pengzhao-intel closed issue #12703: Performance Regression from #12019
URL: https://github.com/apache/incubator-mxnet/issues/12703
This is an automated message from the Apache Git Service.
To respond to the message, please lo
pengzhao-intel commented on issue #12703: Performance Regression from #12019
URL:
https://github.com/apache/incubator-mxnet/issues/12703#issuecomment-428848050
fix is merged so close.
This is an automated message from the Apa
zheng-da commented on issue #11948: [MXNET-766] add unroll RNN for HybridBlock
URL: https://github.com/apache/incubator-mxnet/pull/11948#issuecomment-428848233
i don't think this PR will be merged. I'll close it.
This is an au
zheng-da closed pull request #11948: [MXNET-766] add unroll RNN for HybridBlock
URL: https://github.com/apache/incubator-mxnet/pull/11948
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
KellenSunderland commented on a change in pull request #12791: [MXNET-1093] Add
python3 Docker images for each MXNet release
URL: https://github.com/apache/incubator-mxnet/pull/12791#discussion_r224353651
##
File path: docker/docker-python/Dockerfile.mxnet.python3.cpu
##
@
KellenSunderland commented on a change in pull request #12791: [MXNET-1093] Add
python3 Docker images for each MXNet release
URL: https://github.com/apache/incubator-mxnet/pull/12791#discussion_r224353651
##
File path: docker/docker-python/Dockerfile.mxnet.python3.cpu
##
@
andrewfayres commented on a change in pull request #12772: [MXNET-984] Add Java
NDArray and introduce Java Operator Builder class
URL: https://github.com/apache/incubator-mxnet/pull/12772#discussion_r224356345
##
File path:
scala-package/core/src/main/scala/org/apache/mxnet/javaapi
andrewfayres commented on a change in pull request #12772: [MXNET-984] Add Java
NDArray and introduce Java Operator Builder class
URL: https://github.com/apache/incubator-mxnet/pull/12772#discussion_r224362087
##
File path: scala-package/scalastyle-config.xml
##
@@ -64,7 +
andrewfayres commented on issue #12772: [MXNET-984] Add Java NDArray and
introduce Java Operator Builder class
URL: https://github.com/apache/incubator-mxnet/pull/12772#issuecomment-428873527
I like the idea of getting a test coverage measurement working early on the
java api. I think that
marcoabreu commented on issue #12792: Fix broken amalgamation
URL: https://github.com/apache/incubator-mxnet/pull/12792#issuecomment-428894576
Thanks for your contribution! Could you elaborate how you compile? We run
amalgamation in our CI and this should have been caught
-
KellenSunderland commented on issue #12782: WIP: [MXNET-674] Speed up GPU
builds.
URL: https://github.com/apache/incubator-mxnet/pull/12782#issuecomment-428897001
Including cubins for sm52 and sm70 for all builds only marginally increases
build times (looks like an extra ~1.5 minute builds
KellenSunderland edited a comment on issue #12782: WIP: [MXNET-674] Speed up
GPU builds.
URL: https://github.com/apache/incubator-mxnet/pull/12782#issuecomment-428897001
Including cubins for sm52 and sm70 for all builds only marginally increases
build times (looks like an extra ~1.5 minute
marcoabreu commented on issue #12782: [MXNET-674] Speed up GPU builds in CI
URL: https://github.com/apache/incubator-mxnet/pull/12782#issuecomment-428898037
Sounds good to me
This is an automated message from the Apache Git Se
KellenSunderland commented on issue #12782: [MXNET-674] Speed up GPU builds in
CI
URL: https://github.com/apache/incubator-mxnet/pull/12782#issuecomment-428898454
Cool beans. Ok let's call this one ready for review. Any other feedback
appreciated @marcoabreu or @sandeep-krishnamurthy.
daedaesplayground1911 commented on issue #11612: Find government repositories
URL:
https://github.com/apache/incubator-mxnet/issues/11612#issuecomment-428906676
Yes I'm specifically searching for government liquidation sites and places
where I can aquire documents relating to services that
daedaesplayground1911 commented on issue #11612: Find government repositories
URL:
https://github.com/apache/incubator-mxnet/issues/11612#issuecomment-428907412
I'm trying to find government liquidation places that specifically
specialize in selling vehicles, weapons, used smartphones, and
sxjscience commented on issue #12310: Flaky test: test_ndarray.test_order
URL:
https://github.com/apache/incubator-mxnet/issues/12310#issuecomment-428914285
The bug is caused by memory misalignment. As stated in
https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#device-memory
marcoabreu commented on a change in pull request #12791: [MXNET-1093] Add
python3 Docker images for each MXNet release
URL: https://github.com/apache/incubator-mxnet/pull/12791#discussion_r224409081
##
File path: docker/docker-python/build_python_dockerfile.sh
##
@@ -117,1
marcoabreu commented on a change in pull request #12791: [MXNET-1093] Add
python3 Docker images for each MXNet release
URL: https://github.com/apache/incubator-mxnet/pull/12791#discussion_r224408388
##
File path: docker/docker-python/README.md
##
@@ -12,13 +12,21 @@ It use
marcoabreu commented on a change in pull request #12791: [MXNET-1093] Add
python3 Docker images for each MXNet release
URL: https://github.com/apache/incubator-mxnet/pull/12791#discussion_r224408013
##
File path: docker/docker-python/Dockerfile.mxnet.python3.cpu
##
@@ -0,0
marcoabreu commented on a change in pull request #12791: [MXNET-1093] Add
python3 Docker images for each MXNet release
URL: https://github.com/apache/incubator-mxnet/pull/12791#discussion_r224408602
##
File path: docker/docker-python/README.md
##
@@ -39,3 +47,10 @@ Credent
marcoabreu commented on a change in pull request #12782: [MXNET-674] Speed up
GPU builds in CI
URL: https://github.com/apache/incubator-mxnet/pull/12782#discussion_r224410262
##
File path: ci/docker/runtime_functions.sh
##
@@ -559,13 +563,15 @@ build_ubuntu_gpu_mkldnn_nocu
lostella commented on issue #12783: Name conflict when serializing
HybridSequentialRNNCell
URL:
https://github.com/apache/incubator-mxnet/issues/12783#issuecomment-428921900
I simplified the MWE:
https://gist.github.com/lostella/9a790fd89726c1741a1fcf4194a5dac6
It seems like it's u
jlcontreras commented on issue #11945: python/setup.py modified to accept a
package name argument.
URL: https://github.com/apache/incubator-mxnet/pull/11945#issuecomment-428938287
Sorry for the late reply, I'm away until the beginning of next week and then
I'll answer everything!
soeque1 edited a comment on issue #12778: Cannot print parameter summary of
embedding layer
URL:
https://github.com/apache/incubator-mxnet/issues/12778#issuecomment-428455018
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/visualization.py
It seems that the Embe
KellenSunderland commented on a change in pull request #12782: [MXNET-674]
Speed up GPU builds in CI
URL: https://github.com/apache/incubator-mxnet/pull/12782#discussion_r224435265
##
File path: ci/docker/runtime_functions.sh
##
@@ -559,13 +563,15 @@ build_ubuntu_gpu_mkldn
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 cef6092 Bump the publish ti
KellenSunderland commented on a change in pull request #12791: [MXNET-1093] Add
python3 Docker images for each MXNet release
URL: https://github.com/apache/incubator-mxnet/pull/12791#discussion_r224437572
##
File path: docker/docker-python/build_python_dockerfile.sh
##
@@
KellenSunderland commented on issue #12782: [MXNET-674] Speed up GPU builds in
CI
URL: https://github.com/apache/incubator-mxnet/pull/12782#issuecomment-428947180
Yeah took a look at windows, everything seems good there (configured to
build for sm52)
--
lostella opened a new pull request #12794: fixed symbols naming in RNNCell and
LSTMCell
URL: https://github.com/apache/incubator-mxnet/pull/12794
## Description ##
This partially fixes #12783, by assigning *all* nodes in `hybrid_forward` a
unique `name`. Some operations were in fact per
larroy commented on a change in pull request #12094: [MXNET-793] ★ Virtualized
testing in CI with QEMU ★
URL: https://github.com/apache/incubator-mxnet/pull/12094#discussion_r224442806
##
File path: ci/qemu/run.sh
##
@@ -0,0 +1,32 @@
+#!/bin/bash
Review comment:
done
larroy commented on a change in pull request #12094: [MXNET-793] ★ Virtualized
testing in CI with QEMU ★
URL: https://github.com/apache/incubator-mxnet/pull/12094#discussion_r224442902
##
File path: ci/qemu/run.sh
##
@@ -0,0 +1,32 @@
+#!/bin/bash
+
+# Licensed to the Apach
soeque1 edited a comment on issue #12778: Cannot print parameter summary of
embedding layer
URL:
https://github.com/apache/incubator-mxnet/issues/12778#issuecomment-428455018
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/visualization.py
It seems that the Embe
vafl opened a new issue #12795: Deserialization problem with gluon `ValueError:
There are multiple outputs with name ...`
URL: https://github.com/apache/incubator-mxnet/issues/12795
## Description
For a simple HybridBlock, saving and deserializing the symbol fails with
mxnet 1.3 whe
marcoabreu commented on issue #12795: Deserialization problem with gluon
`ValueError: There are multiple outputs with name ...`
URL:
https://github.com/apache/incubator-mxnet/issues/12795#issuecomment-428971005
@srochel
This
marcoabreu edited a comment on issue #12795: Deserialization problem with gluon
`ValueError: There are multiple outputs with name ...`
URL:
https://github.com/apache/incubator-mxnet/issues/12795#issuecomment-428971005
@srochel @lupesko
-
apeforest commented on a change in pull request #11952: [MXNET-707] Add unit
test for mxnet to coreml converter
URL: https://github.com/apache/incubator-mxnet/pull/11952#discussion_r224465056
##
File path: ci/travis/install.sh
##
@@ -22,5 +22,5 @@ export HOMEBREW_NO_AUTO_U
apeforest commented on issue #10171: CoreML tool, wrong mlmodelc file name for
Inception-BN
URL:
https://github.com/apache/incubator-mxnet/issues/10171#issuecomment-428972880
@piyushghai I will take a look.
This is an autom
lostella commented on issue #12794: fixed symbols naming in RNNCell and LSTMCell
URL: https://github.com/apache/incubator-mxnet/pull/12794#issuecomment-428973988
In light of #12795, this PR is more like a workaround. There might be some
deeper issue with `+` and `*` (and `-` and `/`?)
marcoabreu commented on a change in pull request #12094: [MXNET-793] ★
Virtualized testing in CI with QEMU ★
URL: https://github.com/apache/incubator-mxnet/pull/12094#discussion_r224472418
##
File path: ci/docker/runtime_functions.py
##
@@ -0,0 +1,131 @@
+#!/usr/bin/env py
marcoabreu commented on a change in pull request #12094: [MXNET-793] ★
Virtualized testing in CI with QEMU ★
URL: https://github.com/apache/incubator-mxnet/pull/12094#discussion_r224472937
##
File path: tests/nightly/apache_rat_license_check/rat-excludes
##
@@ -57,3 +57,4
marcoabreu commented on a change in pull request #12782: [MXNET-674] Speed up
GPU builds in CI
URL: https://github.com/apache/incubator-mxnet/pull/12782#discussion_r224475752
##
File path: ci/docker/runtime_functions.sh
##
@@ -559,13 +563,15 @@ build_ubuntu_gpu_mkldnn_nocu
marcoabreu commented on a change in pull request #12791: [MXNET-1093] Add
python3 Docker images for each MXNet release
URL: https://github.com/apache/incubator-mxnet/pull/12791#discussion_r224476519
##
File path: docker/docker-python/build_python_dockerfile.sh
##
@@ -117,1
soeque1 opened a new pull request #12796: Add embedding to print_summary
URL: https://github.com/apache/incubator-mxnet/pull/12796
## Description ##
(Brief description on what this PR is about)
## Checklist ##
### Essentials ###
Please feel free to remove inapplicable items f
piyushghai commented on issue #12795: Deserialization problem with gluon
`ValueError: There are multiple outputs with name ...`
URL:
https://github.com/apache/incubator-mxnet/issues/12795#issuecomment-428999167
@mxnet-label-bot [Bug, Gluon]
piyushghai commented on issue #12795: Deserialization problem with gluon
`ValueError: There are multiple outputs with name ...`
URL:
https://github.com/apache/incubator-mxnet/issues/12795#issuecomment-428999426
This seems similar to #12783
lostella commented on issue #12795: Deserialization problem with gluon
`ValueError: There are multiple outputs with name ...`
URL:
https://github.com/apache/incubator-mxnet/issues/12795#issuecomment-429003322
Looks like the problem occurs with `Dense` as well, so the issue probably
lies i
aaronmarkham opened a new issue #12797: test_subgraph.test_pos_conv_bn_sum_relu
fails in CI
URL: https://github.com/apache/incubator-mxnet/issues/12797
Unrelated .js update failed in CI:
http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/PR-12781/4/p
aaronmarkham commented on issue #12781: Fixed issue #12745
URL: https://github.com/apache/incubator-mxnet/pull/12781#issuecomment-429012821
Looks like we're hung up on a flaky test. I opened an issue here:
https://github.com/apache/incubator-mxnet/issues/12797
I'll retrigger CI, but it m
ufoym commented on issue #12792: Fix broken amalgamation
URL: https://github.com/apache/incubator-mxnet/pull/12792#issuecomment-429015317
I've just tried to compile amalgamation in Ubuntu and everything seems OK.
But it is not the case when compiling in OS X. So more specifically, the fix
marcoabreu commented on issue #12792: Fix broken amalgamation
URL: https://github.com/apache/incubator-mxnet/pull/12792#issuecomment-429016088
Ah! @lebeg wdyt?
This is an automated message from the Apache Git Service.
To respo
sxjscience opened a new pull request #12798: Fix Flaky Topk
URL: https://github.com/apache/incubator-mxnet/pull/12798
## Description ##
Continues https://github.com/apache/incubator-mxnet/pull/12446. Fixes fixes
the flaky topk test reported in
https://github.com/apache/incubator-mxnet/p
sxjscience closed pull request #12446: [WIP][Bugfix] Fix flaky topk
URL: https://github.com/apache/incubator-mxnet/pull/12446
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 pu
sxjscience commented on issue #12798: Fix Flaky Topk
URL: https://github.com/apache/incubator-mxnet/pull/12798#issuecomment-429018551
@kalyc @ankkhedia @lebeg
This is an automated message from the Apache Git Service.
To respon
sandeep-krishnamurthy closed issue #12797:
test_subgraph.test_pos_conv_bn_sum_relu fails in CI
URL: https://github.com/apache/incubator-mxnet/issues/12797
This is an automated message from the Apache Git Service.
To respond
sxjscience edited a comment on issue #12798: Fix Flaky Topk
URL: https://github.com/apache/incubator-mxnet/pull/12798#issuecomment-429018551
@kalyc @ankkhedia @lebeg @marcoabreu
This is an automated message from the Apache Git
sandeep-krishnamurthy commented on issue #12797:
test_subgraph.test_pos_conv_bn_sum_relu fails in CI
URL:
https://github.com/apache/incubator-mxnet/issues/12797#issuecomment-429019751
I have reverted my PR on Batchnorm that brought in this issue to master.
Unblocking master and PR builds
sxjscience edited a comment on issue #12798: Fix Flaky Topk
URL: https://github.com/apache/incubator-mxnet/pull/12798#issuecomment-429018551
@kalyc @ankkhedia @lebeg @marcoabreu @vandanavk
This is an automated message from the
larroy commented on a change in pull request #12094: [MXNET-793] ★ Virtualized
testing in CI with QEMU ★
URL: https://github.com/apache/incubator-mxnet/pull/12094#discussion_r224514254
##
File path: tests/nightly/apache_rat_license_check/rat-excludes
##
@@ -57,3 +57,4 @@ a
aaronmarkham commented on issue #12781: Fixed issue #12745
URL: https://github.com/apache/incubator-mxnet/pull/12781#issuecomment-429023905
Now tracking https://github.com/apache/incubator-mxnet/pull/12788 - might
need to rebase after this PR gets merged.
--
piyushghai commented on issue #12431: [R] use of mx.io.arrayiter completely
crashes R environment
URL:
https://github.com/apache/incubator-mxnet/issues/12431#issuecomment-429029325
@some-guy1 The PR for fixing this got merged. Can you verify and close this
issue if the bug is bug ?
sandeep-krishnamurthy commented on issue #12794: fixed symbols naming in
RNNCell and LSTMCell
URL: https://github.com/apache/incubator-mxnet/pull/12794#issuecomment-429033286
@piyushghai @szha
This is an automated message fr
marcoabreu commented on issue #12798: Fix Flaky Topk
URL: https://github.com/apache/incubator-mxnet/pull/12798#issuecomment-429033493
Thanks for the fix! Could you submit a test case that would have failed
previously? Also, do we have any performance benchmarks for this kernel change?
anirudhacharya commented on a change in pull request #12748: [MXNET-1028]
Adding CELU Activation function
URL: https://github.com/apache/incubator-mxnet/pull/12748#discussion_r224525239
##
File path: python/mxnet/gluon/nn/activations.py
##
@@ -181,6 +181,25 @@ def hybrid_f
piyushghai commented on issue #12799: ERROR: test_ndarray.test_order fails on
ARMv7 (Qemu)
URL:
https://github.com/apache/incubator-mxnet/issues/12799#issuecomment-429033855
@mxnet-label-bot [Flaky]
This is an automated mes
larroy commented on issue #12799: ERROR: test_ndarray.test_order fails on ARMv7
(Qemu)
URL:
https://github.com/apache/incubator-mxnet/issues/12799#issuecomment-429034160
@piyushghai is not flaky.
This is an automated messag
larroy commented on issue #12799: ERROR: test_ndarray.test_order fails on ARMv7
(Qemu)
URL:
https://github.com/apache/incubator-mxnet/issues/12799#issuecomment-429034472
@mxnet-label-bot, please add labels :[ARM, NDArray, Bug]
--
piyushghai commented on issue #12794: fixed symbols naming in RNNCell and
LSTMCell
URL: https://github.com/apache/incubator-mxnet/pull/12794#issuecomment-429036417
@lostella You are right. The issue is with the ```-``` and ```/``` as well.
Just tried the Dense Layer example as well.
soeque1 commented on issue #12796: Add embedding to print_summary
URL: https://github.com/apache/incubator-mxnet/pull/12796#issuecomment-429038277
The Post (#12778) does not have enough information to run the code.
So I added something to the sample code.
```
from mxnet import
vandanavk commented on a change in pull request #12785: [MXNET-906] ONNX
export/import: Selu
URL: https://github.com/apache/incubator-mxnet/pull/12785#discussion_r224532125
##
File path: tests/python-pytest/onnx/import/test_cases.py
##
@@ -98,6 +98,7 @@
'test_MaxPool'
ArmageddonKnight commented on a change in pull request #11364: [MXNET-490]
Added OpenLSTMRNN together with benchmarks and Tensorboard callback routines.
URL: https://github.com/apache/incubator-mxnet/pull/11364#discussion_r224534555
##
File path: src/operator/contrib/cu_open_lstm_rn
soeque1 edited a comment on issue #12796: Add embedding to print_summary
URL: https://github.com/apache/incubator-mxnet/pull/12796#issuecomment-429038277
The Post (#12778) does not have enough information to run the code.
So I added something to the sample code.
```
from mxnet
marcoabreu commented on issue #12796: Add embedding to print_summary
URL: https://github.com/apache/incubator-mxnet/pull/12796#issuecomment-429051934
Am I catching it wrong or is test_print_summary() actually not doing any
validation?
--
Vikas89 opened a new issue #12800: getting segfault while running
train_cifar10.py program in example directory
URL: https://github.com/apache/incubator-mxnet/issues/12800
I am trying to run this command:
python example/image-classification/train_cifar10.py
And getting the segfau
aaronmarkham commented on issue #12689: Search update
URL: https://github.com/apache/incubator-mxnet/pull/12689#issuecomment-429055202
@sandeep-krishnamurthy @safrooze I've re-enabled the preview. You should
clear your cache to make sure you're getting the latest .htaccess file.
This wa
marcoabreu commented on a change in pull request #12094: [MXNET-793] ★
Virtualized testing in CI with QEMU ★
URL: https://github.com/apache/incubator-mxnet/pull/12094#discussion_r224546252
##
File path: tests/nightly/apache_rat_license_check/rat-excludes
##
@@ -57,3 +57,4
ankkhedia commented on issue #12798: Fix Flaky Topk
URL: https://github.com/apache/incubator-mxnet/pull/12798#issuecomment-429063399
Hi @sxjscience, request you to please look into the failing CI for the PR.
This is an automat
bhavinthaker commented on a change in pull request #12791: [MXNET-1093] Add
python3 Docker images for each MXNet release
URL: https://github.com/apache/incubator-mxnet/pull/12791#discussion_r224554242
##
File path: docker/docker-python/Dockerfile.mxnet.python3.cpu
##
@@ -0
mbaijal commented on a change in pull request #12791: [MXNET-1093] Add python3
Docker images for each MXNet release
URL: https://github.com/apache/incubator-mxnet/pull/12791#discussion_r224557167
##
File path: docker/docker-python/Dockerfile.mxnet.python3.cpu
##
@@ -0,0 +1
mbaijal commented on a change in pull request #12791: [MXNET-1093] Add python3
Docker images for each MXNet release
URL: https://github.com/apache/incubator-mxnet/pull/12791#discussion_r224558239
##
File path: docker/docker-python/README.md
##
@@ -39,3 +47,10 @@ Credential
mbaijal commented on a change in pull request #12791: [MXNET-1093] Add python3
Docker images for each MXNet release
URL: https://github.com/apache/incubator-mxnet/pull/12791#discussion_r224558542
##
File path: docker/docker-python/Dockerfile.mxnet.python3.cpu
##
@@ -0,0 +1
vandanavk commented on issue #12785: [MXNET-906] ONNX export/import: Selu
URL: https://github.com/apache/incubator-mxnet/pull/12785#issuecomment-429074177
Adding @Roshrini @zhreshold for review
This is an automated message fro
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 9c8d603 Bump the publish ti
ChaiBapchya opened a new issue #12801: Inconsistent function signature of
NDarray.randn
URL: https://github.com/apache/incubator-mxnet/issues/12801
## Description
NDArray.randn function signature
`def rand (*shape, **kwargs):`
is incorrect and inconsistent with rest of the signatu
anirudhacharya commented on issue #12801: Inconsistent function signature of
NDarray.randn
URL:
https://github.com/apache/incubator-mxnet/issues/12801#issuecomment-429083958
can you please add links and references in the above issue description
szha commented on issue #11585: Adds support for finding cudnn inside cuda path
URL: https://github.com/apache/incubator-mxnet/pull/11585#issuecomment-429084947
@lebeg is this something you can pick up?
This is an automated me
szha commented on issue #12783: Name conflict when serializing LSTMCell
URL:
https://github.com/apache/incubator-mxnet/issues/12783#issuecomment-429086305
The problem is in the sample code. Since HybridSequentialRNNCell is a
container block, you need to use its name_scope if you intend to
szha edited a comment on issue #12783: Name conflict when serializing LSTMCell
URL:
https://github.com/apache/incubator-mxnet/issues/12783#issuecomment-429086305
For the first piece of code, the problem is not using the container's
name_scope. Since HybridSequentialRNNCell is a container b
danithaca commented on issue #6783: mx.set.seed() does not work using GPU
URL:
https://github.com/apache/incubator-mxnet/issues/6783#issuecomment-429089214
Read the code again, and the results and explanation makes sense. Thanks
@ankkhedia. @nswamy -- thanks for reopening the issue. I agre
lostella commented on issue #12783: Name conflict when serializing LSTMCell
URL:
https://github.com/apache/incubator-mxnet/issues/12783#issuecomment-429091366
Unfortunately, that does not seem to solve the issue. See also the simpler
example in my previous comment, which does not involve H
szha commented on issue #12783: Name conflict when serializing LSTMCell
URL:
https://github.com/apache/incubator-mxnet/issues/12783#issuecomment-429092412
For the problem with LSTM alone, the problem is in not naming some of the
elementwise operations. In LSTM there are three names that ar
szha commented on issue #12783: Name conflict when serializing LSTMCell
URL:
https://github.com/apache/incubator-mxnet/issues/12783#issuecomment-429092674
This problem exists in RNN and GRU as well, so all three needs to be patched.
-
ankkhedia commented on issue #6783: mx.set.seed() does not work using GPU
URL:
https://github.com/apache/incubator-mxnet/issues/6783#issuecomment-429095180
@nswamy @sandeep-krishnamurthy Could you please close the issue as the
confusion has been resolved.
-
piyushghai commented on issue #12800: getting segfault while running
train_cifar10.py program in example directory
URL:
https://github.com/apache/incubator-mxnet/issues/12800#issuecomment-429095935
Thanks for reporting this issue @Vikas89
@mxnet-label-bot [Example, Bug]
--
sameer-kapoor opened a new issue #12802: Crash while loading libmxnet-scala.so
URL: https://github.com/apache/incubator-mxnet/issues/12802
## Description
Random crashes while loading libmxnet-scala.so with both Mxnet 1.2.0 and
1.3.0
## Environment info (Required)
```
piyushghai commented on issue #11542: Import to gluon not working for ONNX model
URL:
https://github.com/apache/incubator-mxnet/issues/11542#issuecomment-429096687
Similar issues perhaps : #12795 , #12794
This is an automate
azai91 commented on issue #12724: simplify mac mkldnn build
URL: https://github.com/apache/incubator-mxnet/pull/12724#issuecomment-429097045
@xinyu-intel
I get the following when running on resnet-50 with MKLDNN
```
INFO:root:batch size 1, dtype float32, images/sec: 16.791932
I
vandanavk commented on issue #12783: Name conflict when serializing LSTMCell
URL:
https://github.com/apache/incubator-mxnet/issues/12783#issuecomment-429097068
Similar error occurred in
https://github.com/apache/incubator-mxnet/issues/11542
piyushghai commented on issue #12802: Crash while loading libmxnet-scala.so
URL:
https://github.com/apache/incubator-mxnet/issues/12802#issuecomment-429097144
Thank you for reporting this issue. We will look into it.
@mxnet-label-bot [Scala, Breaking]
---
zhanghang1989 opened a new pull request #12803: Fix Extremely Unbalanced Issue
in split_and_load
URL: https://github.com/apache/incubator-mxnet/pull/12803
## Description ##
Existing implementation:
```python
import mxnet as mx
x = mx.nd.ones((38, 3, 4, 4))
ctx = [mx.gpu(i) fo
1 - 100 of 175 matches
Mail list logo