a change in pull request #10605: [MXNET-310]
[ONNX-MXNet] API to import ONNX models into Gluon.
URL: https://github.com/apache/incubator-mxnet/pull/10605#discussion_r182600988
##
File path: tests/python-pytest/onnx/gluon_backend.py
##
@@ -0,0
Roshrini commented on issue #10480: Could you help build windows pypi package
with cuda9.0
URL:
https://github.com/apache/incubator-mxnet/issues/10480#issuecomment-382562152
@nswamy Could you please add label-
installation, windows
aaronmarkham opened a new pull request #60: Site refresh
URL: https://github.com/apache/incubator-mxnet-site/pull/60
## Description ##
Manual site refresh for 1.2.0 RC0
This is an automated message from the Apache Git
haojin2 commented on issue #10585: [MXNET-329] support SparseEmbedding with
dense weight
URL: https://github.com/apache/incubator-mxnet/pull/10585#issuecomment-382557084
LGTM
This is an automated message from the Apache Git
eric-haibin-lin commented on a change in pull request #10585: [MXNET-329]
support SparseEmbedding with dense weight
URL: https://github.com/apache/incubator-mxnet/pull/10585#discussion_r182593627
##
File path: tests/python/unittest/test_sparse_operator.py
##
@@ -1638,10
anirudh2290 commented on issue #10606: Additional v1.2.0 changes - install docs
changes, website changes
URL: https://github.com/apache/incubator-mxnet/pull/10606#issuecomment-382553520
The CI running on the branch was successful:
zhreshold commented on a change in pull request #10595: Update mobilenetv2
symbol definition
URL: https://github.com/apache/incubator-mxnet/pull/10595#discussion_r182583557
##
File path: example/image-classification/symbols/mobilenetv2.py
##
@@ -17,60 +17,239 @@
# -*-
zhreshold commented on a change in pull request #10595: Update mobilenetv2
symbol definition
URL: https://github.com/apache/incubator-mxnet/pull/10595#discussion_r182583210
##
File path: example/image-classification/symbols/mobilenetv2.py
##
@@ -17,60 +17,239 @@
# -*-
eric-haibin-lin commented on issue #10606: Additional v1.2.0 changes - install
docs changes, website changes
URL: https://github.com/apache/incubator-mxnet/pull/10606#issuecomment-382532583
I'll merge these now since they're just cherrypicked commits.
eric-haibin-lin closed pull request #10606: Additional v1.2.0 changes - install
docs changes, website changes
URL: https://github.com/apache/incubator-mxnet/pull/10606
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the
anirudh2290 commented on issue #10606: Additional v1.2.0 changes - install docs
changes, website changes
URL: https://github.com/apache/incubator-mxnet/pull/10606#issuecomment-382529735
@eric-haibin-lin can this be merged without squashing. We don't have to wait
for CI to pass since the
eric-haibin-lin closed pull request #10449: [MXNET-293] Navigation updates
URL: https://github.com/apache/incubator-mxnet/pull/10449
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As this is a
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 19c5212 [MXNET-293] Navigation
haojin2 commented on a change in pull request #10585: [MXNET-329] support
SparseEmbedding with dense weight
URL: https://github.com/apache/incubator-mxnet/pull/10585#discussion_r182565612
##
File path: tests/python/unittest/test_sparse_operator.py
##
@@ -1638,10 +1638,10
ThomasDelteil commented on issue #10485: [MXNET-304][RFC] Jenkins docs build
URL: https://github.com/apache/incubator-mxnet/pull/10485#issuecomment-382522569
@aaronmarkham re the "fixing scala issue" commit, we don't need to install
the java sdk to run scala compile docs? Or does it come
anirudh2290 commented on issue #10449: [MXNET-293] Navigation updates
URL: https://github.com/apache/incubator-mxnet/pull/10449#issuecomment-382521646
@piiswrong @mli gluon has been added back. Is this good to merge ?
This is
anirudh2290 opened a new pull request #10606: [WIP] Additional v1.2.0 changes -
install docs changes, website changes
URL: https://github.com/apache/incubator-mxnet/pull/10606
## Description ##
PR to the release branch to add : #10587 and #10449
## Checklist ##
### Essentials
eric-haibin-lin commented on issue #10585: [MXNET-329] support SparseEmbedding
with dense weight
URL: https://github.com/apache/incubator-mxnet/pull/10585#issuecomment-382514788
@ZiyueHuang @haojin2 could you review?
This is
This is an automated email from the ASF dual-hosted git repository.
cjolivier01 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 869b156 Remove pip upgrade step
cjolivier01 closed pull request #10587: Remove pip upgrade step from install
docs
URL: https://github.com/apache/incubator-mxnet/pull/10587
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As this
eric-haibin-lin commented on issue #10509: [Improvement] Distinguish
Sparse/Dense operator for temp resource request
URL:
https://github.com/apache/incubator-mxnet/issues/10509#issuecomment-382503640
After some discussion with @piiswrong and @zheng-da I think the following
additional
tdomhan commented on issue #10354: Expose the number of GPUs.
URL: https://github.com/apache/incubator-mxnet/pull/10354#issuecomment-382502963
so for CPU tests we run MXNet without CUDA built in, meaning that we compile
once for CPU tests and then also for GPU tests with different flags?
rahul003 commented on issue #10531: Float16 Support for dot
URL:
https://github.com/apache/incubator-mxnet/issues/10531#issuecomment-382488353
Can we close this now?
This is an automated message from the Apache Git Service.
rahul003 commented on issue #9629: do not save gpu memory during fp16 training
URL:
https://github.com/apache/incubator-mxnet/issues/9629#issuecomment-382485782
@tornadomeet / @315386775 Were you both using such low batch sizes? I
imagine in such cases, the CUDA overhead might dominate
zhreshold closed pull request #10603: fix MobileNetV2 structure according to
1801.04381 v3
URL: https://github.com/apache/incubator-mxnet/pull/10603
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
This is an automated email from the ASF dual-hosted git repository.
zhreshold 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 7661fee fix MobileNetV2 structure
kpmurali commented on issue #10449: [MXNET-293] Navigation updates
URL: https://github.com/apache/incubator-mxnet/pull/10449#issuecomment-382482818
I have added Gluon back to the navbar. Please review @aaronmarkham @mli
@piiswrong @srochel
aaronmarkham commented on issue #10449: [MXNET-293] Navigation updates
URL: https://github.com/apache/incubator-mxnet/pull/10449#issuecomment-382482209
@srochel - this was intended to go out after 1.2 RC, but when ONNX was
removed from the home page to make room for the 60-min crash
anirudh2290 commented on issue #10587: Remove pip upgrade step from install docs
URL: https://github.com/apache/incubator-mxnet/pull/10587#issuecomment-382450910
@gsemet Thanks for your inputs. Even with the `--user` flag the system pip
breaks on Ubuntu 16.04:
```
$ pip install
piiswrong closed pull request #10572: Make frontpage nav fonts bolder
URL: https://github.com/apache/incubator-mxnet/pull/10572
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As this is a foreign
This is an automated email from the ASF dual-hosted git repository.
jxie pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git
The following commit(s) were added to refs/heads/master by this push:
new 1b0afc0 Make frontpage nav fonts bolder
piiswrong commented on issue #10604: fix strip prefix in HybridBlock with
parameter to control it.
URL: https://github.com/apache/incubator-mxnet/pull/10604#issuecomment-382474076
You can use net.collect_params().load to load the exported network.
strip_prefix should be False by
indhub commented on issue #10568: [WIP] [MXNET-325] Model parallelism tutorial.
URL: https://github.com/apache/incubator-mxnet/pull/10568#issuecomment-382472168
@ThomasDelteil When I run it on two GPUs, I see 100% utilization of both
GPUs. There are no blocking operations in the code.
spidyDev commented on a change in pull request #10605: [MXNET-310] [ONNX-MXNet]
API to import ONNX models into Gluon.
URL: https://github.com/apache/incubator-mxnet/pull/10605#discussion_r182505547
##
File path: python/mxnet/contrib/onnx/_import/import_onnx.py
##
@@
spidyDev commented on a change in pull request #10605: [MXNET-310] [ONNX-MXNet]
API to import ONNX models into Gluon.
URL: https://github.com/apache/incubator-mxnet/pull/10605#discussion_r182510986
##
File path: python/mxnet/contrib/onnx/_import/import_onnx.py
##
@@
spidyDev commented on a change in pull request #10605: [MXNET-310] [ONNX-MXNet]
API to import ONNX models into Gluon.
URL: https://github.com/apache/incubator-mxnet/pull/10605#discussion_r182511252
##
File path: python/mxnet/contrib/onnx/_import/import_to_gluon.py
##
@@
spidyDev commented on a change in pull request #10605: [MXNET-310] [ONNX-MXNet]
API to import ONNX models into Gluon.
URL: https://github.com/apache/incubator-mxnet/pull/10605#discussion_r182510350
##
File path: python/mxnet/contrib/onnx/_import/import_model.py
##
@@
spidyDev commented on a change in pull request #10605: [MXNET-310] [ONNX-MXNet]
API to import ONNX models into Gluon.
URL: https://github.com/apache/incubator-mxnet/pull/10605#discussion_r182512877
##
File path: tests/python-pytest/onnx/gluon_backend.py
##
@@ -0,0 +1,148
spidyDev commented on a change in pull request #10605: [MXNET-310] [ONNX-MXNet]
API to import ONNX models into Gluon.
URL: https://github.com/apache/incubator-mxnet/pull/10605#discussion_r182507348
##
File path: tests/python-pytest/onnx/gluon_backend_test.py
##
@@ -0,0
piiswrong commented on issue #10605: [MXNET-310] [ONNX-MXNet] API to import
ONNX models into Gluon.
URL: https://github.com/apache/incubator-mxnet/pull/10605#issuecomment-382470853
This is unnecessary. You can use gluon's SymbolBlock to directly load symbol
models.
haojin2 commented on a change in pull request #10557: [MXNET-322]
Elemwise_mul(row_sparse, dense) = row_sparse on CPU
URL: https://github.com/apache/incubator-mxnet/pull/10557#discussion_r182510261
##
File path: tests/python/unittest/test_sparse_operator.py
##
@@ -322,17
kpmurali commented on issue #10449: [MXNET-293] Navigation updates
URL: https://github.com/apache/incubator-mxnet/pull/10449#issuecomment-382468370
Ok, I will add Gluon and its links back to the Navbar unless anyone has any
objections.
indhub commented on issue #10568: [WIP] [MXNET-325] Model parallelism tutorial.
URL: https://github.com/apache/incubator-mxnet/pull/10568#issuecomment-382466399
@mli
This is an automated message from the Apache Git Service.
anirudh2290 commented on issue #10587: Remove pip upgrade step from install docs
URL: https://github.com/apache/incubator-mxnet/pull/10587#issuecomment-382461854
Yes it's in the path. You can see that I have outputted path. which pip
returns ~/.local/bin/pip. It's weird because when I
anirudhacharya opened a new pull request #10605: [MXNET-310] [ONNX-MXNet] API
to import ONNX models into Gluon.
URL: https://github.com/apache/incubator-mxnet/pull/10605
## Description ##
API and corresponding tests to import ONNX models into Gluon
## Checklist ##
###
gsemet commented on issue #10587: Remove pip upgrade step from install docs
URL: https://github.com/apache/incubator-mxnet/pull/10587#issuecomment-382455740
what does `which pip` returns? You need to ensure you have ~/.local/bin in
your PATH
anirudh2290 commented on issue #10587: Remove pip upgrade step from install docs
URL: https://github.com/apache/incubator-mxnet/pull/10587#issuecomment-382450910
@gsemet Thanks for your inputs. Even with the `--user` flag the system pip
breaks on Ubuntu 16.04:
```
$ pip install
marcoabreu commented on issue #10586: Cannot build ubuntu_gpu CI docker image
(java8 sdk missing)
URL:
https://github.com/apache/incubator-mxnet/issues/10586#issuecomment-382430009
I'm indifferent to that.
This is an
marcoabreu commented on issue #10433: [MXNET-290] MKLDNN support for model
quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#issuecomment-382426955
@reminisce could you please review this PR?
This is an
marcoabreu commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r182468883
##
File path: src/operator/quantization/quantized_conv.cc
##
@@ -86,12 +89,10 @@ bool
marcoabreu commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r182468602
##
File path: src/operator/quantization/quantize_graph_pass.cc
##
@@ -198,7 +198,7 @@
marcoabreu commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r182468565
##
File path: src/operator/quantization/quantize_graph_pass.cc
##
@@ -159,7 +159,7 @@
marcoabreu commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r182467890
##
File path: src/c_api/c_api_symbolic.cc
##
@@ -595,7 +597,12 @@ int
marcoabreu commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r182467604
##
File path: python/mxnet/contrib/quantization.py
##
@@ -72,7 +72,7 @@ def
marcoabreu commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r182467087
##
File path: include/mxnet/c_api.h
##
@@ -1409,13 +1409,17 @@ MXNET_DLL int
marcoabreu commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r182466761
##
File path: ci/docker/runtime_functions.sh
##
@@ -407,26 +405,28 @@
marcoabreu commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r182466678
##
File path: ci/docker/runtime_functions.sh
##
@@ -407,26 +405,28 @@
marcoabreu commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r182466521
##
File path: ci/docker/runtime_functions.sh
##
@@ -407,26 +405,28 @@
ariwaranosai opened a new pull request #10604: fix strip prefix in HybridBlock
with parameter to control it.
URL: https://github.com/apache/incubator-mxnet/pull/10604
## Description ##
fix #10601
add strip_prefix parameter and strip prefix during export model.
code
srochel commented on issue #10449: [MXNET-293] Navigation updates
URL: https://github.com/apache/incubator-mxnet/pull/10449#issuecomment-382398507
The PR is blocking the 1.2 RC creation. Can we keep Gluon in navigation bar
until we have a complete proposal in place, reviewed and agreed? We
dwSun opened a new pull request #10603: fix MobileNetV2 structure according to
1801.04381 v3
URL: https://github.com/apache/incubator-mxnet/pull/10603
## Description ##
fix net structure according to 1801.04381 v3.
relu -> relu6 in MobileNetV2.
## Checklist ##
- [ Y ]
chinakook commented on issue #10599: [MKLDNN Bug] MKLDNN eats lots of memory
and then crash down.
URL:
https://github.com/apache/incubator-mxnet/issues/10599#issuecomment-382382640
Thanks! As I've tested before, when inferencing a scene segmentation network
such as Deeplab or U-Net,
pengzhao-intel commented on issue #10599: [MKLDNN Bug] MKLDNN eats lots of
memory and then crash down.
URL:
https://github.com/apache/incubator-mxnet/issues/10599#issuecomment-382380820
The case consumes about 40G peak memory w/ MKL-DNN in 8.4 seconds while 12G
w/ default CPU path in
chinakook commented on issue #10600: Gluon Interpolation Layer
URL:
https://github.com/apache/incubator-mxnet/issues/10600#issuecomment-382375451
I don't know. I think this op is weird.
This is an automated message from the
crazyleg opened a new issue #10602: mxnet.model.server custom Service
model.get_outputs() returns only one result
URL: https://github.com/apache/incubator-mxnet/issues/10602
## Description
When using mxnet.model.server custom Service model.get_outputs() returns
only one result, while
marcoabreu commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r182411485
##
File path: ci/docker/runtime_functions.sh
##
@@ -371,7 +371,6 @@
marcoabreu commented on a change in pull request #10433: [MXNET-290] MKLDNN
support for model quantization
URL: https://github.com/apache/incubator-mxnet/pull/10433#discussion_r182411505
##
File path: ci/docker/runtime_functions.sh
##
@@ -382,7 +381,6 @@
PKurnikov commented on issue #10600: Gluon Interpolation Layer
URL:
https://github.com/apache/incubator-mxnet/issues/10600#issuecomment-382372938
@chinakook ,
Thanks, this way is work!
But the function BilinearResize2D does not work with 'width' more then 1000.
Are there any
ariwaranosai opened a new issue #10601: load parameters error with the same
code exported by HybridBlock
URL: https://github.com/apache/incubator-mxnet/issues/10601
Note: Providing complete information in the most concise form is the best
way to get help. This issue template serves as
chinakook commented on issue #10600: Gluon Interpolation Layer
URL:
https://github.com/apache/incubator-mxnet/issues/10600#issuecomment-382346176
```
def hybrid_forward(self, F, x):
return F.contrib.BilinearResize2D(x, self.height, self.width)
```
or
```
PKurnikov commented on issue #10600: Gluon Interpolation Layer
URL:
https://github.com/apache/incubator-mxnet/issues/10600#issuecomment-382344098
Unfortunately, I can't to use this approach
> nn.Block
, because I use BatchNorm, HybridSequential classes which contains
chinakook commented on issue #10600: Gluon Interpolation Layer
URL:
https://github.com/apache/incubator-mxnet/issues/10600#issuecomment-382339302
del hybrid
```
class BilinearInterp(nn.Block):
def __init__(self, target_size, **kwargs):
super(BilinearInterp,
PKurnikov opened a new issue #10600: Gluon Interpolation Layer
URL: https://github.com/apache/incubator-mxnet/issues/10600
I tried to implement the layer on gluon like this:
```
class BilinearInterp(nn.HybridBlock):
def __init__(self, target_size, **kwargs):
1 - 100 of 109 matches
Mail list logo