Alicia1529 opened a new pull request #17390: [Numpy]add op random.beta
URL: https://github.com/apache/incubator-mxnet/pull/17390
add np.random.beta
This is an automated message from the Apache Git Service.
To respond to the me
reminisce opened a new issue #17389: Indexing ndarray should give empty array
output
URL: https://github.com/apache/incubator-mxnet/issues/17389
## Description
When indexing an ndarray with an index greater than the dim size, it should
produce an empty ndarray as in NumPy, instead of th
This is an automated email from the ASF dual-hosted git repository.
aaronmarkham 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 2738d10 Bump the publis
rongzha1 commented on a change in pull request #17170: add mkldnn softmax
backward
URL: https://github.com/apache/incubator-mxnet/pull/17170#discussion_r368829691
##
File path: src/operator/nn/mkldnn/mkldnn_softmax.cc
##
@@ -131,6 +143,80 @@ void MKLDNNSoftmaxForward(cons
KellenSunderland commented on issue #15547: Pybind11 license issue from onnx
tensorrt
URL:
https://github.com/apache/incubator-mxnet/issues/15547#issuecomment-576528385
I'd also propose we do this in the 1.7 release.
This is
eric-haibin-lin commented on a change in pull request #15124: [MXNET-1294]
Priority-based parameter propagation for improved data parallel training
throughput
URL: https://github.com/apache/incubator-mxnet/pull/15124#discussion_r368809705
##
File path: src/kvstore/p3store.h
##
eric-haibin-lin commented on a change in pull request #15124: [MXNET-1294]
Priority-based parameter propagation for improved data parallel training
throughput
URL: https://github.com/apache/incubator-mxnet/pull/15124#discussion_r368813696
##
File path: python/mxnet/kvstore/kvstore.
eric-haibin-lin commented on a change in pull request #15124: [MXNET-1294]
Priority-based parameter propagation for improved data parallel training
throughput
URL: https://github.com/apache/incubator-mxnet/pull/15124#discussion_r368811146
##
File path: include/mxnet/c_api.h
##
eric-haibin-lin commented on a change in pull request #15124: [MXNET-1294]
Priority-based parameter propagation for improved data parallel training
throughput
URL: https://github.com/apache/incubator-mxnet/pull/15124#discussion_r368807494
##
File path: docs/static_site/src/pages/ap
eric-haibin-lin commented on a change in pull request #15124: [MXNET-1294]
Priority-based parameter propagation for improved data parallel training
throughput
URL: https://github.com/apache/incubator-mxnet/pull/15124#discussion_r368814411
##
File path: python/mxnet/kvstore/kvstore.
eric-haibin-lin commented on a change in pull request #15124: [MXNET-1294]
Priority-based parameter propagation for improved data parallel training
throughput
URL: https://github.com/apache/incubator-mxnet/pull/15124#discussion_r368807735
##
File path: python/mxnet/kvstore/base.py
eric-haibin-lin commented on a change in pull request #15124: [MXNET-1294]
Priority-based parameter propagation for improved data parallel training
throughput
URL: https://github.com/apache/incubator-mxnet/pull/15124#discussion_r368809867
##
File path: src/kvstore/p3store.h
##
eric-haibin-lin commented on a change in pull request #15124: [MXNET-1294]
Priority-based parameter propagation for improved data parallel training
throughput
URL: https://github.com/apache/incubator-mxnet/pull/15124#discussion_r368809460
##
File path: src/kvstore/p3store.h
##
eric-haibin-lin commented on a change in pull request #15124: [MXNET-1294]
Priority-based parameter propagation for improved data parallel training
throughput
URL: https://github.com/apache/incubator-mxnet/pull/15124#discussion_r368808125
##
File path: python/mxnet/kvstore/kvstore.
eric-haibin-lin commented on a change in pull request #15124: [MXNET-1294]
Priority-based parameter propagation for improved data parallel training
throughput
URL: https://github.com/apache/incubator-mxnet/pull/15124#discussion_r368809279
##
File path: src/kvstore/p3store.h
##
eric-haibin-lin commented on a change in pull request #15124: [MXNET-1294]
Priority-based parameter propagation for improved data parallel training
throughput
URL: https://github.com/apache/incubator-mxnet/pull/15124#discussion_r368813200
##
File path: python/mxnet/kvstore/base.py
eric-haibin-lin commented on a change in pull request #15124: [MXNET-1294]
Priority-based parameter propagation for improved data parallel training
throughput
URL: https://github.com/apache/incubator-mxnet/pull/15124#discussion_r368813825
##
File path: python/mxnet/kvstore/kvstore.
vexilligera commented on a change in pull request #17387: [TVM][Windows] TVM
Ops windows build
URL: https://github.com/apache/incubator-mxnet/pull/17387#discussion_r368794934
##
File path: CMakeLists.txt
##
@@ -134,7 +134,7 @@ if(MSVC)
if(USE_F16C)
message("F16C in
vexilligera commented on a change in pull request #17387: [TVM][Windows] TVM
Ops windows build
URL: https://github.com/apache/incubator-mxnet/pull/17387#discussion_r368794934
##
File path: CMakeLists.txt
##
@@ -134,7 +134,7 @@ if(MSVC)
if(USE_F16C)
message("F16C in
vexilligera commented on a change in pull request #17014: [NumPy] Add NumPy
support for norm
URL: https://github.com/apache/incubator-mxnet/pull/17014#discussion_r368790115
##
File path: src/operator/numpy/linalg/np_norm-inl.h
##
@@ -0,0 +1,830 @@
+/*
+ * Licensed to the A
vexilligera commented on a change in pull request #17014: [NumPy] Add NumPy
support for norm
URL: https://github.com/apache/incubator-mxnet/pull/17014#discussion_r368790115
##
File path: src/operator/numpy/linalg/np_norm-inl.h
##
@@ -0,0 +1,830 @@
+/*
+ * Licensed to the A
leezu commented on a change in pull request #17387: [TVM][Windows] TVM Ops
windows build
URL: https://github.com/apache/incubator-mxnet/pull/17387#discussion_r368792178
##
File path: CMakeLists.txt
##
@@ -419,7 +419,7 @@ if(USE_OPENMP)
# could a Windows user please fix?
vexilligera commented on a change in pull request #17014: [NumPy] Add NumPy
support for norm
URL: https://github.com/apache/incubator-mxnet/pull/17014#discussion_r368791962
##
File path: src/operator/numpy/linalg/broadcast_reduce_customized-inl.cuh
##
@@ -0,0 +1,416 @@
+/*
vexilligera commented on a change in pull request #17014: [NumPy] Add NumPy
support for norm
URL: https://github.com/apache/incubator-mxnet/pull/17014#discussion_r368790115
##
File path: src/operator/numpy/linalg/np_norm-inl.h
##
@@ -0,0 +1,830 @@
+/*
+ * Licensed to the A
access2rohit commented on issue #17388: skipping randint flaky test for large
vector and reordering op execution
URL: https://github.com/apache/incubator-mxnet/pull/17388#issuecomment-576485523
@ChaiBapchya @apeforest PR ready for review
---
access2rohit edited a comment on issue #17388: skipping randint flaky test for
large vector and reordering op execution
URL: https://github.com/apache/incubator-mxnet/pull/17388#issuecomment-576485460
@mxnet-label-bot add [pr-awaiting-review]
--
access2rohit commented on issue #17388: skipping randint flaky test for large
vector and reordering op execution
URL: https://github.com/apache/incubator-mxnet/pull/17388#issuecomment-576485460
@mxnet-label-bot add [pre-awaiting-review]
access2rohit opened a new pull request #17388: skipping randint flaky test for
large vector and reordering op execution
URL: https://github.com/apache/incubator-mxnet/pull/17388
## Description ##
Disables flaky test randint since natively none of the randing APIs used
internally suppor
haojin2 commented on issue #17384: Error in numpy unit test
URL:
https://github.com/apache/incubator-mxnet/issues/17384#issuecomment-576485062
I have 1.18.1, I'll try with my numpy downgraded.
This is an automated message fro
heaseny commented on issue #17384: Error in numpy unit test
URL:
https://github.com/apache/incubator-mxnet/issues/17384#issuecomment-576484737
The test can fail with numpy==1.16.4, so should I upgrade my numpy?
This is an au
vexilligera commented on a change in pull request #17387: [TVM][Windows] TVM
Ops windows build
URL: https://github.com/apache/incubator-mxnet/pull/17387#discussion_r368778676
##
File path: CMakeLists.txt
##
@@ -134,7 +134,7 @@ if(MSVC)
if(USE_F16C)
message("F16C in
vexilligera commented on a change in pull request #17387: [TVM][Windows] TVM
Ops windows build
URL: https://github.com/apache/incubator-mxnet/pull/17387#discussion_r368778573
##
File path: CMakeLists.txt
##
@@ -419,7 +419,7 @@ if(USE_OPENMP)
# could a Windows user pleas
vexilligera commented on a change in pull request #17387: [TVM][Windows] TVM
Ops windows build
URL: https://github.com/apache/incubator-mxnet/pull/17387#discussion_r368778350
##
File path: python/mxnet/space.py
##
@@ -1 +1,214 @@
-../../contrib/tvmop/space.py
Review com
rondogency commented on issue #17382: document of lib_custom_op support gpu
without rebuilding mxnet
URL:
https://github.com/apache/incubator-mxnet/issues/17382#issuecomment-576471945
Hi @a550461053, the lib_custom_op in c++ doesn't support GPU for now, and we
are working on both c++ cust
This is an automated email from the ASF dual-hosted git repository.
aaronmarkham 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 2f9471d Bump the publis
leezu edited a comment on issue #17121: Include 3rdparty/jemalloc (Fix for
Ubuntu 18.10+, Debian Stable, etc.)
URL: https://github.com/apache/incubator-mxnet/pull/17121#issuecomment-576464949
Close this until stable upstream. See
https://github.com/jemalloc/jemalloc/issues/303.
http
leezu commented on issue #17121: Include 3rdparty/jemalloc (Fix for Ubuntu
18.10+, Debian Stable, etc.)
URL: https://github.com/apache/incubator-mxnet/pull/17121#issuecomment-576464949
Close this until stable upstream. See
https://github.com/jemalloc/jemalloc/issues/303.
-
leezu closed pull request #17121: Include 3rdparty/jemalloc (Fix for Ubuntu
18.10+, Debian Stable, etc.)
URL: https://github.com/apache/incubator-mxnet/pull/17121
This is an automated message from the Apache Git Service.
To
larroy edited a comment on issue #16995: CI timeout on unix-cpu Python2 test
URL:
https://github.com/apache/incubator-mxnet/issues/16995#issuecomment-576458573
Thanks the team will look into it in the following days.
This is
larroy commented on issue #16995: CI timeout on unix-cpu Python2 test
URL:
https://github.com/apache/incubator-mxnet/issues/16995#issuecomment-576458573
Thanks the team will look into this next week.
This is an automated mess
haojin2 commented on issue #17384: Error in numpy unit test
URL:
https://github.com/apache/incubator-mxnet/issues/17384#issuecomment-576454166
Hi, I'm not able to reproduce with latest master branch, which branch were
you using? and what kind of build are you using?
```
nosetests -s
haojin2 commented on issue #17317: Floor division operator "//" in deepnumpy
URL:
https://github.com/apache/incubator-mxnet/issues/17317#issuecomment-576450916
working on it
This is an automated message from the Apache Git Se
This is an automated email from the ASF dual-hosted git repository.
lausen pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 28742cf [numpy] add op random.exponential (#17280)
add 425319c Set USE_JEMALLOC=OFF by default due to
This is an automated email from the ASF dual-hosted git repository.
lausen pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 28742cf [numpy] add op random.exponential (#17280)
add 425319c Set USE_JEMALLOC=OFF by default due to
leezu merged pull request #17324: USE_JEMALLOC=OFF by default
URL: https://github.com/apache/incubator-mxnet/pull/17324
This is an automated message from the Apache Git Service.
To respond to the message, please log on to Git
leezu commented on a change in pull request #17387: [TVM][Windows] TVM Ops
windows build
URL: https://github.com/apache/incubator-mxnet/pull/17387#discussion_r368691253
##
File path: python/mxnet/space.py
##
@@ -1 +1,214 @@
-../../contrib/tvmop/space.py
Review comment:
leezu commented on a change in pull request #17387: [TVM][Windows] TVM Ops
windows build
URL: https://github.com/apache/incubator-mxnet/pull/17387#discussion_r368690456
##
File path: CMakeLists.txt
##
@@ -419,7 +419,7 @@ if(USE_OPENMP)
# could a Windows user please fix?
leezu commented on a change in pull request #17387: [TVM][Windows] TVM Ops
windows build
URL: https://github.com/apache/incubator-mxnet/pull/17387#discussion_r368690375
##
File path: CMakeLists.txt
##
@@ -134,7 +134,7 @@ if(MSVC)
if(USE_F16C)
message("F16C instruct
This is an automated email from the ASF dual-hosted git repository.
aaronmarkham 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 d66628f Bump the publis
sxjscience commented on issue #17317: Floor division operator "//" in deepnumpy
URL:
https://github.com/apache/incubator-mxnet/issues/17317#issuecomment-576372804
@haojin2 @reminisce
This is an automated message from the Apa
sxjscience closed issue #17303: New Basic Layer "nn.bias_add"
URL: https://github.com/apache/incubator-mxnet/issues/17303
This is an automated message from the Apache Git Service.
To respond to the message, please log on to G
sxjscience commented on issue #17303: New Basic Layer "nn.bias_add"
URL:
https://github.com/apache/incubator-mxnet/issues/17303#issuecomment-576364762
Given that we can use broadcasting, it should not be necessary to introduce
`bias_add`. Also, in numpy, we can directly use the auto-broadc
xidulu commented on a change in pull request #17385: [NumPy] add
random.geometric op
URL: https://github.com/apache/incubator-mxnet/pull/17385#discussion_r368596365
##
File path: src/operator/numpy/random/np_geometric_op.h
##
@@ -0,0 +1,177 @@
+/*
+ * Licensed to the Apach
xidulu commented on a change in pull request #17385: [NumPy] add
random.geometric op
URL: https://github.com/apache/incubator-mxnet/pull/17385#discussion_r368595176
##
File path: src/operator/numpy/random/np_geometric_op.h
##
@@ -0,0 +1,177 @@
+/*
+ * Licensed to the Apach
xidulu commented on a change in pull request #17385: [NumPy] add
random.geometric op
URL: https://github.com/apache/incubator-mxnet/pull/17385#discussion_r368595176
##
File path: src/operator/numpy/random/np_geometric_op.h
##
@@ -0,0 +1,177 @@
+/*
+ * Licensed to the Apach
xidulu commented on a change in pull request #17385: [NumPy] add
random.geometric op
URL: https://github.com/apache/incubator-mxnet/pull/17385#discussion_r368594348
##
File path: src/operator/numpy/random/np_geometric_op.h
##
@@ -0,0 +1,177 @@
+/*
+ * Licensed to the Apach
xidulu commented on a change in pull request #17385: [NumPy] add
random.geometric op
URL: https://github.com/apache/incubator-mxnet/pull/17385#discussion_r368592023
##
File path: src/operator/numpy/random/np_geometric_op.h
##
@@ -0,0 +1,177 @@
+/*
+ * Licensed to the Apach
TaoLv commented on a change in pull request #17170: add mkldnn softmax backward
URL: https://github.com/apache/incubator-mxnet/pull/17170#discussion_r368582465
##
File path: src/operator/nn/mkldnn/mkldnn_softmax.cc
##
@@ -131,6 +143,80 @@ void MKLDNNSoftmaxForward(const nn
szha commented on a change in pull request #17376: [MXNET-1404] Implement
storage tagging, the first half of the memory profiler
URL: https://github.com/apache/incubator-mxnet/pull/17376#discussion_r368549404
##
File path: include/mxnet/storage.h
##
@@ -26,15 +26,52 @@
#d
This is an automated email from the ASF dual-hosted git repository.
aaronmarkham 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 17a6bd6 Bump the publis
vexilligera opened a new pull request #17387: [TVM][Windows] TVM Ops windows
build
URL: https://github.com/apache/incubator-mxnet/pull/17387
## Description ##
Linux link and Windows works differently, so I had to paste
contrib/tvmop/space.py to python/mxnet/space.py, otherwise python fr
liuzh91 edited a comment on issue #17380: pip install error
URL:
https://github.com/apache/incubator-mxnet/issues/17380#issuecomment-576188540
@ResearchingDexter Have you `make` under the `incubator-mxnet` directory
before doing `pip install`?
After cloning the git directory, you s
liuzh91 edited a comment on issue #17380: pip install error
URL:
https://github.com/apache/incubator-mxnet/issues/17380#issuecomment-576188540
@ResearchingDexter Have you `make` under the `incubator-mxnet` directory
before doing `pip install`?
After cloning the git directory, you s
liuzh91 commented on issue #17380: pip install error
URL:
https://github.com/apache/incubator-mxnet/issues/17380#issuecomment-576188540
@ResearchingDexter Have you `make` under the `incubator-mxnet` directory
before doing `pip install`?
---
TaoLv commented on issue #17386: fixed seed for mkldnn test
URL: https://github.com/apache/incubator-mxnet/pull/17386#issuecomment-576176991
@apeforest This change should fix the problem in #17352. Could you please
review and check?
-
TaoLv commented on a change in pull request #17386: [MXNET-17352]fixed seed for
mkldnn test
URL: https://github.com/apache/incubator-mxnet/pull/17386#discussion_r368431866
##
File path: tests/python/mkl/test_mkldnn.py
##
@@ -95,7 +97,7 @@ def __getitem__(self, key):
f
heaseny opened a new pull request #17386: [MXNET-17352]fixed seed for mkldnn
test
URL: https://github.com/apache/incubator-mxnet/pull/17386
## Description ##
This PR is intended to fix the issue in JIRA 17352 with fixed seed.
https://github.com/apache/incubator-mxnet/issues/17352
AntiZpvoh opened a new pull request #17385: [NumPy] add random.geometric op
URL: https://github.com/apache/incubator-mxnet/pull/17385
## Description ##
add numpy.random.geometric op and write the unit test
## Checklist ##
### Essentials ###
Please feel free to remove inapplic
TaoLv commented on issue #16995: CI timeout on unix-cpu Python2 test
URL:
https://github.com/apache/incubator-mxnet/issues/16995#issuecomment-576163296
Timeout again:
http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/mxnet-validation%2Funix-cpu/detail/PR-17355/3/pipeline/
69 matches
Mail list logo