zheng-da commented on a change in pull request #11498: Fix InferStorage for
sparse fallback in FullyConnected
URL: https://github.com/apache/incubator-mxnet/pull/11498#discussion_r199313642
##
File path: src/operator/nn/fully_connected.cc
##
@@ -210,17 +210,17 @@ inline
yukang2017 opened a new issue #11508: 【Question】Are there any examples for
gradients clipping in gluon?
URL: https://github.com/apache/incubator-mxnet/issues/11508
This is what I guess. Is it right?
trainer.allreduce_grads()
with autograd.record():
szha commented on a change in pull request #11134: [MXNET-507] Set dtype=int32
for ret_indices in ordering ops
URL: https://github.com/apache/incubator-mxnet/pull/11134#discussion_r199310908
##
File path: src/operator/tensor/ordering_op-inl.h
##
@@ -79,6 +80,16 @@ struct
This is an automated email from the ASF dual-hosted git repository.
zhasheng 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 9a0099f [MXNET-595] Install page
szha closed pull request #11404: [MXNET-595] Install page not loading properly
in Internet Explorer
URL: https://github.com/apache/incubator-mxnet/pull/11404
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
szha closed pull request #11405: Update large word language model example
URL: https://github.com/apache/incubator-mxnet/pull/11405
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.
zhasheng 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 806b41b Update large word language
marcoabreu opened a new pull request #11507: Disable flaky test test_conv in
gluon
URL: https://github.com/apache/incubator-mxnet/pull/11507
## Description ##
Seems like test_conv is actually broken as well...
https://github.com/apache/incubator-mxnet/issues/11506
## Checklist
marcoabreu opened a new issue #11506: Flaky test_operator_gpu.test_conv causes
other tests to fail on Windows
URL: https://github.com/apache/incubator-mxnet/issues/11506
http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/master/1106/pipeline
```
marcoabreu commented on a change in pull request #10696: [MXNET-366]Extend
MXNet Distributed Training by AllReduce
URL: https://github.com/apache/incubator-mxnet/pull/10696#discussion_r199309383
##
File path: tests/nightly/test_all.sh
##
@@ -44,6 +44,7 @@ USE_CUDA=1
marcoabreu edited a comment on issue #11475: Amending ONNX importer/exporter
#11213
URL:
https://github.com/apache/incubator-mxnet/issues/11475#issuecomment-401508772
We currently have the following structure:
```
tests/python-pytest
tests/python
tests/cpp
```
The
marcoabreu commented on issue #11475: Amending ONNX importer/exporter #11213
URL:
https://github.com/apache/incubator-mxnet/issues/11475#issuecomment-401508772
We currently have the following structure:
```
tests/python-pytest
tests/python
tests/cpp
```
The current
aaronmarkham edited a comment on issue #11503: updating installation info to
have latest packages and more clarity
URL: https://github.com/apache/incubator-mxnet/pull/11503#issuecomment-401507051
The table is rendering incorrectly, so I'll need to fix that first. It is
supposed to look
vrakesh opened a new pull request #11505: [MXNET-615] Fix for flaky
test_svmoutput_with_type
URL: https://github.com/apache/incubator-mxnet/pull/11505
## Description ##
Fixes flaky test #8288
## Checklist ##
### Essentials ###
Please feel free to remove inapplicable
eric-haibin-lin closed issue #10920: Flaky test
test_operator_gpu.test_sparse_dot
URL: https://github.com/apache/incubator-mxnet/issues/10920
This is an automated message from the Apache Git Service.
To respond to the
aaronmarkham commented on issue #11503: updating installation info to have
latest packages and more clarity
URL: https://github.com/apache/incubator-mxnet/pull/11503#issuecomment-401507051
The table is rendering incorrectly, so I'll need to fix that first. It is
supposed to look like
aaronmarkham removed a comment on issue #11503: updating installation info to
have latest packages and more clarity
URL: https://github.com/apache/incubator-mxnet/pull/11503#issuecomment-401503263
@nswamy @marcoabreu @szha - Can you please review?
eric-haibin-lin closed pull request #10889: [MXNET-382] Shape and Size Operator
URL: https://github.com/apache/incubator-mxnet/pull/10889
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
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 33022f8 [MXNET-382] Shape and Size
rahul003 commented on a change in pull request #11406: [MXNET-599] Partial
shape infer for Slice
URL: https://github.com/apache/incubator-mxnet/pull/11406#discussion_r199307912
##
File path: src/operator/tensor/matrix_op-inl.h
##
@@ -674,13 +675,23 @@ inline void
aaronmarkham commented on issue #11152: updating Scala IntelliJ tutorial &
installation instructions
URL: https://github.com/apache/incubator-mxnet/pull/11152#issuecomment-401506350
@nswamy I've addressed all of your concerns. Please confirm/merge. Thanks!
roywei commented on issue #8044: Fix the Test
test_operator_gpu.test_batchnorm_training
URL:
https://github.com/apache/incubator-mxnet/issues/8044#issuecomment-401506153
it may due to cuDNN implementation in BatchNorm.
changed
```
numeric_eps=1e-3
atol=1e-3
```
running
rahul003 commented on a change in pull request #11406: [MXNET-599] Partial
shape infer for Slice
URL: https://github.com/apache/incubator-mxnet/pull/11406#discussion_r199307425
##
File path: src/operator/tensor/matrix_op-inl.h
##
@@ -690,9 +701,10 @@ inline bool
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 e00cc55 Bump the publish
anirudhacharya commented on issue #10756: L1 Normalization operator not present
in MXNet
URL:
https://github.com/apache/incubator-mxnet/issues/10756#issuecomment-401504206
https://github.com/apache/incubator-mxnet/pull/11229
anirudhacharya closed issue #10756: L1 Normalization operator not present in
MXNet
URL: https://github.com/apache/incubator-mxnet/issues/10756
This is an automated message from the Apache Git Service.
To respond to the
anirudhacharya commented on issue #10889: [MXNET-382] Shape and Size Operator
URL: https://github.com/apache/incubator-mxnet/pull/10889#issuecomment-401504000
@piiswrong @reminisce please merge this.
This is an automated
vrakesh commented on issue #8288: test_svmoutput_with_type fails in CI builds
URL:
https://github.com/apache/incubator-mxnet/issues/8288#issuecomment-401503518
The above script from @anirudh2290 , was expanded with a test for gpu as well
```python
import mxnet as mx
import numpy
aaronmarkham commented on issue #11503: updating installation info to have
latest packages and more clarity
URL: https://github.com/apache/incubator-mxnet/pull/11503#issuecomment-401503263
@nswamy @marcoabreu @szha - Can you please review?
aaronmarkham opened a new pull request #11503: updating installation info to
have latest packages and more clarity
URL: https://github.com/apache/incubator-mxnet/pull/11503
## Description ##
There hasn't been an update on installation detail instructions in a while.
Many new
eric-haibin-lin closed pull request #11229: [MXNET-379] L1 Normalization
URL: https://github.com/apache/incubator-mxnet/pull/11229
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 f783a66 [MXNET-379] L1 Normalization
szha commented on issue #11482: make gluon rnn layers hybrid blocks
URL: https://github.com/apache/incubator-mxnet/pull/11482#issuecomment-401501912
I can override forward but it would be pretty much equivalent. The reason I
cannot do this in hybrid_forward is that when given partial
anirudhacharya commented on issue #11229: [MXNET-379] L1 Normalization
URL: https://github.com/apache/incubator-mxnet/pull/11229#issuecomment-401501298
@eric-haibin-lin can you please merge this.
This is an automated message
ctcyang commented on a change in pull request #10696: [MXNET-366]Extend MXNet
Distributed Training by AllReduce
URL: https://github.com/apache/incubator-mxnet/pull/10696#discussion_r199300072
##
File path: src/kvstore/collectives/src/collectives.cc
##
@@ -0,0 +1,779 @@
zhanghang1989 opened a new pull request #11502: [MXNET-614] Adding Synchronized
Batch Normalization
URL: https://github.com/apache/incubator-mxnet/pull/11502
## Description ##
Adding Synchronized Batch Normalization
## Checklist ##
### Essentials ###
Please feel free to
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 a937c7ce Bump the publish
piiswrong commented on issue #11482: make gluon rnn layers hybrid blocks
URL: https://github.com/apache/incubator-mxnet/pull/11482#issuecomment-401497076
I see. Then could you do it without overriding `__call__`?
This is
szha closed pull request #11287: [MXNET-548] fixed path for auto_module_index.js
URL: https://github.com/apache/incubator-mxnet/pull/11287
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
szha closed issue #11238: UX for ONNX Documentation is broken
URL: https://github.com/apache/incubator-mxnet/issues/11238
This is an automated message from the Apache Git Service.
To respond to the message, please log on
This is an automated email from the ASF dual-hosted git repository.
zhasheng 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 355de66 [MXNET-548] fixed path for
szha commented on issue #11004: Only allocate cudnn-rnn dropout memory if
dropout p > 0 and acquire descriptors during initialization
URL: https://github.com/apache/incubator-mxnet/pull/11004#issuecomment-401494044
@sxjscience see above. This seems like a good justification for reusing the
eric-haibin-lin commented on issue #11004: Only allocate cudnn-rnn dropout
memory if dropout p > 0 and acquire descriptors during initialization
URL: https://github.com/apache/incubator-mxnet/pull/11004#issuecomment-401493149
According to @safrooze this PR makes the RNN cells in his model
anirudh2290 commented on issue #11478: 1.2.1 release notes
URL: https://github.com/apache/incubator-mxnet/pull/11478#issuecomment-401492746
@srochel your changes have been incorporated.
This is an automated message from the
apeforest commented on a change in pull request #11466: [MXNET-560] Add
temperature parameter in Softmax operator
URL: https://github.com/apache/incubator-mxnet/pull/11466#discussion_r199296955
##
File path: src/operator/nn/softmax-inl.h
##
@@ -53,7 +53,7 @@ struct
roywei commented on issue #8044: Fix the Test
test_operator_gpu.test_batchnorm_training
URL:
https://github.com/apache/incubator-mxnet/issues/8044#issuecomment-401490912
issue causing the ctx error is tracked at #11448
working on flakiness:
tried changing numeric_eps=1e-3
using
gigasquid commented on a change in pull request #11205: Clojure Contrib Package
URL: https://github.com/apache/incubator-mxnet/pull/11205#discussion_r199295055
##
File path: contrib/clojure-package/src/org/apache/clojure_mxnet/gen/ndarray.clj
##
@@ -0,0 +1,2312 @@
+(ns
zhreshold closed pull request #11145: [MXNET-517] add sample ratio for ROI Align
URL: https://github.com/apache/incubator-mxnet/pull/11145
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
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 e892301 [MXNET-517] add sample
zhreshold closed issue #11077: The sampling_ratio in roi-align can be
specified(not adaptive size)!
URL: https://github.com/apache/incubator-mxnet/issues/11077
This is an automated message from the Apache Git Service.
To
frankfliu commented on issue #11496: Selu Operator not present in MXNet
URL:
https://github.com/apache/incubator-mxnet/issues/11496#issuecomment-401485460
Hi @anirudhacharya, thanks for submitting issue. @sandeep-krishnamurthy
requesting this be labeled.
frankfliu commented on issue #11501: Tracking issue for tests with fixed random
seed
URL:
https://github.com/apache/incubator-mxnet/issues/11501#issuecomment-401485304
Thank you for submitting the issue! @sandeep-krishnamurthy requesting this
be labeled under feature request.
vdantu commented on issue #9864: Flaky hanging test_operator.test_laop_3
URL:
https://github.com/apache/incubator-mxnet/issues/9864#issuecomment-401485090
@marcoabreu @KellenSunderland : Please suggest how we could go about
narrowing this down and resolve this.
frankfliu commented on issue #11495: Bug for matrices of multiple dimension,
with one dimension much larger
URL:
https://github.com/apache/incubator-mxnet/issues/11495#issuecomment-401485161
Hi @altosaar, thanks for submitting issue. @sandeep-krishnamurthy requesting
this be labeled.
szha commented on issue #11482: make gluon rnn layers hybrid blocks
URL: https://github.com/apache/incubator-mxnet/pull/11482#issuecomment-401484871
@piiswrong turns out I cannot do that only in infer_shape. The reason is
that sometimes the block is used as a child block of other blocks,
frankfliu commented on issue #11497: Feature request: where operator should
support scalar operand
URL:
https://github.com/apache/incubator-mxnet/issues/11497#issuecomment-401484809
Thank you for submitting the request! We are labeling it so MXNet community
members can help resolve it.
haojin2 commented on issue #10920: Flaky test test_operator_gpu.test_sparse_dot
URL:
https://github.com/apache/incubator-mxnet/issues/10920#issuecomment-401483849
@eric-haibin-lin should be fixed at this time.
This is an
ankkhedia edited a comment on issue #9116: unit test
test_ndarray.test_ndarray_indexing fails
URL:
https://github.com/apache/incubator-mxnet/issues/9116#issuecomment-399574168
Couldn't reproduce the issue for 1 runs on the setup mentioned in the
issue.
@zheng-da
lanking520 commented on a change in pull request #11205: Clojure Contrib Package
URL: https://github.com/apache/incubator-mxnet/pull/11205#discussion_r199287183
##
File path: contrib/clojure-package/src/org/apache/clojure_mxnet/gen/ndarray.clj
##
@@ -0,0 +1,2312 @@
+(ns
azai91 commented on issue #11466: [MXNET-560] Add temperature parameter in
Softmax operator
URL: https://github.com/apache/incubator-mxnet/pull/11466#issuecomment-401480335
@pengzhao-intel @zheng-da do you know if MKLDNN supports the temperature
parameter? If not then we will just have to
anooprh closed issue #11500: Unable to build docker image locally
URL: https://github.com/apache/incubator-mxnet/issues/11500
This is an automated message from the Apache Git Service.
To respond to the message, please log on
anooprh commented on issue #11500: Unable to build docker image locally
URL:
https://github.com/apache/incubator-mxnet/issues/11500#issuecomment-401476755
My bad. This issue got fixed after I allocated more memory to docker (4GB)
and more swap(2GB). Closing the issue.
szha commented on a change in pull request #11482: make gluon rnn layers hybrid
blocks
URL: https://github.com/apache/incubator-mxnet/pull/11482#discussion_r199282336
##
File path: python/mxnet/gluon/rnn/rnn_layer.py
##
@@ -173,67 +177,63 @@ def begin_state(self,
szha commented on a change in pull request #11482: make gluon rnn layers hybrid
blocks
URL: https://github.com/apache/incubator-mxnet/pull/11482#discussion_r199281972
##
File path: python/mxnet/gluon/rnn/rnn_layer.py
##
@@ -173,67 +177,63 @@ def begin_state(self,
szha commented on a change in pull request #11482: make gluon rnn layers hybrid
blocks
URL: https://github.com/apache/incubator-mxnet/pull/11482#discussion_r199281972
##
File path: python/mxnet/gluon/rnn/rnn_layer.py
##
@@ -173,67 +177,63 @@ def begin_state(self,
szha commented on a change in pull request #11482: make gluon rnn layers hybrid
blocks
URL: https://github.com/apache/incubator-mxnet/pull/11482#discussion_r199281929
##
File path: python/mxnet/gluon/rnn/rnn_layer.py
##
@@ -173,67 +177,63 @@ def begin_state(self,
aaronmarkham commented on issue #11287: [MXNET-548] fixed path for
auto_module_index.js
URL: https://github.com/apache/incubator-mxnet/pull/11287#issuecomment-401475808
@szha rebased and passed!
This is an automated message
zhreshold closed pull request #11434: add ignore_reinit to initialize to skip
warnings
URL: https://github.com/apache/incubator-mxnet/pull/11434
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
szha closed pull request #11499: Test for new int64 type in CSVIter
URL: https://github.com/apache/incubator-mxnet/pull/11499
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
lupesko commented on issue #11463: Removing FlakyTest: memory_test.cc
URL: https://github.com/apache/incubator-mxnet/pull/11463#issuecomment-401473248
Adding @cjolivier01 who is the author of these tests.
This is an automated
haojin2 opened a new issue #11501: Tracking issue for tests with fixed random
seed
URL: https://github.com/apache/incubator-mxnet/issues/11501
@eric-haibin-lin and @azai91 discovered that some unit tests have been using
fixed seed to mask flakiness, opening this issue to keep track of
anirudh2290 commented on issue #11470: Fix build issue with USE_CUDNN=0
URL: https://github.com/apache/incubator-mxnet/pull/11470#issuecomment-401468703
@szha Yep, planning to fix that.
This is an automated message from the
haojin2 commented on a change in pull request #10889: [MXNET-382] Shape and
Size Operator
URL: https://github.com/apache/incubator-mxnet/pull/10889#discussion_r199275013
##
File path: src/operator/tensor/elemwise_unary_op_basic.cc
##
@@ -398,6 +398,98 @@
zhreshold commented on issue #11434: add ignore_reinit to initialize to skip
warnings
URL: https://github.com/apache/incubator-mxnet/pull/11434#issuecomment-401468140
Yes, I agree that it’s not a good practice to add an option to disable
warning.
However in this particular case, if
piiswrong commented on a change in pull request #11482: make gluon rnn layers
hybrid blocks
URL: https://github.com/apache/incubator-mxnet/pull/11482#discussion_r199273106
##
File path: python/mxnet/gluon/rnn/rnn_layer.py
##
@@ -173,67 +177,63 @@ def begin_state(self,
piiswrong commented on a change in pull request #11482: make gluon rnn layers
hybrid blocks
URL: https://github.com/apache/incubator-mxnet/pull/11482#discussion_r199273220
##
File path: python/mxnet/gluon/rnn/rnn_layer.py
##
@@ -173,67 +177,63 @@ def begin_state(self,
anirudhacharya edited a comment on issue #10889: [MXNET-382] Shape and Size
Operator
URL: https://github.com/apache/incubator-mxnet/pull/10889#issuecomment-401413402
@haojin2 @piiswrong can this be merged?
This is an
haojin2 commented on a change in pull request #11466: [MXNET-560] Add
temperature parameter in Softmax operator
URL: https://github.com/apache/incubator-mxnet/pull/11466#discussion_r199268467
##
File path: src/operator/nn/softmax-inl.h
##
@@ -53,7 +53,7 @@ struct
anooprh opened a new issue #11500: Unable to build docker image locally
URL: https://github.com/apache/incubator-mxnet/issues/11500
Note: Providing complete information in the most concise form is the best
way to get help. This issue template serves as the checklist for essential
szha commented on issue #11499: Test for new int64 type in CSVIter
URL: https://github.com/apache/incubator-mxnet/pull/11499#issuecomment-401452682
@haojin2 thanks for adding the test for #11446
This is an automated message
haojin2 commented on issue #11499: Test for new int64 type in CSVIter
URL: https://github.com/apache/incubator-mxnet/pull/11499#issuecomment-401448212
@szha
This is an automated message from the Apache Git Service.
To
haojin2 opened a new pull request #11499: Test for new int64 type in CSVIter
URL: https://github.com/apache/incubator-mxnet/pull/11499
## Description ##
Previous int64 data type for CSVIter PR was merged but forgot a special test
for it.
## Checklist ##
### Essentials ###
-
zhanghang1989 commented on issue #11145: [MXNET-517] add sample ratio for ROI
Align
URL: https://github.com/apache/incubator-mxnet/pull/11145#issuecomment-401447551
Any updates or comments @piiswrong
This is an automated
This is an automated email from the ASF dual-hosted git repository.
zhasheng 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 3a62200 support int64 data type in
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 a6d2a6c Bump the publish
This is an automated email from the ASF dual-hosted git repository.
zhasheng 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 275cd8e update docs: deps using CI
gigasquid commented on a change in pull request #11205: Clojure Contrib Package
URL: https://github.com/apache/incubator-mxnet/pull/11205#discussion_r199245552
##
File path: contrib/clojure-package/src/org/apache/clojure_mxnet/gen/ndarray.clj
##
@@ -0,0 +1,2312 @@
+(ns
haojin2 commented on issue #10988: Flaky test:
test_operator_gpu.test_countsketch
URL:
https://github.com/apache/incubator-mxnet/issues/10988#issuecomment-401435901
From the reproduced error we can see that only part of the grad ndarray is
filled:
```
ctcyang edited a comment on issue #10696: [MXNET-366]Extend MXNet Distributed
Training by AllReduce
URL: https://github.com/apache/incubator-mxnet/pull/10696#issuecomment-401432358
Thanks for your help @threeleafzerg I was able to build it with
`USE_ALLREDUCE_DIST_KVSTORE = 1`. On AWS EC2
ctcyang edited a comment on issue #10696: [MXNET-366]Extend MXNet Distributed
Training by AllReduce
URL: https://github.com/apache/incubator-mxnet/pull/10696#issuecomment-401432358
Thanks for your help @threeleafzerg I was able to build it with
USE_ALLREDUCE_DIST_KVSTORE = 1. On AWS EC2
ctcyang edited a comment on issue #10696: [MXNET-366]Extend MXNet Distributed
Training by AllReduce
URL: https://github.com/apache/incubator-mxnet/pull/10696#issuecomment-401432358
Thanks for your help @threeleafzerg I was able to build it with
USE_ALLREDUCE_DIST_KVSTORE = 1. On AWS EC2
ctcyang edited a comment on issue #10696: [MXNET-366]Extend MXNet Distributed
Training by AllReduce
URL: https://github.com/apache/incubator-mxnet/pull/10696#issuecomment-401432358
Thanks for your help @threeleafzerg I was able to build it with
USE_ALLREDUCE_DIST_KVSTORE = 1. On AWS EC2
ctcyang edited a comment on issue #10696: [MXNET-366]Extend MXNet Distributed
Training by AllReduce
URL: https://github.com/apache/incubator-mxnet/pull/10696#issuecomment-401432358
Thanks for your help @threeleafzerg I was able to build it with
USE_ALLREDUCE_DIST_KVSTORE = 1. On AWS EC2
ctcyang edited a comment on issue #10696: [MXNET-366]Extend MXNet Distributed
Training by AllReduce
URL: https://github.com/apache/incubator-mxnet/pull/10696#issuecomment-401432358
Thanks for your help @threeleafzerg I was able to build it with
USE_ALLREDUCE_DIST_KVSTORE = 1. On AWS EC2
eric-haibin-lin commented on a change in pull request #11498: Fix InferStorage
for sparse fallback in FullyConnected
URL: https://github.com/apache/incubator-mxnet/pull/11498#discussion_r199239656
##
File path: src/operator/nn/fully_connected.cc
##
@@ -210,17 +210,17 @@
ctcyang commented on issue #10696: [MXNET-366]Extend MXNet Distributed Training
by AllReduce
URL: https://github.com/apache/incubator-mxnet/pull/10696#issuecomment-401432358
Thanks for your help @threeleafzerg I was able to build it with
USE_ALLREDUCE_DIST_KVSTORE=1. On AWS EC2 instances,
yzhliu commented on a change in pull request #11205: Clojure Contrib Package
URL: https://github.com/apache/incubator-mxnet/pull/11205#discussion_r199239591
##
File path: contrib/clojure-package/src/org/apache/clojure_mxnet/gen/ndarray.clj
##
@@ -0,0 +1,2312 @@
+(ns
eric-haibin-lin commented on a change in pull request #10696: [MXNET-366]Extend
MXNet Distributed Training by AllReduce
URL: https://github.com/apache/incubator-mxnet/pull/10696#discussion_r199239103
##
File path: tests/nightly/test_all.sh
##
@@ -44,6 +44,7 @@ USE_CUDA=1
szha commented on issue #11429: Improve sparse pull performance for gluon
trainer
URL: https://github.com/apache/incubator-mxnet/pull/11429#issuecomment-401430706
how does the performance look? @eric-haibin-lin @leezu
This
szha commented on a change in pull request #11429: Improve sparse pull
performance for gluon trainer
URL: https://github.com/apache/incubator-mxnet/pull/11429#discussion_r199237863
##
File path: python/mxnet/gluon/trainer.py
##
@@ -169,9 +190,9 @@ def _init_kvstore(self):
1 - 100 of 243 matches
Mail list logo