eric-haibin-lin opened a new issue #12729: Feature request: operator "sparse
softmax"
URL: https://github.com/apache/incubator-mxnet/issues/12729
The sparse softmax is useful for the transformer model:
https://www.tensorflow.org/api_docs/python/tf/sparse_softmax
apeforest commented on issue #12637: [MXNET-912] Refactoring ctc loss operator
URL: https://github.com/apache/incubator-mxnet/pull/12637#issuecomment-426515746
@anirudh2290 Please review the change to ctc_include folder.
This
apeforest commented on issue #11742: [MXNET-623] Fixing an integer overflow bug
in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#issuecomment-426511498
@lebeg Changed unit test style following your suggestion. Please review
again. Thanks
sandeep-krishnamurthy commented on a change in pull request #12625: [MXNET-979]
Add fix_beta support in BatchNorm
URL: https://github.com/apache/incubator-mxnet/pull/12625#discussion_r222179922
##
File path: tests/python/gpu/test_operator_gpu.py
##
@@ -303,35 +303,35 @@
sandeep-krishnamurthy commented on a change in pull request #12625: [MXNET-979]
Add fix_beta support in BatchNorm
URL: https://github.com/apache/incubator-mxnet/pull/12625#discussion_r222178605
##
File path: src/operator/nn/batch_norm.cc
##
@@ -164,18 +164,42 @@ void
sandeep-krishnamurthy commented on issue #12625: [MXNET-979] Add fix_beta
support in BatchNorm
URL: https://github.com/apache/incubator-mxnet/pull/12625#issuecomment-426506686
@szha - cudnn batchnorm is covered in this PR.
Since SyncBatchNorm is independent of BatchNorm operator, I
sandeep-krishnamurthy opened a new issue #12728: support fix_beta in
sync_batchnorm operator
URL: https://github.com/apache/incubator-mxnet/issues/12728
Add fix_beta parameter in sync_batchnorm. Related PR for adding in BatchNorm
- https://github.com/apache/incubator-mxnet/pull/12625
apivovarov commented on issue #12711: change numpy version to 1.15.2 in python
install_requires
URL: https://github.com/apache/incubator-mxnet/pull/12711#issuecomment-426490649
Good catch. I just made it `numpy<=1.15.2,>=1.8.2` in all files.
vrakesh commented on issue #12255: Pretty high cpu load when import mxnet
URL:
https://github.com/apache/incubator-mxnet/issues/12255#issuecomment-426483000
@fighting-liu
similar results to above were observed running your example
```bash
>>> def mxnet_worker():
...
nswamy commented on a change in pull request #12636: [MXNET-982] Provide
example to illustrate usage of CSVIter in C++ API
URL: https://github.com/apache/incubator-mxnet/pull/12636#discussion_r222157803
##
File path: cpp-package/README.md
##
@@ -1,42 +1,21 @@
# MXNet C++
apivovarov commented on issue #12711: change numpy version to 1.15.2 in python
install_requires
URL: https://github.com/apache/incubator-mxnet/pull/12711#issuecomment-426480328
Default numpy version in The Python Package Index (PyPI) is 1.15.2
https://pypi.org/project/numpy/
pip/pip3
nswamy commented on issue #12641: Add docker file which includes sample jupyter
notebook (#12611)
URL: https://github.com/apache/incubator-mxnet/pull/12641#issuecomment-426479708
you can use the code in this script
thomelane commented on issue #12392: How to use BlockGrad with weight decay
URL:
https://github.com/apache/incubator-mxnet/issues/12392#issuecomment-426479639
Answering your question a little more directly (if you haven't moved over to
Gluon), you could set the weight decay to zero for
thomelane commented on issue #12392: How to use BlockGrad with weight decay
URL:
https://github.com/apache/incubator-mxnet/issues/12392#issuecomment-426478951
Hi @VanDavv,
Given the original context of transfer learning, with Gluon you can just
provide the parameters for the last
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 78b56a9 Bump the publish
thomelane commented on issue #12547: [Scala] [macOS] GPU / CUDA?
URL:
https://github.com/apache/incubator-mxnet/issues/12547#issuecomment-426475940
What issue were you having with MXNet on Ubuntu @mariussoutier? Check out
AWS EC2 DLAMI for Ubuntu that has MXNet and CUDA pre-installed if
lebeg commented on a change in pull request #12726: Fix failing GPU test on
single GPU host (kvstore) #10977
URL: https://github.com/apache/incubator-mxnet/pull/12726#discussion_r222153443
##
File path: tests/python/gpu/test_kvstore_gpu.py
##
@@ -59,13 +59,17 @@ def
lebeg commented on a change in pull request #12726: Fix failing GPU test on
single GPU host (kvstore) #10977
URL: https://github.com/apache/incubator-mxnet/pull/12726#discussion_r222153443
##
File path: tests/python/gpu/test_kvstore_gpu.py
##
@@ -59,13 +59,17 @@ def
larroy commented on a change in pull request #12726: Fix failing GPU test on
single GPU host (kvstore) #10977
URL: https://github.com/apache/incubator-mxnet/pull/12726#discussion_r222152057
##
File path: tests/python/gpu/test_kvstore_gpu.py
##
@@ -59,13 +59,17 @@ def
KellenSunderland commented on issue #12656: [MXNET-703] Show perf info for
TensorRT during tests
URL: https://github.com/apache/incubator-mxnet/pull/12656#issuecomment-426472284
No by default it's disabled and this enables it. Terrible API.
On Wed, Oct 3, 2018, 12:32 AM Anton
lebeg commented on a change in pull request #11742: [MXNET-623] Fixing an
integer overflow bug in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#discussion_r222146174
##
File path: tests/nightly/test_large_array.py
##
@@ -0,0 +1,31 @@
+# Licensed
nswamy commented on a change in pull request #12639: [MXNET-947] Expand scala
imclassification example with resnet
URL: https://github.com/apache/incubator-mxnet/pull/12639#discussion_r222143828
##
File path:
nswamy commented on a change in pull request #12639: [MXNET-947] Expand scala
imclassification example with resnet
URL: https://github.com/apache/incubator-mxnet/pull/12639#discussion_r222142030
##
File path:
nswamy commented on a change in pull request #12639: [MXNET-947] Expand scala
imclassification example with resnet
URL: https://github.com/apache/incubator-mxnet/pull/12639#discussion_r222143289
##
File path:
apeforest commented on a change in pull request #11742: [MXNET-623] Fixing an
integer overflow bug in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#discussion_r222143271
##
File path: src/operator/nn/upsampling-inl.h
##
@@ -48,8 +48,8 @@ enum
apeforest commented on a change in pull request #11742: [MXNET-623] Fixing an
integer overflow bug in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#discussion_r222142443
##
File path: tests/nightly/test_large_array.py
##
@@ -0,0 +1,31 @@
+#
apeforest commented on a change in pull request #11742: [MXNET-623] Fixing an
integer overflow bug in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#discussion_r222142443
##
File path: tests/nightly/test_large_array.py
##
@@ -0,0 +1,31 @@
+#
apeforest commented on a change in pull request #11742: [MXNET-623] Fixing an
integer overflow bug in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#discussion_r222142164
##
File path: src/operator/tensor/ordering_op-inl.h
##
@@ -378,16 +378,19 @@
vrakesh edited a comment on issue #12255: Pretty high cpu load when import mxnet
URL:
https://github.com/apache/incubator-mxnet/issues/12255#issuecomment-426460487
@kardoszc using the example provided by you , i was able to get the same
result on run 1 on a MAC, on consecutive runs of the
lebeg commented on a change in pull request #11742: [MXNET-623] Fixing an
integer overflow bug in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#discussion_r222141043
##
File path: src/operator/nn/upsampling-inl.h
##
@@ -48,8 +48,8 @@ enum
apeforest commented on a change in pull request #11742: [MXNET-623] Fixing an
integer overflow bug in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#discussion_r222140090
##
File path: src/operator/nn/upsampling-inl.h
##
@@ -48,8 +48,8 @@ enum
vrakesh commented on issue #12255: Pretty high cpu load when import mxnet
URL:
https://github.com/apache/incubator-mxnet/issues/12255#issuecomment-426460487
@kardoszc using the example provided by you , i was able to get the same
result on run 1 on a MAC, on consecutive runs of the above,
lebeg commented on a change in pull request #12356: [MXNET-860] Use modernized
ranged loops where possible
URL: https://github.com/apache/incubator-mxnet/pull/12356#discussion_r222135483
##
File path: src/imperative/cached_op.cc
##
@@ -658,9 +658,9 @@ void
lebeg commented on a change in pull request #12356: [MXNET-860] Use modernized
ranged loops where possible
URL: https://github.com/apache/incubator-mxnet/pull/12356#discussion_r222135326
##
File path: src/executor/graph_executor.cc
##
@@ -1551,16 +1550,16 @@ static
lebeg commented on a change in pull request #12356: [MXNET-860] Use modernized
ranged loops where possible
URL: https://github.com/apache/incubator-mxnet/pull/12356#discussion_r222134518
##
File path: src/c_api/c_api_symbolic.cc
##
@@ -268,8 +268,8 @@ int
lebeg commented on a change in pull request #12356: [MXNET-860] Use modernized
ranged loops where possible
URL: https://github.com/apache/incubator-mxnet/pull/12356#discussion_r222134743
##
File path: src/common/exec_utils.h
##
@@ -594,11 +594,11 @@ inline nnvm::Graph
lebeg commented on a change in pull request #12356: [MXNET-860] Use modernized
ranged loops where possible
URL: https://github.com/apache/incubator-mxnet/pull/12356#discussion_r222135194
##
File path: src/executor/attach_op_execs_pass.cc
##
@@ -58,12 +58,10 @@ class
lebeg commented on a change in pull request #12356: [MXNET-860] Use modernized
ranged loops where possible
URL: https://github.com/apache/incubator-mxnet/pull/12356#discussion_r222135633
##
File path: src/imperative/cached_op.cc
##
@@ -1283,8 +1283,8 @@ void
thomelane commented on issue #12577: Training with fc and multi-gpu is much
slower than single gpu
URL:
https://github.com/apache/incubator-mxnet/issues/12577#issuecomment-426453274
Hi @liu6381810,
Seems like there's an overhead to using multiple GPUs, and one possible
source is
lebeg commented on a change in pull request #12656: [MXNET-703] Show perf info
for TensorRT during tests
URL: https://github.com/apache/incubator-mxnet/pull/12656#discussion_r222132940
##
File path: ci/docker/runtime_functions.sh
##
@@ -741,7 +741,7 @@
lebeg commented on issue #12725: [Backport] Fix for test_ndarray.test_order
failing on CI (v1.3.x)
URL: https://github.com/apache/incubator-mxnet/pull/12725#issuecomment-426449863
@ankkhedia strange, I see your
lebeg commented on a change in pull request #11742: [MXNET-623] Fixing an
integer overflow bug in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#discussion_r222128077
##
File path: src/operator/convolution_v1-inl.h
##
@@ -335,12 +335,10 @@ class
lebeg commented on a change in pull request #11742: [MXNET-623] Fixing an
integer overflow bug in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#discussion_r222129669
##
File path: tests/nightly/test_large_array.py
##
@@ -0,0 +1,31 @@
+# Licensed
lebeg commented on a change in pull request #11742: [MXNET-623] Fixing an
integer overflow bug in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#discussion_r222128294
##
File path: src/operator/nn/upsampling-inl.h
##
@@ -48,8 +48,8 @@ enum
lebeg commented on a change in pull request #11742: [MXNET-623] Fixing an
integer overflow bug in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#discussion_r222128932
##
File path: src/operator/tensor/ordering_op-inl.h
##
@@ -378,16 +378,19 @@
lebeg commented on a change in pull request #11742: [MXNET-623] Fixing an
integer overflow bug in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#discussion_r222126397
##
File path: src/io/iter_image_recordio_2.cc
##
@@ -328,7 +328,8 @@ inline bool
aaronmarkham commented on a change in pull request #12696: Refine mxnet python
installation
URL: https://github.com/apache/incubator-mxnet/pull/12696#discussion_r222128823
##
File path: docs/install/windows_setup.md
##
@@ -228,7 +228,7 @@ opencv_world341.dll (in OpenCV
thomelane commented on issue #1905: Random scaling in image augmentation
URL:
https://github.com/apache/incubator-mxnet/issues/1905#issuecomment-426447418
Hi @TD,
I've just replied to a similar post
[here](https://github.com/apache/incubator-mxnet/issues/12671) which hopefully
mbrookhart commented on issue #12502: [NGRAPH] MXNet - nGraph initial
integration
URL: https://github.com/apache/incubator-mxnet/pull/12502#issuecomment-426447034
@zheng-da @reminisce Any other concerns?
This is an
thomelane commented on issue #12671: Could you tell me how to use aug_seq at
mx.io.ImageRecordIter?
URL:
https://github.com/apache/incubator-mxnet/issues/12671#issuecomment-426446193
Aside from the above, I'd take a look at Gluon's Transforms. You'll probably
find they're an easier way
thomelane commented on issue #12671: Could you tell me how to use aug_seq at
mx.io.ImageRecordIter?
URL:
https://github.com/apache/incubator-mxnet/issues/12671#issuecomment-426445643
Hi @shawncheeer,
As I understand things, `aug_seq` should be a string referring to the
augmenter
perdasilva commented on issue #12717: Re-enables test_operator.test_dropout
URL: https://github.com/apache/incubator-mxnet/pull/12717#issuecomment-426444971
Done
This is an automated message from the Apache Git Service.
To
zhreshold commented on a change in pull request #12646: ONNX export: Fully
connected operator with no bias
URL: https://github.com/apache/incubator-mxnet/pull/12646#discussion_r222123917
##
File path: python/mxnet/contrib/onnx/mx2onnx/_op_translations.py
##
@@ -214,17
Roshrini commented on issue #12646: ONNX export: Fully connected operator with
no bias
URL: https://github.com/apache/incubator-mxnet/pull/12646#issuecomment-426439862
Looks good to me. @zhreshold Can you please take a look once?
Roshrini commented on issue #12653: ONNX export: Square and ReduceSum operators
URL: https://github.com/apache/incubator-mxnet/pull/12653#issuecomment-426439102
LGTM
This is an automated message from the Apache Git Service.
aaronmarkham commented on a change in pull request #12696: Refine mxnet python
installation
URL: https://github.com/apache/incubator-mxnet/pull/12696#discussion_r222118851
##
File path: docs/install/windows_setup.md
##
@@ -206,13 +207,28 @@ We have installed MXNet core
vrakesh commented on issue #9527: Python Segmentation fault with commit
afd82410165576dec7c673cd0438a1a9d39a9af6
URL:
https://github.com/apache/incubator-mxnet/issues/9527#issuecomment-426436558
@panovr is this issue resolved for you, if so, requesting to close this issue
lebeg commented on a change in pull request #12726: Fix failing GPU test on
single GPU host (kvstore) #10977
URL: https://github.com/apache/incubator-mxnet/pull/12726#discussion_r222112076
##
File path: tests/python/gpu/test_kvstore_gpu.py
##
@@ -59,13 +59,17 @@ def
lebeg commented on a change in pull request #12726: Fix failing GPU test on
single GPU host (kvstore) #10977
URL: https://github.com/apache/incubator-mxnet/pull/12726#discussion_r222113617
##
File path: tests/python/gpu/test_kvstore_gpu.py
##
@@ -21,7 +21,7 @@
import
stu1130 commented on a change in pull request #12696: Refine mxnet python
installation
URL: https://github.com/apache/incubator-mxnet/pull/12696#discussion_r222111717
##
File path: docs/install/windows_setup.md
##
@@ -206,13 +207,28 @@ We have installed MXNet core
larroy opened a new pull request #12726: Fix failing GPU test on single GPU
host (kvstore) #10977
URL: https://github.com/apache/incubator-mxnet/pull/12726
## Description ##
See description
## Checklist ##
### Essentials ###
Please feel free to remove inapplicable items for
This is an automated email from the ASF dual-hosted git repository.
skm pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 42cc960 Fix static / dynamic linking of gperftools and jemalloc
(#12714)
add 5c86143 Disable test
aaronmarkham commented on a change in pull request #12696: Refine mxnet python
installation
URL: https://github.com/apache/incubator-mxnet/pull/12696#discussion_r222108128
##
File path: docs/install/windows_setup.md
##
@@ -206,13 +207,28 @@ We have installed MXNet core
sandeep-krishnamurthy closed pull request #12716: Disable test batchnorm slice
URL: https://github.com/apache/incubator-mxnet/pull/12716
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
lebeg commented on issue #12725: [Backport] Fix for testorder failing on CI
fro v1.3.x backported from Master
URL: https://github.com/apache/incubator-mxnet/pull/12725#issuecomment-426425134
You have a typo in the title ;) I suggest for the title:
```[Backport] Fix for
lebeg commented on a change in pull request #12725: [Backport] Fix for
testorder failing on CI fro v1.3.x backported from Master
URL: https://github.com/apache/incubator-mxnet/pull/12725#discussion_r222105981
##
File path: tests/python/unittest/test_ndarray.py
##
@@
stu1130 commented on issue #12696: Refine mxnet python installation
URL: https://github.com/apache/incubator-mxnet/pull/12696#issuecomment-426421529
@aaronmarkham let me know if you have any question
This is an automated
ankkhedia commented on issue #12725: [Backport] Fix for testorder failing on
CI fro v1.3.x backported from Master
URL: https://github.com/apache/incubator-mxnet/pull/12725#issuecomment-426421406
@marcoabreu @lebeg
This is
ankkhedia opened a new pull request #12725: [Backport] Fix for testorder
failing on CI fro v1.3.x backported from Master
URL: https://github.com/apache/incubator-mxnet/pull/12725
## Description ##
Backporting fix for flaky test: test_ndarray.test_order from master as the
test
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 9ad8c48 Bump the publish
gaurav-gireesh edited a comment on issue #12697: [WIP] [MXNET -1004] Poisson
NegativeLog Likelihood loss
URL: https://github.com/apache/incubator-mxnet/pull/12697#issuecomment-426095278
> @gaurav-gireesh what is "in progress" for this PR? The PR description says
that you have completed
anirudhacharya commented on a change in pull request #12653: ONNX export:
Square and ReduceSum operators
URL: https://github.com/apache/incubator-mxnet/pull/12653#discussion_r222095800
##
File path: python/mxnet/contrib/onnx/mx2onnx/_op_translations.py
##
@@ -2077,3
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 6fb81af [MXNET-637] Multidimensional LSTM example for MXNetR (#12664)
add 42cc960 Fix static /
szha closed pull request #12714: Fix static / dynamic linking of gperftools and
jemalloc
URL: https://github.com/apache/incubator-mxnet/pull/12714
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:
gaurav-gireesh commented on a change in pull request #12697: [WIP] [MXNET
-1004] Poisson NegativeLog Likelihood loss
URL: https://github.com/apache/incubator-mxnet/pull/12697#discussion_r222090478
##
File path: python/mxnet/gluon/loss.py
##
@@ -706,3 +707,64 @@ def
gaurav-gireesh commented on a change in pull request #12697: [WIP] [MXNET
-1004] Poisson NegativeLog Likelihood loss
URL: https://github.com/apache/incubator-mxnet/pull/12697#discussion_r222089775
##
File path: tests/python/unittest/test_loss.py
##
@@ -342,12 +343,45 @@
gaurav-gireesh commented on a change in pull request #12697: [WIP] [MXNET
-1004] Poisson NegativeLog Likelihood loss
URL: https://github.com/apache/incubator-mxnet/pull/12697#discussion_r222089775
##
File path: tests/python/unittest/test_loss.py
##
@@ -342,12 +343,45 @@
vrakesh commented on issue #12724: simplify mac mkldnn build
URL: https://github.com/apache/incubator-mxnet/pull/12724#issuecomment-426405537
@mxnet-label-bot [pr-awaiting-review]
This is an automated message from the Apache
vrakesh commented on issue #12723: [MXNET-716][DO NOT Merge] Scala Benchmark
Extension pack
URL: https://github.com/apache/incubator-mxnet/pull/12723#issuecomment-426405368
@mxnet-label-bot [pr-awaiting-review, Scala]
This
vrakesh commented on issue #12722: Add support for
CUDNN_TENSOR_OP_MATH_ALLOW_CONVERSION
URL: https://github.com/apache/incubator-mxnet/pull/12722#issuecomment-426404435
@mxnet-label-bot [pr-awaiting-review]
This is an
vrakesh commented on issue #12721: [MXNET-716] Adding Scala Inference Benchmarks
URL: https://github.com/apache/incubator-mxnet/pull/12721#issuecomment-426404284
@mxnet-label-bot [scala]
This is an automated message from the
vrakesh commented on issue #12721: [MXNET-716] Adding Scala Inference Benchmarks
URL: https://github.com/apache/incubator-mxnet/pull/12721#issuecomment-426403970
@mxnet-label-bot [pr-awaiting-review]
This is an automated
azai91 opened a new pull request #12724: simplify mac mkldnn build
URL: https://github.com/apache/incubator-mxnet/pull/12724
## Description ##
mxnet-mkldnn on mac requires that openmp be enabled (else performance is
degraded). There are currently instructions specifying that the
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 12bc9a7 Bump the publish
vandanavk commented on issue #12717: Re-enables test_operator.test_dropout
URL: https://github.com/apache/incubator-mxnet/pull/12717#issuecomment-426383187
Please add
> Fixes https://github.com/apache/incubator-mxnet/issues/12329
in the PR description so that the issue is
vandanavk commented on issue #12718: Reenable test_gluon.test_conv
URL: https://github.com/apache/incubator-mxnet/pull/12718#issuecomment-426382707
Please add
> Fixes https://github.com/apache/incubator-mxnet/issues/11687
to automatically close the issue when this PR is
vandanavk commented on a change in pull request #12653: ONNX export: Square and
ReduceSum operators
URL: https://github.com/apache/incubator-mxnet/pull/12653#discussion_r222061870
##
File path: python/mxnet/contrib/onnx/mx2onnx/_op_translations.py
##
@@ -2077,3 +2077,85
piyushghai commented on issue #12721: [MXNET-716] Adding Scala Inference
Benchmarks
URL: https://github.com/apache/incubator-mxnet/pull/12721#issuecomment-426379905
@lanking520 @andrewfayres
This is an automated message
vandanavk commented on a change in pull request #12653: ONNX export: Square and
ReduceSum operators
URL: https://github.com/apache/incubator-mxnet/pull/12653#discussion_r222058465
##
File path: python/mxnet/contrib/onnx/mx2onnx/_op_translations.py
##
@@ -2077,3 +2077,85
eli2014 opened a new issue #12403: roi_align.cc bug?
URL: https://github.com/apache/incubator-mxnet/issues/12403
There maybe a bug in file roi_align.cc.
#Line 177 and Line 178
T roi_width = std::max(roi_end_w - roi_start_w, (T)1.);
T roi_height = std::max(roi_end_h -
vrakesh commented on issue #12720: [R] CNN Tutorial Fix
URL: https://github.com/apache/incubator-mxnet/pull/12720#issuecomment-426373325
@anirudhacharya Requesting to look into the build failure
@mxnet-label-bot [pr-awaiting-review, Docs]
vrakesh commented on issue #12719: fix cnn visualization tutorial
URL: https://github.com/apache/incubator-mxnet/pull/12719#issuecomment-426372968
@mxnet-label-bot [pr-awaiting-review, Docs]
This is an automated message from
vrakesh commented on issue #12718: Reenable test_gluon.test_conv
URL: https://github.com/apache/incubator-mxnet/pull/12718#issuecomment-426372736
@mxnet-label-bot [pr-awaiting-review, Tests]
This is an automated message from
vrakesh commented on issue #12717: Re-enables test_operator.test_dropout
URL: https://github.com/apache/incubator-mxnet/pull/12717#issuecomment-426370451
@mxnet-label-bot [pr-awaiting-review]
This is an automated message
anirudhacharya commented on a change in pull request #12653: ONNX export:
Square and ReduceSum operators
URL: https://github.com/apache/incubator-mxnet/pull/12653#discussion_r222049383
##
File path: python/mxnet/contrib/onnx/mx2onnx/_op_translations.py
##
@@ -2077,3
anirudhacharya commented on a change in pull request #12653: ONNX export:
Square and ReduceSum operators
URL: https://github.com/apache/incubator-mxnet/pull/12653#discussion_r222049383
##
File path: python/mxnet/contrib/onnx/mx2onnx/_op_translations.py
##
@@ -2077,3
apeforest commented on issue #11742: [MXNET-623] Fixing an integer overflow bug
in large NDArray
URL: https://github.com/apache/incubator-mxnet/pull/11742#issuecomment-426366897
@sandeep-krishnamurthy @nswamy This PR is ready for review. Could you please
change the label to
anirudhacharya opened a new pull request #12720: [R] CNN Tutorial Fix
URL: https://github.com/apache/incubator-mxnet/pull/12720
## Description ##
Fix the Cats vs Dogs tutorial for R API.
## Checklist ##
### Essentials ###
Please feel free to remove inapplicable items for
vrakesh commented on issue #12714: Fix static / dynamic linking of gperftools
and jemalloc
URL: https://github.com/apache/incubator-mxnet/pull/12714#issuecomment-426365130
@mxnet-label-bot [pr-awaiting-merge]
This is an
yuxihu commented on issue #12403: roi_align.cc bug?
URL:
https://github.com/apache/incubator-mxnet/issues/12403#issuecomment-426364135
@eli2014 Thanks for your response. We will investigate and verify the
algorithm based on what you provide. Since our algorithm for this operator is
1 - 100 of 134 matches
Mail list logo
