JiadingGai commented on issue #16063: Allow loading from model with empty
weights.
URL: https://github.com/apache/incubator-mxnet/pull/16063#issuecomment-529803893
@ChaiBapchya Thanks for pointing it. Closing this request out and look
forward to getting the master one pushed.
JiadingGai closed pull request #16063: Allow loading from model with empty
weights.
URL: https://github.com/apache/incubator-mxnet/pull/16063
This is an automated message from the Apache Git Service.
To respond to the messag
JiadingGai commented on issue #16062: Allow loading from model with empty
weights.
URL: https://github.com/apache/incubator-mxnet/pull/16062#issuecomment-529804242
Thanks @szha and @ChaiBapchya for pointing out the policy. Closing this
request out and look forward to getting the master one
JiadingGai closed pull request #16062: Allow loading from model with empty
weights.
URL: https://github.com/apache/incubator-mxnet/pull/16062
This is an automated message from the Apache Git Service.
To respond to the messag
Zheweiqiu edited a comment on issue #16042: Error when calling
get_backend_symbol
URL:
https://github.com/apache/incubator-mxnet/issues/16042#issuecomment-529804310
> @Zheweiqiu Can you show the whole process / script that you did for
building MXNet ?
I am using retinaface mxnet mo
Zheweiqiu commented on issue #16042: Error when calling get_backend_symbol
URL:
https://github.com/apache/incubator-mxnet/issues/16042#issuecomment-529804310
> @Zheweiqiu Can you show the whole process / script that you did for
building MXNet ?
I am using retinaface mxnet model and
Zheweiqiu edited a comment on issue #16042: Error when calling
get_backend_symbol
URL:
https://github.com/apache/incubator-mxnet/issues/16042#issuecomment-529804310
> @Zheweiqiu Can you show the whole process / script that you did for
building MXNet ?
I am using retinaface mxnet mo
Zheweiqiu edited a comment on issue #16042: Error when calling
get_backend_symbol
URL:
https://github.com/apache/incubator-mxnet/issues/16042#issuecomment-529804310
> @Zheweiqiu Can you show the whole process / script that you did for
building MXNet ?
I am using retinaface mxnet mo
haojin2 commented on a change in pull request #16132: [Numpy] Numpy eye
URL: https://github.com/apache/incubator-mxnet/pull/16132#discussion_r322600224
##
File path: python/mxnet/ndarray/numpy/_op.py
##
@@ -2363,3 +2363,32 @@ def var(a, axis=None, dtype=None, out=None, ddof
haojin2 commented on a change in pull request #16132: [Numpy] Numpy eye
URL: https://github.com/apache/incubator-mxnet/pull/16132#discussion_r322600373
##
File path: python/mxnet/symbol/numpy/_symbol.py
##
@@ -2678,4 +2678,33 @@ def var(a, axis=None, dtype=None, out=None, d
haojin2 commented on a change in pull request #16132: [Numpy] Numpy eye
URL: https://github.com/apache/incubator-mxnet/pull/16132#discussion_r322600348
##
File path: python/mxnet/numpy/multiarray.py
##
@@ -3808,3 +3808,28 @@ def var(a, axis=None, dtype=None, out=None, ddof=
haojin2 commented on a change in pull request #16132: [Numpy] Numpy eye
URL: https://github.com/apache/incubator-mxnet/pull/16132#discussion_r322600420
##
File path: python/mxnet/symbol/numpy/_symbol.py
##
@@ -2678,4 +2678,33 @@ def var(a, axis=None, dtype=None, out=None, d
haojin2 commented on a change in pull request #16132: [Numpy] Numpy eye
URL: https://github.com/apache/incubator-mxnet/pull/16132#discussion_r322600149
##
File path: python/mxnet/ndarray/numpy/_op.py
##
@@ -2363,3 +2363,32 @@ def var(a, axis=None, dtype=None, out=None, ddof
haojin2 commented on a change in pull request #16132: [Numpy] Numpy eye
URL: https://github.com/apache/incubator-mxnet/pull/16132#discussion_r322600324
##
File path: python/mxnet/numpy/multiarray.py
##
@@ -3808,3 +3808,28 @@ def var(a, axis=None, dtype=None, out=None, ddof=
leezu commented on a change in pull request #16097: [numpy] array ufunc and
array function protocols
URL: https://github.com/apache/incubator-mxnet/pull/16097#discussion_r322640975
##
File path: python/mxnet/numpy_dispatch_protocol.py
##
@@ -0,0 +1,214 @@
+# Licensed to th
mxnet-label-bot commented on issue #16134: Incorrect subtraction
URL:
https://github.com/apache/incubator-mxnet/issues/16134#issuecomment-529905850
Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some labels so
that the appropriate MXNet co
RomanSteinberg opened a new issue #16134: Incorrect subtraction
URL: https://github.com/apache/incubator-mxnet/issues/16134
## Description
The result of subtraction is influenced by third party code.
## Environment info (Required)
```
--Python Info--
Ve
leezu commented on issue #16134: Incorrect subtraction
URL:
https://github.com/apache/incubator-mxnet/issues/16134#issuecomment-529913938
I can't reproduce this on Ubuntu 18.04 with neither MXNet 1.5 from pypi nor
a local build from latest master.
Could you please share more informa
Caenorst commented on issue #16042: Error when calling get_backend_symbol
URL:
https://github.com/apache/incubator-mxnet/issues/16042#issuecomment-529946006
@Zheweiqiu Sorry, I asked you about the process for BUILDING MXNet.
lostella opened a new issue #16135: Sampling fails on mxnet-mkl==1.5.0?
URL: https://github.com/apache/incubator-mxnet/issues/16135
## Description
Some random sampling functions seem not to be working on mxnet-mkl version
1.5.0.
## Environment info (Required)
```
---
mxnet-label-bot commented on issue #16135: Sampling fails on mxnet-mkl==1.5.0?
URL:
https://github.com/apache/incubator-mxnet/issues/16135#issuecomment-529948986
Hey, this is the MXNet Label Bot.
Thank you for submitting the issue! I will try and suggest some labels so
that the approp
lostella commented on issue #16135: Sampling fails on mxnet-mkl==1.5.0?
URL:
https://github.com/apache/incubator-mxnet/issues/16135#issuecomment-529970623
This still did not occur in `mxnet_mkl-1.4.1`, and appears to be still
occurring in `mxnet_mkl-1.6.0b20190909`
---
access2rohit commented on issue #16112: [DO NOT MERGE]Revert custom profiler
URL: https://github.com/apache/incubator-mxnet/pull/16112#issuecomment-530022581
> why?
Only for debugging purposes this will not be merged. After merging Zhaoqi's
PR there is an observed increase in unix-gp
lostella edited a comment on issue #16135: Sampling fails on mxnet-mkl==1.5.0?
URL:
https://github.com/apache/incubator-mxnet/issues/16135#issuecomment-529970623
This did not occur in `mxnet_mkl-1.4.1`, and appears to be still occurring
in `mxnet_mkl-1.6.0b20190909`
--
eric-haibin-lin commented on issue #16112: [DO NOT MERGE]Revert custom profiler
URL: https://github.com/apache/incubator-mxnet/pull/16112#issuecomment-530042713
thanks for the explanation. Is it the increase in the time for build, or
running tests?
-
Caenorst commented on a change in pull request #16122: Add fast implementation
of LARS
URL: https://github.com/apache/incubator-mxnet/pull/16122#discussion_r322887660
##
File path: tests/python/gpu/test_optimizer.py
##
@@ -0,0 +1,93 @@
+# Licensed to the Apache Software Fo
samskalicky commented on issue #16131: [WIP] fix for duplicate subgraph
inputs/outputs
URL: https://github.com/apache/incubator-mxnet/pull/16131#issuecomment-530074025
@zheng-da @reminisce could you please review?
This is an
samskalicky commented on issue #16131: [WIP] fix for duplicate subgraph
inputs/outputs
URL: https://github.com/apache/incubator-mxnet/pull/16131#issuecomment-530074176
@mxnet-label-bot add [pr-awaiting-review]
This is an auto
reminisce commented on issue #16097: [numpy] array ufunc and array function
protocols
URL: https://github.com/apache/incubator-mxnet/pull/16097#issuecomment-530086810
@leezu Thanks for the review. I tried the commands you showed, but it works
for me. Could you confirm that you are using Nu
reminisce commented on a change in pull request #16097: [numpy] array ufunc and
array function protocols
URL: https://github.com/apache/incubator-mxnet/pull/16097#discussion_r322928210
##
File path: python/mxnet/numpy_dispatch_protocol.py
##
@@ -0,0 +1,214 @@
+# Licensed t
reminisce edited a comment on issue #16097: [numpy] array ufunc and array
function protocols
URL: https://github.com/apache/incubator-mxnet/pull/16097#issuecomment-530110567
@leezu Re: Can test be improved? Yes, absolutely. This PR is for building
the skeleton of testing suite. It enables
reminisce commented on issue #16097: [numpy] array ufunc and array function
protocols
URL: https://github.com/apache/incubator-mxnet/pull/16097#issuecomment-530110567
@leezu Re: Can test be improved? Yes, absolutely. This PR is for building
the skeleton of testing suite. It enables `mxnet.
ChaiBapchya commented on issue #16134: Incorrect subtraction
URL:
https://github.com/apache/incubator-mxnet/issues/16134#issuecomment-530114764
@mxnet-label-bot add [Bug, Question]
This is an automated message from the Apache
ChaiBapchya commented on issue #16090: ONNX dosen't support SwapAxis
URL:
https://github.com/apache/incubator-mxnet/issues/16090#issuecomment-530115184
@mxnet-label-bot add [ONNX]
This is an automated message from the Apache
ChaiBapchya commented on issue #16090: ONNX dosen't support SwapAxis
URL:
https://github.com/apache/incubator-mxnet/issues/16090#issuecomment-530115332
@vandanavk take a look
@Tianxiaomo thanks for pointing out
This is an
ChaiBapchya commented on issue #16060: [bug] mxnet.ndarray.sparse.norm fallback
regression in 1.5.0 and master
URL:
https://github.com/apache/incubator-mxnet/issues/16060#issuecomment-530115636
@mxnet-label-bot add [Bug, Operator]
--
ChaiBapchya commented on issue #16135: Sampling fails on mxnet==1.5.0 on Linux?
URL:
https://github.com/apache/incubator-mxnet/issues/16135#issuecomment-530140670
@mxnet-label-bot add [Operator, Bug]
This is an automated mess
ChaiBapchya commented on issue #16135: Sampling fails on mxnet==1.5.0 on Linux?
URL:
https://github.com/apache/incubator-mxnet/issues/16135#issuecomment-530141237
Thanks for pointing out.
I was able to reproduce this issue for `pip install mxnet==1.5.0`, `pip
install mxnet==1.6.0`
access2rohit commented on issue #16136: Reducing memory footprint of one_hot
for Large Vector Testing
URL: https://github.com/apache/incubator-mxnet/pull/16136#issuecomment-530146058
@mxnet-label-bot add [pr-awaiting-review]
access2rohit opened a new pull request #16136: Reducing memory footprint of
one_hot for Large Vector Testing
URL: https://github.com/apache/incubator-mxnet/pull/16136
## Description ##
Using single element one hot location Ndarray multiple times to ensure
onr_hot vector is single row on
access2rohit commented on issue #16136: Reducing memory footprint of one_hot
for Large Vector Testing
URL: https://github.com/apache/incubator-mxnet/pull/16136#issuecomment-530146612
@apeforest @ChaiBapchya PR is ready for review
---
apeforest commented on a change in pull request #16136: Reducing memory
footprint of one_hot for Large Vector Testing
URL: https://github.com/apache/incubator-mxnet/pull/16136#discussion_r322993500
##
File path: tests/nightly/test_large_vector.py
##
@@ -715,12 +715,13 @@ d
apeforest commented on issue #16136: Reducing memory footprint of one_hot for
Large Vector Testing
URL: https://github.com/apache/incubator-mxnet/pull/16136#issuecomment-530148872
Actually, after checking test_large_array.py, I think this test is
redundant. We can remove it from test_large
aaronmarkham commented on issue #16105: Update python dependencies
URL: https://github.com/apache/incubator-mxnet/pull/16105#issuecomment-530149891
Wouldn't this imply that CI's python install scripts need to be updated
accordingly?
-
access2rohit commented on a change in pull request #16136: Reducing memory
footprint of one_hot for Large Vector Testing
URL: https://github.com/apache/incubator-mxnet/pull/16136#discussion_r322995113
##
File path: tests/nightly/test_large_vector.py
##
@@ -715,12 +715,13 @
apeforest commented on a change in pull request #16136: Reducing memory
footprint of one_hot for Large Vector Testing
URL: https://github.com/apache/incubator-mxnet/pull/16136#discussion_r322993500
##
File path: tests/nightly/test_large_vector.py
##
@@ -715,12 +715,13 @@ d
access2rohit commented on a change in pull request #16136: Reducing memory
footprint of one_hot for Large Vector Testing
URL: https://github.com/apache/incubator-mxnet/pull/16136#discussion_r322995113
##
File path: tests/nightly/test_large_vector.py
##
@@ -715,12 +715,13 @
access2rohit commented on a change in pull request #16136: Reducing memory
footprint of one_hot for Large Vector Testing
URL: https://github.com/apache/incubator-mxnet/pull/16136#discussion_r322998437
##
File path: tests/nightly/test_large_vector.py
##
@@ -715,12 +715,13 @
zachgk opened a new pull request #16137: Update python dependencies in more
places
URL: https://github.com/apache/incubator-mxnet/pull/16137
## Description ##
Updates python dependencies continuing #16105
@lanking520 @access2rohit @aaronmarkham
## Checklist ##
### Esse
access2rohit commented on a change in pull request #16136: Reducing memory
footprint of one_hot for Large Vector Testing
URL: https://github.com/apache/incubator-mxnet/pull/16136#discussion_r322998437
##
File path: tests/nightly/test_large_vector.py
##
@@ -715,12 +715,13 @
zachgk commented on issue #16105: Update python dependencies
URL: https://github.com/apache/incubator-mxnet/pull/16105#issuecomment-530153590
@aaronmarkham That would make sense. I did a search to find other places to
update (including the CI install scripts) for
https://github.com/apache/
access2rohit commented on a change in pull request #16136: Reducing memory
footprint of one_hot for Large Vector Testing
URL: https://github.com/apache/incubator-mxnet/pull/16136#discussion_r322998437
##
File path: tests/nightly/test_large_vector.py
##
@@ -715,12 +715,13 @
access2rohit commented on a change in pull request #16136: Reducing memory
footprint of one_hot for Large Array Testing
URL: https://github.com/apache/incubator-mxnet/pull/16136#discussion_r322998437
##
File path: tests/nightly/test_large_vector.py
##
@@ -715,12 +715,13 @@
samskalicky commented on issue #16131: [WIP] fix for duplicate subgraph
inputs/outputs
URL: https://github.com/apache/incubator-mxnet/pull/16131#issuecomment-530167932
@ZhennanQin for review. Any suggestions on how to better fix the problem?
eric-haibin-lin commented on issue #15968: 1 bit mask for Dropout
URL:
https://github.com/apache/incubator-mxnet/issues/15968#issuecomment-530169606
@apeforest
This is an automated message from the Apache Git Service.
To re
Zheweiqiu commented on issue #16042: Error when calling get_backend_symbol
URL:
https://github.com/apache/incubator-mxnet/issues/16042#issuecomment-530189067
> @Zheweiqiu Sorry, I asked you about the process for BUILDING MXNet.
You mean the process of building model? I am using the p
Caenorst commented on issue #16042: Error when calling get_backend_symbol
URL:
https://github.com/apache/incubator-mxnet/issues/16042#issuecomment-530204843
We suggested your rebuild MXNet from Source with TensorRT and you said you
did it. Did you ?
---
reminisce commented on issue #16113: [Flaky] test_mkldnn.test_activation
URL:
https://github.com/apache/incubator-mxnet/issues/16113#issuecomment-530204764
@ElaineBao Thanks for checking it. Unfortunately, the log was overwritten by
the follow-up run. I will report it once I see it again.
iblis17 opened a new pull request #16138: julia: fix `mx.forward` kwargs
checking
URL: https://github.com/apache/incubator-mxnet/pull/16138
close https://github.com/dmlc/MXNet.jl/issues/431
- test case included
This is
This is an automated email from the ASF dual-hosted git repository.
iblis pushed a commit to branch ib/fix-forward
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git
commit bb8a6f42b89b090c9b134d5710157606e9b99494
Author: Iblis Lin
AuthorDate: Wed Sep 11 04:18:45 2019 +
This is an automated email from the ASF dual-hosted git repository.
iblis pushed a change to branch ib/fix-forward
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
at bb8a6f4 julia: fix `mx.forward` kwargs checking
This branch includes the following new commits:
iblis17 commented on issue #15613: [Discussion] 1.5.1 Patch Release
URL:
https://github.com/apache/incubator-mxnet/issues/15613#issuecomment-530213936
> @iblis17 Could you help to pick #15608 #15609 to the v1.5.x branch?
@TaoLv Just can I just push to the v1.5.x branch direcetly?
--
iblis17 edited a comment on issue #15613: [Discussion] 1.5.1 Patch Release
URL:
https://github.com/apache/incubator-mxnet/issues/15613#issuecomment-513439987
I will track Julia stuffs that need to be backported here:
- [x] https://github.com/apache/incubator-mxnet/pull/15609
- [x] htt
iblis17 edited a comment on issue #15613: [Discussion] 1.5.1 Patch Release
URL:
https://github.com/apache/incubator-mxnet/issues/15613#issuecomment-513439987
I will track Julia stuffs that need to be backported here:
- [x] https://github.com/apache/incubator-mxnet/pull/15609
- [ ] htt
apeforest merged pull request #16136: Reducing memory footprint of one_hot for
Large Array Testing
URL: https://github.com/apache/incubator-mxnet/pull/16136
This is an automated message from the Apache Git Service.
To respon
This is an automated email from the ASF dual-hosted git repository.
apeforest pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 9675a2d Update python dependencies (#16105)
add e87995d Reducing memory footprint of one_hot for L
lostella commented on issue #16135: Sampling fails on mxnet==1.5.0 on Linux?
URL:
https://github.com/apache/incubator-mxnet/issues/16135#issuecomment-530226234
> No error for `pip install mxnet==1.4.0`
> So likely introduced in patch release 1.4.1
I was not able to reproduce on 1.
marcoabreu commented on issue #15613: [Discussion] 1.5.1 Patch Release
URL:
https://github.com/apache/incubator-mxnet/issues/15613#issuecomment-530237141
No, please create pull requests instead
This is an automated message fr
apeforest commented on issue #15613: [Discussion] 1.5.1 Patch Release
URL:
https://github.com/apache/incubator-mxnet/issues/15613#issuecomment-530240272
Haven't we already passed code freeze?
On Tue, Sep 10, 2019 at 11:16 PM Marco de Abreu
wrote:
> No, please create pull r
69 matches
Mail list logo