marcoabreu commented on a change in pull request #11630: [WIP] Fix flaky test
test_deconvolution
URL: https://github.com/apache/incubator-mxnet/pull/11630#discussion_r201789608
##
File path: tests/python/unittest/common.py
##
@@ -94,6 +95,23 @@ def random_seed(seed=None):
nswamy commented on a change in pull request #11605: Updating MultiTask example
to use new infer api and adding test for CI
URL: https://github.com/apache/incubator-mxnet/pull/11605#discussion_r201789116
##
File path:
aaronmarkham closed issue #11645: Cannot train SmileCNN with Keras-MXNet
URL: https://github.com/apache/incubator-mxnet/issues/11645
This is an automated message from the Apache Git Service.
To respond to the message, please
aaronmarkham commented on issue #11645: Cannot train SmileCNN with Keras-MXNet
URL:
https://github.com/apache/incubator-mxnet/issues/11645#issuecomment-404259350
Moving this to https://github.com/awslabs/keras-apache-mxnet/issues/134
haojin2 commented on issue #10755: 'Take' operator should support non-zero axis
values
URL:
https://github.com/apache/incubator-mxnet/issues/10755#issuecomment-404259182
Supported in #11326, can close once that is merged.
haojin2 closed issue #11367: Segfault when running
test_operator_gpu.test_sparse_dot many times
URL: https://github.com/apache/incubator-mxnet/issues/11367
This is an automated message from the Apache Git Service.
To
haojin2 commented on issue #11367: Segfault when running
test_operator_gpu.test_sparse_dot many times
URL:
https://github.com/apache/incubator-mxnet/issues/11367#issuecomment-404258534
Fix already submitted in #11389. No new reports, closing this issue.
haojin2 commented on issue #11498: Fix InferStorage for sparse fallback in
FullyConnected
URL: https://github.com/apache/incubator-mxnet/pull/11498#issuecomment-404257948
@eric-haibin-lin @zheng-da
This is an automated
alireza202 edited a comment on issue #11635: Aborted python session when using
nd.concat
URL:
https://github.com/apache/incubator-mxnet/issues/11635#issuecomment-404255759
The python crashing aside, how would I actually concatenate an empty tensor
with another tensor? I tried both
szha commented on a change in pull request #11626: [MXNET-651] MXNet Model
Backwards Compatibility Checker
URL: https://github.com/apache/incubator-mxnet/pull/11626#discussion_r201785017
##
File path: tests/nightly/model_backwards_compatibility_check/common.py
##
@@
anirudh2290 commented on a change in pull request #11630: [WIP] Fix flaky test
test_deconvolution
URL: https://github.com/apache/incubator-mxnet/pull/11630#discussion_r201784385
##
File path: Jenkinsfile
##
@@ -667,6 +680,21 @@ try {
}
}
},
+
alireza202 commented on issue #11635: Aborted python session when using
nd.concat
URL:
https://github.com/apache/incubator-mxnet/issues/11635#issuecomment-404255759
The python crashing aside, how would I actually concatenate an empty tensor
with another tensor? I tried both dimensions,
zhreshold opened a new pull request #11650: remove windows note for dataloader,
fix deprecated warning in python3.7
URL: https://github.com/apache/incubator-mxnet/pull/11650
## Description ##
- multi worker data loader is supported in windows, update the doc
correspondingly.
- fix a
piyushghai commented on a change in pull request #11626: [MXNET-651] MXNet
Model Backwards Compatibility Checker
URL: https://github.com/apache/incubator-mxnet/pull/11626#discussion_r201782584
##
File path:
ThomasDelteil commented on a change in pull request #11626: [MXNET-651] MXNet
Model Backwards Compatibility Checker
URL: https://github.com/apache/incubator-mxnet/pull/11626#discussion_r201781970
##
File path:
cclauss opened a new pull request #11648: Modernize Python 2 code to get ready
for Python 3
URL: https://github.com/apache/incubator-mxnet/pull/11648
A more focused version of #11615
$ __futurize --stage1 -w .__ #
http://python-future.org/futurize_cheatsheet.html
Minimal,
piyushghai commented on a change in pull request #11626: [MXNET-651] MXNet
Model Backwards Compatibility Checker
URL: https://github.com/apache/incubator-mxnet/pull/11626#discussion_r201780568
##
File path:
mkolod commented on issue #11325: Added TensorRT runtime integration
URL: https://github.com/apache/incubator-mxnet/pull/11325#issuecomment-404252100
@reminisce Thanks, that's exactly what @Caenorst and I were thinking in
terms of how to test the graph's topology after the rewrite, and
piyushghai commented on a change in pull request #11626: [MXNET-651] MXNet
Model Backwards Compatibility Checker
URL: https://github.com/apache/incubator-mxnet/pull/11626#discussion_r201780155
##
File path: tests/nightly/model_backwards_compatibility_check/common.py
##
@@
anirudh2290 commented on issue #11635: Aborted python session when using
nd.concat
URL:
https://github.com/apache/incubator-mxnet/issues/11635#issuecomment-404251216
you should switch to 1.2.0 to prevent the python process from crashing
haojin2 commented on a change in pull request #11625: [MXNET-647] Add Diff
Collator
URL: https://github.com/apache/incubator-mxnet/pull/11625#discussion_r201779496
##
File path: tools/diff_collator.py
##
@@ -0,0 +1,165 @@
+# Licensed to the Apache Software Foundation
haojin2 commented on a change in pull request #11625: [MXNET-647] Add Diff
Collator
URL: https://github.com/apache/incubator-mxnet/pull/11625#discussion_r201778366
##
File path: tools/diff_collator.py
##
@@ -0,0 +1,165 @@
+# Licensed to the Apache Software Foundation
mkolod commented on a change in pull request #11325: Added TensorRT runtime
integration
URL: https://github.com/apache/incubator-mxnet/pull/11325#discussion_r201778476
##
File path: src/operator/contrib/tensorrt.cc
##
@@ -0,0 +1,196 @@
+/*
+ * Licensed to the Apache
cclauss closed pull request #11615: Modernize Python 2 code to get ready for
Python 3
URL: https://github.com/apache/incubator-mxnet/pull/11615
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
haojin2 commented on issue #11635: Aborted python session when using nd.concat
URL:
https://github.com/apache/incubator-mxnet/issues/11635#issuecomment-404247607
The issue could possibly be fixed by changing the InferShape of concat
operator to adopt mutual shape inference.
anirudh2290 commented on a change in pull request #11573: Add stable nrm2
Reducer
URL: https://github.com/apache/incubator-mxnet/pull/11573#discussion_r201775313
##
File path: src/operator/mshadow_op.h
##
@@ -815,6 +868,76 @@ struct nanprod {
}
};
+/*! \brief
ThomasDelteil commented on a change in pull request #11626: [MXNET-651] MXNet
Model Backwards Compatibility Checker
URL: https://github.com/apache/incubator-mxnet/pull/11626#discussion_r201773901
##
File path:
ThomasDelteil commented on a change in pull request #11626: [MXNET-651] MXNet
Model Backwards Compatibility Checker
URL: https://github.com/apache/incubator-mxnet/pull/11626#discussion_r201773565
##
File path: tests/nightly/model_backwards_compatibility_check/common.py
##
ThomasDelteil commented on a change in pull request #11626: [MXNET-651] MXNet
Model Backwards Compatibility Checker
URL: https://github.com/apache/incubator-mxnet/pull/11626#discussion_r201774694
##
File path:
mkolod commented on a change in pull request #11325: Added TensorRT runtime
integration
URL: https://github.com/apache/incubator-mxnet/pull/11325#discussion_r201774632
##
File path: Makefile
##
@@ -94,6 +94,14 @@ else
endif
CFLAGS += -I$(TPARTYDIR)/mshadow/
aaronmarkham commented on issue #11562: [MXNET-244] Update RaspberryPI
instructions
URL: https://github.com/apache/incubator-mxnet/pull/11562#issuecomment-404245844
@larroy - I'd remove the `##` and just `**` bold stuff for section headers.
You could also use html styles and nested ``
aaronmarkham commented on a change in pull request #11049: Add linux and macos
MKLDNN Building Instruction
URL: https://github.com/apache/incubator-mxnet/pull/11049#discussion_r201771261
##
File path: MKLDNN_README.md
##
@@ -0,0 +1,287 @@
+# Build/Install MXNet with
aaronmarkham commented on a change in pull request #11049: Add linux and macos
MKLDNN Building Instruction
URL: https://github.com/apache/incubator-mxnet/pull/11049#discussion_r201768297
##
File path: MKLDNN_README.md
##
@@ -0,0 +1,287 @@
+# Build/Install MXNet with
aaronmarkham commented on a change in pull request #11049: Add linux and macos
MKLDNN Building Instruction
URL: https://github.com/apache/incubator-mxnet/pull/11049#discussion_r201766017
##
File path: MKLDNN_README.md
##
@@ -0,0 +1,302 @@
+# Build/Install MXNet with
aaronmarkham commented on a change in pull request #11049: Add linux and macos
MKLDNN Building Instruction
URL: https://github.com/apache/incubator-mxnet/pull/11049#discussion_r201769386
##
File path: MKLDNN_README.md
##
@@ -0,0 +1,287 @@
+# Build/Install MXNet with
aaronmarkham commented on a change in pull request #11049: Add linux and macos
MKLDNN Building Instruction
URL: https://github.com/apache/incubator-mxnet/pull/11049#discussion_r201766968
##
File path: MKLDNN_README.md
##
@@ -0,0 +1,229 @@
+# Build/Install MXNet with
aaronmarkham commented on a change in pull request #11049: Add linux and macos
MKLDNN Building Instruction
URL: https://github.com/apache/incubator-mxnet/pull/11049#discussion_r201767766
##
File path: MKLDNN_README.md
##
@@ -0,0 +1,287 @@
+# Build/Install MXNet with
aaronmarkham commented on a change in pull request #11049: Add linux and macos
MKLDNN Building Instruction
URL: https://github.com/apache/incubator-mxnet/pull/11049#discussion_r201767466
##
File path: MKLDNN_README.md
##
@@ -0,0 +1,229 @@
+# Build/Install MXNet with
szha commented on issue #11616: Flaky test test_gluon.test_export
URL:
https://github.com/apache/incubator-mxnet/issues/11616#issuecomment-404243226
`export MXNET_GPU_MEM_POOL_RESERVE=100` should effectively disable the pool.
haojin2 commented on a change in pull request #11631: Fix batchnorm problem
with sparse matrices when fix_gamma=True
URL: https://github.com/apache/incubator-mxnet/pull/11631#discussion_r201770974
##
File path: src/operator/nn/batch_norm.cc
##
@@ -452,19 +453,56 @@ static
haojin2 commented on issue #11647: Sparse input matrices are not supported in
batchnorm operator when fix_gamma=True
URL:
https://github.com/apache/incubator-mxnet/issues/11647#issuecomment-404242662
#11631
This is an
haojin2 commented on issue #11647: Sparse input matrices are not supported in
batchnorm operator when fix_gamma=True
URL:
https://github.com/apache/incubator-mxnet/issues/11647#issuecomment-404242519
Suggested label: Feature Request
haojin2 opened a new issue #11647: Sparse input matrices are not supported in
batchnorm operator when fix_gamma=True
URL: https://github.com/apache/incubator-mxnet/issues/11647
As title.
This is an automated message from the
golo314 commented on issue #11543: Seg. Fault installing cpp-package
URL:
https://github.com/apache/incubator-mxnet/issues/11543#issuecomment-404240855
The problem was with OpenBLAS. I installed ATLAS and was able to build.
golo314 closed issue #11543: Seg. Fault installing cpp-package
URL: https://github.com/apache/incubator-mxnet/issues/11543
This is an automated message from the Apache Git Service.
To respond to the message, please log on
szha opened a new issue #11646: download function should avoid race-condition
URL: https://github.com/apache/incubator-mxnet/issues/11646
Currently the download function would directly write to target file path,
which causes file integrity problem when multiple processes use the same logic
marcoabreu edited a comment on issue #11616: Flaky test test_gluon.test_export
URL:
https://github.com/apache/incubator-mxnet/issues/11616#issuecomment-404238313
But yeah, we are regularly testing MXNet CPU on windows and don't seem to
encounter many problems. I also wouldn't know much
apeforest commented on issue #11620: LBSGD does not have momentum
URL:
https://github.com/apache/incubator-mxnet/issues/11620#issuecomment-404238513
@sandeep-krishnamurthy Please label this issue _Example_, _Good for first
time_
marcoabreu commented on issue #11616: Flaky test test_gluon.test_export
URL:
https://github.com/apache/incubator-mxnet/issues/11616#issuecomment-404238313
But yeah, we are regularly testing MXNet CPU on windows and don't seem to
encounter many problems. I also wouldn't know much reason
eric-haibin-lin closed pull request #11577: Fix flaky test
test_gluon.test_hybrid_static_memory_switching
URL: https://github.com/apache/incubator-mxnet/pull/11577
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake
apeforest commented on issue #11634: website install page query string needed
for devices
URL:
https://github.com/apache/incubator-mxnet/issues/11634#issuecomment-404237681
@sandeep-krishnamurthy Please label this _Feature Request_
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 b264f6f Fix flaky test
marcoabreu commented on issue #11616: Flaky test test_gluon.test_export
URL:
https://github.com/apache/incubator-mxnet/issues/11616#issuecomment-404237611
I tried, what you described was exactly my observation: you run out of cuda
memory if you run MXNet in parallel. I didn't know we had
apeforest commented on issue #11635: Aborted python session when using nd.concat
URL:
https://github.com/apache/incubator-mxnet/issues/11635#issuecomment-404237389
@alireza202 I think the reason is because you are concatenating on the wrong
axis. Numpy would issue an error if you do the
szha commented on issue #11616: Flaky test test_gluon.test_export
URL:
https://github.com/apache/incubator-mxnet/issues/11616#issuecomment-404235872
Users don't directly deal with CUDA. They interact with CUDA through MXNet.
MXNet has internal memory pool to hold previously allocated
sandeep-krishnamurthy commented on issue #11645: Cannot train SmileCNN with
Keras-MXNet
URL:
https://github.com/apache/incubator-mxnet/issues/11645#issuecomment-404235388
@aaronmarkham - Can you please open issue with keras-mxnet repo? Looks like
issue in MXNet backend of Keras.
apeforest edited a comment on issue #11645: Cannot train SmileCNN with
Keras-MXNet
URL:
https://github.com/apache/incubator-mxnet/issues/11645#issuecomment-404233406
@kalyc Can you please help to investigate this issue? Thx
apeforest commented on issue #11645: Cannot train SmileCNN with Keras-MXNet
URL:
https://github.com/apache/incubator-mxnet/issues/11645#issuecomment-404233642
@sandeep-krishnamurthy @nswamy Please label this issue _Keras_
apeforest commented on issue #11645: Cannot train SmileCNN with Keras-MXNet
URL:
https://github.com/apache/incubator-mxnet/issues/11645#issuecomment-404233406
@kalyc Can you please help to check this example? Thx
This is an
marcoabreu commented on issue #11616: Flaky test test_gluon.test_export
URL:
https://github.com/apache/incubator-mxnet/issues/11616#issuecomment-404230989
Every process can manage it's own GPU memory - cuda supports multi tenancy
out of the box. You're probably only going to run into out
marcoabreu commented on issue #11640: Write a new operator by myself,error says
Backward is not implemented.but i write it indeed
URL:
https://github.com/apache/incubator-mxnet/issues/11640#issuecomment-404228516
For these kind of things I'd like to to invite a thread at discuss.mxnet.io
szha commented on issue #11616: Flaky test test_gluon.test_export
URL:
https://github.com/apache/incubator-mxnet/issues/11616#issuecomment-404227932
OK. Following this logic, we either have inter-process memory management for
GPU or our CPU users can't run multi-process CPU inference.
marcoabreu edited a comment on issue #11616: Flaky test test_gluon.test_export
URL:
https://github.com/apache/incubator-mxnet/issues/11616#issuecomment-404226812
My stand on that would be that our tests should all be completely self
contained and thus have no other side effects that
marcoabreu commented on issue #11616: Flaky test test_gluon.test_export
URL:
https://github.com/apache/incubator-mxnet/issues/11616#issuecomment-404226812
My stand on that would be that our tests should all be completely self
contained. What do you rgibk
KellenSunderland opened a new issue #11616: Flaky test test_gluon.test_export
URL: https://github.com/apache/incubator-mxnet/issues/11616
Failed build:
http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/incubator-mxnet/detail/master/1148/pipeline
marcoabreu closed issue #11616: Flaky test test_gluon.test_export
URL: https://github.com/apache/incubator-mxnet/issues/11616
This is an automated message from the Apache Git Service.
To respond to the message, please log on
marcoabreu commented on issue #11616: Flaky test test_gluon.test_export
URL:
https://github.com/apache/incubator-mxnet/issues/11616#issuecomment-404226482
Yes, because everything else is undefined behaviour, right? If we design a
system to be used only in a specific way, we don't
zheng-da commented on issue #11641: [WIP] make CachedOp a normal operator
URL: https://github.com/apache/incubator-mxnet/pull/11641#issuecomment-404226450
@larroy thanks for the review. The code, especially the API design, is still
experimental. I'll let you know when the code is ready for
szha commented on issue #11616: Flaky test test_gluon.test_export
URL:
https://github.com/apache/incubator-mxnet/issues/11616#issuecomment-404226152
We can also ask whether one test should affect another in any way, and if
so, why, and why not managed explicitly.
thomelane commented on issue #11629: deconvolution result different with caffe
result
URL:
https://github.com/apache/incubator-mxnet/issues/11629#issuecomment-404225649
See https://github.com/apache/incubator-mxnet/issues/11203.
Seems like a case of the same bug, found with an even
szha commented on issue #11616: Flaky test test_gluon.test_export
URL:
https://github.com/apache/incubator-mxnet/issues/11616#issuecomment-404225226
@marcoabreu and to you, that statement equals "you are only allowed to run a
single instance of MXNet"?
aaronmarkham opened a new issue #11645: Cannot train SmileCNN with Keras-MXNet
URL: https://github.com/apache/incubator-mxnet/issues/11645
## Description
Error during training of the SmileCNN demo using Keras-MXNet. (Error in
operator conv2d_1/conv2d1)
## Environment info
szha opened a new issue #11644: Jenkins problem
URL: https://github.com/apache/incubator-mxnet/issues/11644
```
Remote call on JNLP4-connect connection from
ip-172-31-25-206.us-west-2.compute.internal/172.31.25.206:52920 failed. The
channel is closing down or has closed down
```
ashkanaev edited a comment on issue #6204: [Fast-RCNN] Proposal operator
failed: too many resources requested for launch
URL:
https://github.com/apache/incubator-mxnet/issues/6204#issuecomment-404207841
Is there any update? Maybe someone developers who responsible for pip build
could
ashkanaev commented on issue #6204: [Fast-RCNN] Proposal operator failed: too
many resources requested for launch
URL:
https://github.com/apache/incubator-mxnet/issues/6204#issuecomment-404207841
Is there any update? Maybe someone from developers who responsible for pip
build could
Ericargus commented on issue #470: More informative debug info for errors in
engine
URL: https://github.com/apache/incubator-mxnet/issues/470#issuecomment-404196747
i @piiswrong hello, i met this error:
mxnet.base.MXNetError: [22:01:36] include/mxnet/././tensor_blob.h:235: Check
Ericargus commented on issue #11640: Write a new operator by myself,error says
Backward is not implemented.but i write it indeed
URL:
https://github.com/apache/incubator-mxnet/issues/11640#issuecomment-404186994
@marcoabreu I'm careless.I solved it by copy similar offical operator,and
Ericargus commented on issue #11640: Write a new operator by myself,error says
Backward is not implemented.but i write it indeed
URL:
https://github.com/apache/incubator-mxnet/issues/11640#issuecomment-404184833
@marcoabreu but i have a new problem.(error reports "TBlob.get: device type
marcoabreu commented on issue #11640: Write a new operator by myself,error says
Backward is not implemented.but i write it indeed
URL:
https://github.com/apache/incubator-mxnet/issues/11640#issuecomment-404184787
Awesome! Do you mind sharing what you did to solve the problem?
marcoabreu closed issue #11640: Write a new operator by myself,error says
Backward is not implemented.but i write it indeed
URL: https://github.com/apache/incubator-mxnet/issues/11640
This is an automated message from the
Ericargus commented on issue #11640: Write a new operator by myself,error says
Backward is not implemented.but i write it indeed
URL:
https://github.com/apache/incubator-mxnet/issues/11640#issuecomment-404184247
@marcoabreu i have solved it
Ericargus commented on issue #11640: Write a new operator by myself,error says
Backward is not implemented.but i write it indeed
URL:
https://github.com/apache/incubator-mxnet/issues/11640#issuecomment-404184093
I have solved it.
ifeherva edited a comment on issue #9253: Numpy Diag in MXNet
URL:
https://github.com/apache/incubator-mxnet/issues/9253#issuecomment-403290348
I am working on replicating np.diag(). Once done will do the full diagonal
operator. This is my first operator so I am learning a lot at the
ifeherva opened a new pull request #11643: Added the diag() operator
URL: https://github.com/apache/incubator-mxnet/pull/11643
## Description ##
Added a new tensor operator called diag() replicating numpy.diag()
## Checklist ##
### Essentials ###
Please feel free to remove
xinyu-intel commented on issue #11049: Add linux and macos MKLDNN Building
Instruction
URL: https://github.com/apache/incubator-mxnet/pull/11049#issuecomment-404165835
Somebody said that installation of your software in Mac OS is really hard
xinyu-intel commented on issue #11049: Add linux and macos MKLDNN Building
Instruction
URL: https://github.com/apache/incubator-mxnet/pull/11049#issuecomment-404163601
@larroy Thanks for your suggestions. Doc for windows is based on cmake, and
cmake instructions for linux and mac will be
larroy commented on a change in pull request #11636: [WIP] set MXNET_DATA_DIR
as base for downloaded models through base.data_dir()
URL: https://github.com/apache/incubator-mxnet/pull/11636#discussion_r201677321
##
File path: docs/faq/env_var.md
##
@@ -126,6 +126,11 @@
larroy commented on a change in pull request #11636: [WIP] set MXNET_DATA_DIR
as base for downloaded models through base.data_dir()
URL: https://github.com/apache/incubator-mxnet/pull/11636#discussion_r201677265
##
File path: docs/faq/env_var.md
##
@@ -126,6 +126,11 @@
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 1c3f920 Bump the publish
larroy commented on issue #11049: Add linux and macos MKLDNN Building
Instruction
URL: https://github.com/apache/incubator-mxnet/pull/11049#issuecomment-404144051
Can we document with CMake? we are trying to migrate to CMake, so would be
great to have instructions that are future proof.
yajiedesign commented on issue #11610: import error
URL:
https://github.com/apache/incubator-mxnet/issues/11610#issuecomment-404133908
@adaa libmxnet.dll
This is an automated message from the Apache Git Service.
To
larroy commented on a change in pull request #11641: [WIP] make CachedOp a
normal operator
URL: https://github.com/apache/incubator-mxnet/pull/11641#discussion_r201625843
##
File path: tests/python/unittest/test_operator.py
##
@@ -6968,6 +6968,39 @@ def
larroy commented on a change in pull request #11641: [WIP] make CachedOp a
normal operator
URL: https://github.com/apache/incubator-mxnet/pull/11641#discussion_r201625451
##
File path: src/imperative/cached_op.h
##
@@ -116,6 +116,24 @@ class CachedOp {
larroy commented on a change in pull request #11641: [WIP] make CachedOp a
normal operator
URL: https://github.com/apache/incubator-mxnet/pull/11641#discussion_r201624918
##
File path: src/imperative/cached_op.cc
##
@@ -1047,6 +1047,105 @@ void CachedOp::Backward(
larroy commented on a change in pull request #11641: [WIP] make CachedOp a
normal operator
URL: https://github.com/apache/incubator-mxnet/pull/11641#discussion_r201624481
##
File path: src/imperative/cached_op.cc
##
@@ -1047,6 +1047,105 @@ void CachedOp::Backward(
larroy commented on a change in pull request #11641: [WIP] make CachedOp a
normal operator
URL: https://github.com/apache/incubator-mxnet/pull/11641#discussion_r201623478
##
File path: src/imperative/cached_op.cc
##
@@ -1047,6 +1047,105 @@ void CachedOp::Backward(
larroy commented on a change in pull request #11641: [WIP] make CachedOp a
normal operator
URL: https://github.com/apache/incubator-mxnet/pull/11641#discussion_r201623330
##
File path: src/imperative/cached_op.cc
##
@@ -1047,6 +1047,105 @@ void CachedOp::Backward(
larroy commented on a change in pull request #11641: [WIP] make CachedOp a
normal operator
URL: https://github.com/apache/incubator-mxnet/pull/11641#discussion_r201623080
##
File path: src/c_api/c_api_symbolic.cc
##
@@ -366,6 +367,36 @@ int
larroy commented on a change in pull request #11641: [WIP] make CachedOp a
normal operator
URL: https://github.com/apache/incubator-mxnet/pull/11641#discussion_r201622787
##
File path: src/c_api/c_api_symbolic.cc
##
@@ -366,6 +367,36 @@ int
larroy commented on a change in pull request #11641: [WIP] make CachedOp a
normal operator
URL: https://github.com/apache/incubator-mxnet/pull/11641#discussion_r201622379
##
File path: src/c_api/c_api_symbolic.cc
##
@@ -366,6 +367,36 @@ int
201 - 300 of 325 matches
Mail list logo