AntiZpvoh opened a new pull request #17316: [NumPy] add op random.laplace
URL: https://github.com/apache/incubator-mxnet/pull/17316
## Description ##
Add op random.laplace, which could generate random samples with Laplace
Distribution.
## Checklist ##
### Essentials ###
ronnac opened a new issue #17315: mvn install scala package sigsev
mxnet::TShape::Size()
URL: https://github.com/apache/incubator-mxnet/issues/17315
## Description
mvn install of the scala package fails on armhf 32bit with SIGSEGV at
Building MXNet Scala Package - Core INTERNAL
rondogency commented on issue #16654: Multithreaded Inference Support
URL: https://github.com/apache/incubator-mxnet/pull/16654#issuecomment-574539215
@anirudh2290 Thanks for this big contribution! one suggestion is that lots
of developers complaining about cached op lacking docblocks and
rondogency commented on a change in pull request #16654: Multithreaded
Inference Support
URL: https://github.com/apache/incubator-mxnet/pull/16654#discussion_r366730799
##
File path: src/imperative/cached_op_threadsafe.cc
##
@@ -0,0 +1,329 @@
+/*
+ * Licensed to the
TaoLv commented on issue #16890: Mkldnn fullyConnect bwd bug fix
URL: https://github.com/apache/incubator-mxnet/pull/16890#issuecomment-574526907
@rongzha1 Any update for this PR and the flaky CPP test?
This is an automated
eric-haibin-lin commented on issue #17244: Incorrect installation instruction
URL:
https://github.com/apache/incubator-mxnet/issues/17244#issuecomment-574520361
@aaronmarkham what's the alternative solution you're proposing?
eric-haibin-lin commented on a change in pull request #16654: Multithreaded
Inference Support
URL: https://github.com/apache/incubator-mxnet/pull/16654#discussion_r366715578
##
File path: src/imperative/cached_op_threadsafe.cc
##
@@ -0,0 +1,329 @@
+/*
+ * Licensed to the
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 58edf36 Bump the
apeforest commented on a change in pull request #17305: grouping large array
tests based on type and updating nightly CI funtion
URL: https://github.com/apache/incubator-mxnet/pull/17305#discussion_r366707955
##
File path: tests/nightly/test_large_array.py
##
@@ -35,1658
Tommliu opened a new pull request #17314: all/any_fix [Numpy]
URL: https://github.com/apache/incubator-mxnet/pull/17314
## Description ##
add all/any self method
## Checklist ##
### Essentials ###
Please feel free to remove inapplicable items for your PR.
- [ ] The PR
ptrendx closed pull request #15834: Check expected vs. actual Node inputs.
Many operators non-compliant.
URL: https://github.com/apache/incubator-mxnet/pull/15834
This is an automated message from the Apache Git Service.
ptrendx commented on issue #15834: Check expected vs. actual Node inputs. Many
operators non-compliant.
URL: https://github.com/apache/incubator-mxnet/pull/15834#issuecomment-574501961
Relevant changes from this PR were incorporated in #17049, so closing this
one.
ResearchingDexter edited a comment on issue #15647: Segmentation fault: 11
URL:
https://github.com/apache/incubator-mxnet/issues/15647#issuecomment-574496320
I ran ```python diagnose.py```
then printed
```shell
--Python Info--
Version : 3.6.8
Compiler
ResearchingDexter edited a comment on issue #15647: Segmentation fault: 11
URL:
https://github.com/apache/incubator-mxnet/issues/15647#issuecomment-574496320
I runned ```python diagnose.py```
then printed
```shell
--Python Info--
Version : 3.6.8
ResearchingDexter commented on issue #15647: Segmentation fault: 11
URL:
https://github.com/apache/incubator-mxnet/issues/15647#issuecomment-574496320
I runned ```python python diagnose.py```
then printed
```shell
--Python Info--
Version : 3.6.8
larroy commented on issue #15530: [MXNET-978] Higher Order Gradient Support
`arcsinh`, `arccosh`.
URL: https://github.com/apache/incubator-mxnet/pull/15530#issuecomment-574474424
Flaky test?
DickJC123 commented on issue #17312: Fix flakey test_ndarray.py:test_reduce
URL: https://github.com/apache/incubator-mxnet/pull/17312#issuecomment-574468792
I see now that the relaxing of precision for float32 was already applied by
PR https://github.com/apache/incubator-mxnet/pull/16992.
litaotju commented on issue #17132: TX2 gpu memory cost abnormal
URL:
https://github.com/apache/incubator-mxnet/issues/17132#issuecomment-574467741
Which memory are you referencong here? TX2 shares gpu and host cpu memory.
While in x86 the cpu and gpu does not share mem
This is an automated email from the ASF dual-hosted git repository.
zhasheng pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 71e70f2 adding asnumpy() to output of gather(implicitly called) to
fix gather test in large vector and
This is an automated email from the ASF dual-hosted git repository.
zhasheng pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 71e70f2 adding asnumpy() to output of gather(implicitly called) to
fix gather test in large vector and
szha merged pull request #17308: Fix CosineEmbeddingLoss in when symbol API is
used
URL: https://github.com/apache/incubator-mxnet/pull/17308
This is an automated message from the Apache Git Service.
To respond to the
TaoLv opened a new pull request #17313: [WIP] Update 3rdparty/mkldnn to v1.2.0
URL: https://github.com/apache/incubator-mxnet/pull/17313
## Description ##
Currently it's the release candidate for v1.2.0. Will change to formal
release once it's published.
## Checklist ##
###
anjishnu edited a comment on issue #17298: [MXNET-1438] Adding SDML loss
function
URL: https://github.com/apache/incubator-mxnet/pull/17298#issuecomment-574458020
It looks a little tricky to port this into the 'fit' and 'score' paradigm
since this is a retrieval specific loss function
anjishnu edited a comment on issue #17298: [MXNET-1438] Adding SDML loss
function
URL: https://github.com/apache/incubator-mxnet/pull/17298#issuecomment-574458020
It looks a little tricky to port this into the 'fit' and 'score' paradigm
since this is a retrieval specific loss function
TaoLv commented on issue #17300: Align repr() function for rnn cell and rnn
layer of Gluon
URL: https://github.com/apache/incubator-mxnet/pull/17300#issuecomment-574458209
@szha Is this change good to you?
This is an
anjishnu commented on issue #17298: [MXNET-1438] Adding SDML loss function
URL: https://github.com/apache/incubator-mxnet/pull/17298#issuecomment-574458020
It looks a little tricky to port this into the 'fit' and 'score' paradigm
since this is a retrieval specific loss function which uses
zixuanweeei commented on issue #17300: Align repr() function for rnn cell and
rnn layer of Gluon
URL: https://github.com/apache/incubator-mxnet/pull/17300#issuecomment-574456481
> Looks good, as it's just a helper message for LSTM/GRU cell, not related
to the functionality. I don't think
ChaiBapchya commented on issue #14728: [MXNET-1386] fix for shape mismatch
URL: https://github.com/apache/incubator-mxnet/pull/14728#issuecomment-574456214
@ZhennanQin does #15518 address `test_subgraph.test_pos_conv_add2` issue?
DickJC123 opened a new pull request #17312: Fix flakey
test_ndarray.py:test_reduce
URL: https://github.com/apache/incubator-mxnet/pull/17312
## Description ##
This PR improves the reliability of the test_reduce unittest in
test_ndarray.py. Two seen test failure outputs are shown below
rongzha1 commented on a change in pull request #17170: add mkldnn softmax
backward
URL: https://github.com/apache/incubator-mxnet/pull/17170#discussion_r366658974
##
File path: src/operator/nn/mkldnn/mkldnn_softmax.cc
##
@@ -131,6 +143,80 @@ void
rongzha1 commented on a change in pull request #17170: add mkldnn softmax
backward
URL: https://github.com/apache/incubator-mxnet/pull/17170#discussion_r366658426
##
File path: src/operator/nn/mkldnn/mkldnn_softmax.cc
##
@@ -131,6 +143,80 @@ void
ciyongch commented on issue #17300: Align repr() function for rnn cell and rnn
layer of Gluon
URL: https://github.com/apache/incubator-mxnet/pull/17300#issuecomment-574448410
Looks good, as it's just a helper message for LSTM/GRU cell, not related to
the functionality. I don't think it's
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 0737692 Bump the
access2rohit commented on issue #17305: grouping large array tests based on
type and updating nightly CI funtion
URL: https://github.com/apache/incubator-mxnet/pull/17305#issuecomment-574441116
> LGTM. How about we regroup/rename the groups to test_tensor (including
array_creation and
blchu opened a new pull request #17311: Added beamsearch_set_finished Operator
URL: https://github.com/apache/incubator-mxnet/pull/17311
## Description ##
Added a new operator, beamsearch_set_finished, for updating the scores of
finished beams in beam search more efficiently than using
apeforest commented on a change in pull request #16654: Multithreaded Inference
Support
URL: https://github.com/apache/incubator-mxnet/pull/16654#discussion_r366631962
##
File path: src/imperative/cached_op_threadsafe.cc
##
@@ -0,0 +1,329 @@
+/*
+ * Licensed to the Apache
apeforest commented on a change in pull request #16654: Multithreaded Inference
Support
URL: https://github.com/apache/incubator-mxnet/pull/16654#discussion_r366631849
##
File path: src/imperative/cached_op_threadsafe.cc
##
@@ -0,0 +1,329 @@
+/*
+ * Licensed to the Apache
apeforest commented on a change in pull request #16654: Multithreaded Inference
Support
URL: https://github.com/apache/incubator-mxnet/pull/16654#discussion_r366630967
##
File path: src/imperative/cached_op.cc
##
@@ -32,247 +32,12 @@
ronnac opened a new issue #17310: mvn compile of the scala package fails on
armhf because of hardcoded sse and f16c flags
URL: https://github.com/apache/incubator-mxnet/issues/17310
## Description
mvn compile of the scala package fails on armhf because of hardcoded sse and
f16c flags
haojin2 commented on issue #17298: [MXNET-1438] Adding SDML loss function
URL: https://github.com/apache/incubator-mxnet/pull/17298#issuecomment-574418897
@anjishnu Like this:
https://github.com/apache/incubator-mxnet/pull/17298/files#diff-b129d416cc6980021cb35420a66e4d3aR334
anjishnu commented on issue #17298: [MXNET-1438] Adding SDML loss function
URL: https://github.com/apache/incubator-mxnet/pull/17298#issuecomment-574415884
Sure will address the sanity cases -
Can you give an example of a unit test that is appropriately randomized so I
can base it
anjishnu edited a comment on issue #17298: [MXNET-1438] Adding SDML loss
function
URL: https://github.com/apache/incubator-mxnet/pull/17298#issuecomment-574415884
@haojin2 Sure will address the sanity cases.
Can you give an example of a unit test that is appropriately randomized so
access2rohit commented on a change in pull request #17309: adding docs for
64bit C APIs of large tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309#discussion_r366609591
##
File path: include/mxnet/c_api.h
##
@@ -585,6
ChaiBapchya commented on a change in pull request #17309: adding docs for 64bit
C APIs of large tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309#discussion_r366608184
##
File path: include/mxnet/c_api.h
##
@@
ChaiBapchya commented on a change in pull request #17309: adding docs for 64bit
C APIs of large tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309#discussion_r366607671
##
File path: include/mxnet/c_api.h
##
@@ -1840,6
ChaiBapchya commented on a change in pull request #17309: adding docs for 64bit
C APIs of large tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309#discussion_r366608014
##
File path: include/mxnet/c_api.h
##
@@
ChaiBapchya commented on a change in pull request #17309: adding docs for 64bit
C APIs of large tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309#discussion_r366606053
##
File path: include/mxnet/c_api.h
##
@@ -1840,6
ChaiBapchya commented on a change in pull request #17309: adding docs for 64bit
C APIs of large tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309#discussion_r366606476
##
File path: include/mxnet/c_api.h
##
@@ -1727,6
ChaiBapchya commented on a change in pull request #17309: adding docs for 64bit
C APIs of large tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309#discussion_r366606351
##
File path: include/mxnet/c_api.h
##
@@ -1727,6
ChaiBapchya commented on a change in pull request #17309: adding docs for 64bit
C APIs of large tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309#discussion_r366606283
##
File path: include/mxnet/c_api.h
##
@@ -1727,6
ChaiBapchya commented on a change in pull request #17309: adding docs for 64bit
C APIs of large tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309#discussion_r366605962
##
File path: include/mxnet/c_api.h
##
@@ -1840,6
ChaiBapchya commented on a change in pull request #17309: adding docs for 64bit
C APIs of large tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309#discussion_r366605886
##
File path: include/mxnet/c_api.h
##
@@ -1840,6
apeforest commented on issue #17159: Performance regression from 1.4.1 to 1.5.1
URL:
https://github.com/apache/incubator-mxnet/issues/17159#issuecomment-574402879
No regression found when building mxnet from source with 1.4, 1.5 and 1.6.
The difference in performance could be due to the
apeforest closed issue #17159: Performance regression from 1.4.1 to 1.5.1
URL: https://github.com/apache/incubator-mxnet/issues/17159
This is an automated message from the Apache Git Service.
To respond to the message,
zhhoper commented on issue #17231: cannot quantization example
URL:
https://github.com/apache/incubator-mxnet/issues/17231#issuecomment-574399327
@wuxun-zhang Sorry that I haven't been able to touch that after reporting
the bug. Will take a look at that and let you know if the bug is
aaronmarkham commented on a change in pull request #17206: Windows dev
environment configuration, update install instructions from source in the docs.
URL: https://github.com/apache/incubator-mxnet/pull/17206#discussion_r366591319
##
File path:
aaronmarkham commented on a change in pull request #17206: Windows dev
environment configuration, update install instructions from source in the docs.
URL: https://github.com/apache/incubator-mxnet/pull/17206#discussion_r366594527
##
File path:
aaronmarkham commented on a change in pull request #17206: Windows dev
environment configuration, update install instructions from source in the docs.
URL: https://github.com/apache/incubator-mxnet/pull/17206#discussion_r366595186
##
File path:
aaronmarkham commented on a change in pull request #17206: Windows dev
environment configuration, update install instructions from source in the docs.
URL: https://github.com/apache/incubator-mxnet/pull/17206#discussion_r366592778
##
File path:
aaronmarkham commented on a change in pull request #17206: Windows dev
environment configuration, update install instructions from source in the docs.
URL: https://github.com/apache/incubator-mxnet/pull/17206#discussion_r366593590
##
File path:
aaronmarkham commented on issue #17244: Incorrect installation instruction
URL:
https://github.com/apache/incubator-mxnet/issues/17244#issuecomment-574389324
The content looks fine in here:
apeforest commented on a change in pull request #17309: adding docs for 64bit C
APIs of large tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309#discussion_r366588180
##
File path: include/mxnet/c_api.h
##
@@ -585,6
access2rohit commented on issue #17309: adding docs for 64bit C APIs of large
tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309#issuecomment-574378284
@apeforest @ChaiBapchya PR ready for review
access2rohit commented on issue #17309: adding docs for 64bit C APIs of large
tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309#issuecomment-574378172
@mxnet-label-bot add [pr-awaiting-review]
access2rohit opened a new pull request #17309: adding docs for 64bit C APIs of
large tensor and removing not required int64 C APIs
URL: https://github.com/apache/incubator-mxnet/pull/17309
## Description ##
adding docs for 64bit C APIs of large tensor and removing not required int64
C
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 9f2e73f Support R-package with cmake build and fix installation
instructions (#17228)
add 71e70f2
apeforest merged pull request #17290: Fixing gather op in large vector and array
URL: https://github.com/apache/incubator-mxnet/pull/17290
This is an automated message from the Apache Git Service.
To respond to the message,
larroy commented on issue #17293: [Build] Add a reasonable default for
CMAKE_CUDA_COMPILER in *nix
URL: https://github.com/apache/incubator-mxnet/pull/17293#issuecomment-574364704
/usr/local/cuda/bin/nvcc is usually NOT in the path.
larroy commented on issue #17293: [Build] Add a reasonable default for
CMAKE_CUDA_COMPILER in *nix
URL: https://github.com/apache/incubator-mxnet/pull/17293#issuecomment-574364596
nvcc is normally not in the PATH in ubuntu nvidia packages it goes in the
path that you can see in this PR.
dithyrambe closed issue #17275: CosineEmbeddingLoss won't work after hybridize
(ie. with symbol API)
URL: https://github.com/apache/incubator-mxnet/issues/17275
This is an automated message from the Apache Git Service.
To
D-Roberts commented on issue #17307: NDArray does not have `std()` method
URL:
https://github.com/apache/incubator-mxnet/issues/17307#issuecomment-574361269
@vietspaceanh In case it helps your usecase, there is:
`import mxnet.numpy as np`
`a = np.random.randn(3,3)`
dithyrambe opened a new pull request #17308: Fix CosineEmbeddingLoss in when
symbol API is used
URL: https://github.com/apache/incubator-mxnet/pull/17308
Fixes #17275
## Description ##
Fix inconsistances between `ndarray` & `symbol` API regarding `.reshape`
method
dithyrambe commented on issue #17275: CosineEmbeddingLoss won't work after
hybridize (ie. with symbol API)
URL:
https://github.com/apache/incubator-mxnet/issues/17275#issuecomment-574348641
All right,
Thanks for your help.
vietspaceanh opened a new issue #17307: NDArray does not have `std()` method
URL: https://github.com/apache/incubator-mxnet/issues/17307
Although `ndarray.mean()` exists, I cannot find the standard deviation
method for `NDArray'.
Example:
```python
from mxnet import nd
a =
apeforest edited a comment on issue #17304: [flaky]
test_higher_order_grad.test_tanh
URL:
https://github.com/apache/incubator-mxnet/issues/17304#issuecomment-574326240
@kshitij12345 Would you mind taking a look at this test? Thanks! It looks
like we can increase the precision tolerance
apeforest edited a comment on issue #17304: [flaky]
test_higher_order_grad.test_tanh
URL:
https://github.com/apache/incubator-mxnet/issues/17304#issuecomment-574326240
@kshitij12345 Could you please take a look at this test? Thanks! It looks
like we can increase the precision tolerance
apeforest commented on issue #17304: [flaky] test_higher_order_grad.test_tanh
URL:
https://github.com/apache/incubator-mxnet/issues/17304#issuecomment-574326240
@kshitij12345 Could you please take a look at this test? Thanks!
access2rohit commented on issue #17306: grouping large vector tests based on
their type and adding their nightly test function
URL: https://github.com/apache/incubator-mxnet/pull/17306#issuecomment-574323103
@apeforest @ChaiBapchya PR is ready for review
access2rohit commented on issue #17306: grouping large vector tests based on
their type and adding their nightly test function
URL: https://github.com/apache/incubator-mxnet/pull/17306#issuecomment-574322969
@mxnet-label-bot add [pr-awaiting-review]
access2rohit opened a new pull request #17306: grouping large vector tests
based on their type and adding their nightly test function
URL: https://github.com/apache/incubator-mxnet/pull/17306
Grouped large vector tests for large tensor support into 4:
creation
manipulation
dnn_ops
ptrendx commented on issue #17278: [Performance] CachedOp/Gluon performance
thread
URL:
https://github.com/apache/incubator-mxnet/issues/17278#issuecomment-574321205
There are a few performance issues that we saw in `CachedOp` compared to
graph executor.
-
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 9189d46 Bump the
ptrendx commented on issue #17303: New Basic Layer "nn.bias_add"
URL:
https://github.com/apache/incubator-mxnet/issues/17303#issuecomment-574316444
You should be able to use `broadcast_add` there, no?
This is an automated
access2rohit commented on issue #17305: grouping large array tests based on
type and updating nightly CI funtion
URL: https://github.com/apache/incubator-mxnet/pull/17305#issuecomment-574315357
@apeforest @ChaiBapchya PR is ready for review
access2rohit commented on issue #17305: grouping large array tests based on
type and updating nightly CI funtion
URL: https://github.com/apache/incubator-mxnet/pull/17305#issuecomment-574315071
@mxnet-label-bot add [pr_awaiting-review]
access2rohit edited a comment on issue #17305: grouping large array tests based
on type and updating nightly CI funtion
URL: https://github.com/apache/incubator-mxnet/pull/17305#issuecomment-574315071
@mxnet-label-bot add [pr-awaiting-review]
reminisce opened a new issue #17304: [flaky]
URL: https://github.com/apache/incubator-mxnet/issues/17304
```
==
FAIL: test_higher_order_grad.test_tanh
access2rohit opened a new pull request #17305: grouping large array tests based
on type and updating nightly CI funtion
URL: https://github.com/apache/incubator-mxnet/pull/17305
## Description ##
Grouped Large Array tests for large tensor support into 4:
creation
manipulation
leezu commented on issue #17275: CosineEmbeddingLoss won't work after hybridize
(ie. with symbol API)
URL:
https://github.com/apache/incubator-mxnet/issues/17275#issuecomment-574306610
If you think it's required to change line 925, 926, 927 above to use tuple
`(-1, 1)` please create a
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 3971938 Fix #17267, add expected and got datatype for concat error
msgs (#17271)
add 9f2e73f
leezu merged pull request #17228: Support R-package with cmake build and fix
installation instructions
URL: https://github.com/apache/incubator-mxnet/pull/17228
This is an automated message from the Apache Git Service.
To
leezu closed issue #17103: R installation instructions out of date
URL: https://github.com/apache/incubator-mxnet/issues/17103
This is an automated message from the Apache Git Service.
To respond to the message, please log
This is an automated email from the ASF dual-hosted git repository.
haibin pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 058de55 Fix image display in python autograd tutorial (#17243)
add 3971938 Fix #17267, add expected
eric-haibin-lin commented on issue #17271: Fix #17267, add expected and got
datatype for concat error msgs
URL: https://github.com/apache/incubator-mxnet/pull/17271#issuecomment-574297555
Great. Thansk for the fix! @litaotju
eric-haibin-lin closed issue #17267: Unclear error message in np.dstack
URL: https://github.com/apache/incubator-mxnet/issues/17267
This is an automated message from the Apache Git Service.
To respond to the message, please
eric-haibin-lin merged pull request #17271: Fix #17267, add expected and got
datatype for concat error msgs
URL: https://github.com/apache/incubator-mxnet/pull/17271
This is an automated message from the Apache Git Service.
dithyrambe commented on issue #17275: CosineEmbeddingLoss won't work after
hybridize (ie. with symbol API)
URL:
https://github.com/apache/incubator-mxnet/issues/17275#issuecomment-574267259
Nice,
I guess the issue is solved then.
What about the fix ?
Regards
leezu commented on issue #17181: Fail to build mxnet from source
URL:
https://github.com/apache/incubator-mxnet/issues/17181#issuecomment-574248904
@apeforest @mjsML please provide the output of your `cmake` commands.
It's unclear whether MKL is found during your compilation attempt
xidulu commented on a change in pull request #17302: [numpy]add op
random.logistic
URL: https://github.com/apache/incubator-mxnet/pull/17302#discussion_r366410192
##
File path: tests/python/unittest/test_numpy_op.py
##
@@ -3325,15 +3363,15 @@ def hybrid_forward(self, F,
TaoLv commented on a change in pull request #17170: add mkldnn softmax backward
URL: https://github.com/apache/incubator-mxnet/pull/17170#discussion_r366408821
##
File path: src/operator/nn/mkldnn/mkldnn_softmax.cc
##
@@ -131,6 +143,80 @@ void MKLDNNSoftmaxForward(const
1 - 100 of 129 matches
Mail list logo