TaoLv commented on issue #16735: Use single-bit for mask in dropout operator
URL: https://github.com/apache/incubator-mxnet/pull/16735#issuecomment-586564858
> Does the slow down come from more computation in the new algorithm or the
sub-optimal implementation?
The new implementation
This is an automated email from the ASF dual-hosted git repository.
haoj pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from d352673 Fix transformer.cu interleaved matmul for cuda arch < 5
(#17596)
add 149975c [numpy][Do Not R
This is an automated email from the ASF dual-hosted git repository.
haoj pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from d352673 Fix transformer.cu interleaved matmul for cuda arch < 5
(#17596)
add 149975c [numpy][Do Not R
haojin2 merged pull request #16865: [numpy][Do Not Review]add op insert
URL: https://github.com/apache/incubator-mxnet/pull/16865
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.
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 b73b3ea Bump the publis
leezu opened a new pull request #17603: Backport #17596
URL: https://github.com/apache/incubator-mxnet/pull/17603
Backport #17596
This is an automated message from the Apache Git Service.
To respond to the message, please log
leezu commented on issue #17017: USE_BLAS=apple broken on OSX 10.15
URL:
https://github.com/apache/incubator-mxnet/issues/17017#issuecomment-586560692
Fixed by https://github.com/apache/incubator-mxnet/pull/17602 for cmake build
-
apeforest commented on a change in pull request #17599: Fixed Embedding op for
LT input
URL: https://github.com/apache/incubator-mxnet/pull/17599#discussion_r379766442
##
File path: src/operator/tensor/indexing_op.h
##
@@ -66,7 +66,7 @@ enum QuantizedEmbeddingOpResource {k
apeforest commented on a change in pull request #16735: Use single-bit for mask
in dropout operator
URL: https://github.com/apache/incubator-mxnet/pull/16735#discussion_r379763372
##
File path: src/operator/nn/dropout-inl.h
##
@@ -152,15 +181,20 @@ class DropoutOp {
leezu opened a new pull request #17602: Fix OS X staticbuild and add tests
URL: https://github.com/apache/incubator-mxnet/pull/17602
## Description ##
Fix OS X staticbuild and add tests based on Github Actions.
This is an 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 9ee4f04 Support broadcast assign for `npi_boolean_mask_assign_tensor`
(#17131)
add d352673 Fix trans
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 9ee4f04 Support broadcast assign for `npi_boolean_mask_assign_tensor`
(#17131)
add d352673 Fix trans
leezu merged pull request #17596: Fix transformer.cu interleaved matmul for
cuda arch < 5
URL: https://github.com/apache/incubator-mxnet/pull/17596
This is an automated message from the Apache Git Service.
To respond to the
reminisce opened a new pull request #17601: Port conv3d dilate > 1 fix to 1.5.x
URL: https://github.com/apache/incubator-mxnet/pull/17601
Port https://github.com/apache/incubator-mxnet/pull/17491 to 1.5.x.
This is an automated
TaoLv commented on issue #17595: MKLDNN incompatibility with large tensor (dim
>= 2^32) data
URL:
https://github.com/apache/incubator-mxnet/issues/17595#issuecomment-586552974
I can reproduce the crash.
This is an automated
This is an automated email from the ASF dual-hosted git repository.
reminisce pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from f619c52 Python 2 cleanup (#17583)
add 9ee4f04 Support broadcast assign for `npi_boolean_mask_assig
reminisce merged pull request #17131: Support broadcast assign for
`npi_boolean_mask_assign_tensor`
URL: https://github.com/apache/incubator-mxnet/pull/17131
This is an automated message from the Apache Git Service.
To respo
reminisce opened a new pull request #17600: Fix Non-ASCII character in docstring
URL: https://github.com/apache/incubator-mxnet/pull/17600
## Description ##
As title. Expected to complement
https://github.com/apache/incubator-mxnet/pull/17593 for fixing
https://github.com/apache/incubat
TaoLv commented on issue #17595: MKLDNN incompatibility with large tensor (dim
>= 2^32) data
URL:
https://github.com/apache/incubator-mxnet/issues/17595#issuecomment-586548992
Thank you for reporting the issue. I will take a look at this. But my
initial thought is that MKL-DNN itself alre
TaoLv commented on issue #17576: Fix storage type infer of softmax backward
URL: https://github.com/apache/incubator-mxnet/pull/17576#issuecomment-586548359
@leeze, Got it. Thank you for the merging.
This is an automated messa
TaoLv commented on a change in pull request #16735: Use single-bit for mask in
dropout operator
URL: https://github.com/apache/incubator-mxnet/pull/16735#discussion_r379713840
##
File path: src/operator/nn/dropout-inl.h
##
@@ -152,15 +181,20 @@ class DropoutOp {
rongzha1 commented on a change in pull request #17265: Add bfloat16
floating-point format support based on AMP
URL: https://github.com/apache/incubator-mxnet/pull/17265#discussion_r379713720
##
File path: 3rdparty/mshadow/mshadow/bfloat.h
##
@@ -0,0 +1,167 @@
+/*!
Revi
connorgoggins commented on issue #17599: Fixed Embedding op for LT input
URL: https://github.com/apache/incubator-mxnet/pull/17599#issuecomment-586543671
@access2rohit great question. Here is an example of when sparseEmbedding
params are used without my fix, and the resulting error (same er
zhreshold commented on issue #17265: Add bfloat16 floating-point format support
based on AMP
URL: https://github.com/apache/incubator-mxnet/pull/17265#issuecomment-586543482
@larroy Do you have idea how to display more logs for the edge tests? It
consistently fail at this stage.
leezu commented on issue #17592: Add cmake build support for macOS static build.
URL: https://github.com/apache/incubator-mxnet/pull/17592#issuecomment-586543059
As per
https://github.com/leezu/mxnet/commit/0bbd44c45f34c37bf70c90635e29a5fcf2e6f0cc/checks/447232316/logs
libtiff 4.1 does not
access2rohit edited a comment on issue #17599: Fixed Embedding op for LT input
URL: https://github.com/apache/incubator-mxnet/pull/17599#issuecomment-586540131
@connorgoggins can you paste output of a case where sparseEmbedding params
are used ? Are both used in the same example use case pr
access2rohit commented on issue #17599: Fixed Embedding op for LT input
URL: https://github.com/apache/incubator-mxnet/pull/17599#issuecomment-586540131
@connorgoggins can you paste output of a case where sparseEmbedding params
are used ?
---
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 40c0c54 Fix storage type infer of softmax backward (#17576)
add f619c52 Python 2 cleanup (#17583)
No
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 40c0c54 Fix storage type infer of softmax backward (#17576)
add f619c52 Python 2 cleanup (#17583)
No
leezu merged pull request #17583: Python 2 cleanup
URL: https://github.com/apache/incubator-mxnet/pull/17583
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use
leezu commented on issue #17596: Fix transformer.cu interleaved matmul for cuda
arch < 5
URL: https://github.com/apache/incubator-mxnet/pull/17596#issuecomment-586539281
Verified this patch by finetuning Bert on P2 instance.
Verification was initially blocked / delayed by
https://gi
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 39b158f quantile_scalar (#17572)
add 40c0c54 Fix storage type infer of softmax backward (#17576)
No
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 39b158f quantile_scalar (#17572)
add 40c0c54 Fix storage type infer of softmax backward (#17576)
No
leezu merged pull request #17576: Fix storage type infer of softmax backward
URL: https://github.com/apache/incubator-mxnet/pull/17576
This is an automated message from the Apache Git Service.
To respond to the message, pleas
D-Roberts commented on issue #17590: [numpy] Implement Weibull backward
URL: https://github.com/apache/incubator-mxnet/pull/17590#issuecomment-586538103
@haojin2 @xidulu Ready for review.
This is an automated message from the
leezu commented on issue #17576: Fix storage type infer of softmax backward
URL: https://github.com/apache/incubator-mxnet/pull/17576#issuecomment-586537272
@TaoLv could you next time revert the buggy commit? I spent some time today
to track down the bug, which could have been avoided.
---
connorgoggins commented on issue #17599: Fixed Embedding op for LT input
URL: https://github.com/apache/incubator-mxnet/pull/17599#issuecomment-586535371
@mxnet-label-bot add [pr-awaiting-review]
This is an automated message f
connorgoggins opened a new pull request #17599: Fixed Embedding op for LT input
URL: https://github.com/apache/incubator-mxnet/pull/17599
## Description ##
The Embedding op was previously breaking on large tensor (dimension >= 2^32)
data. With the following input:
```
nd.Embedding(
leezu commented on a change in pull request #17592: Add cmake build support for
macOS static build.
URL: https://github.com/apache/incubator-mxnet/pull/17592#discussion_r379702140
##
File path: tools/dependencies/curl.sh
##
@@ -35,6 +35,7 @@ if [[ ! -f $DEPS_PATH/lib/libcu
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 fe0b694 Bump the publis
apeforest commented on a change in pull request #16735: Use single-bit for mask
in dropout operator
URL: https://github.com/apache/incubator-mxnet/pull/16735#discussion_r379688573
##
File path: src/operator/nn/dropout-inl.h
##
@@ -152,15 +181,20 @@ class DropoutOp {
apeforest edited a comment on issue #16735: Use single-bit for mask in dropout
operator
URL: https://github.com/apache/incubator-mxnet/pull/16735#issuecomment-586519348
> Does the slow down come from more computation in the new algorithm or the
sub-optimal implementation?
@PatricZhao Th
apeforest commented on issue #16735: Use single-bit for mask in dropout operator
URL: https://github.com/apache/incubator-mxnet/pull/16735#issuecomment-586519348
> Does the slow down come from more computation in the new algorithm or the
sub-optimal implementation?
@PatricZhao The slowdo
apeforest commented on a change in pull request #16735: Use single-bit for mask
in dropout operator
URL: https://github.com/apache/incubator-mxnet/pull/16735#discussion_r379688573
##
File path: src/operator/nn/dropout-inl.h
##
@@ -152,15 +181,20 @@ class DropoutOp {
haojin2 commented on a change in pull request #16865: [numpy][Do Not Review]add
op insert
URL: https://github.com/apache/incubator-mxnet/pull/16865#discussion_r379682542
##
File path: src/operator/numpy/np_insert_op_tensor-inl.h
##
@@ -0,0 +1,229 @@
+/*
+ * Licensed to the
haojin2 commented on a change in pull request #16865: [numpy][Do Not Review]add
op insert
URL: https://github.com/apache/incubator-mxnet/pull/16865#discussion_r379682391
##
File path: src/operator/numpy/np_insert_op_slice-inl.h
##
@@ -0,0 +1,200 @@
+/*
+ * Licensed to the
haojin2 commented on a change in pull request #16865: [numpy][Do Not Review]add
op insert
URL: https://github.com/apache/incubator-mxnet/pull/16865#discussion_r379682473
##
File path: src/operator/numpy/np_insert_op_tensor-inl.h
##
@@ -0,0 +1,229 @@
+/*
+ * Licensed to the
haojin2 commented on a change in pull request #16865: [numpy][Do Not Review]add
op insert
URL: https://github.com/apache/incubator-mxnet/pull/16865#discussion_r379682345
##
File path: src/operator/numpy/np_insert_op_slice-inl.h
##
@@ -0,0 +1,200 @@
+/*
+ * Licensed to the
haojin2 commented on a change in pull request #16865: [numpy][Do Not Review]add
op insert
URL: https://github.com/apache/incubator-mxnet/pull/16865#discussion_r379682486
##
File path: src/operator/numpy/np_insert_op_tensor-inl.h
##
@@ -0,0 +1,229 @@
+/*
+ * Licensed to the
haojin2 commented on a change in pull request #16865: [numpy][Do Not Review]add
op insert
URL: https://github.com/apache/incubator-mxnet/pull/16865#discussion_r379682089
##
File path: src/operator/numpy/np_insert_op-inl.h
##
@@ -0,0 +1,371 @@
+/*
+ * Licensed to the Apache
haojin2 commented on a change in pull request #16865: [numpy][Do Not Review]add
op insert
URL: https://github.com/apache/incubator-mxnet/pull/16865#discussion_r379682109
##
File path: src/operator/numpy/np_insert_op-inl.h
##
@@ -0,0 +1,371 @@
+/*
+ * Licensed to the Apache
haojin2 commented on a change in pull request #16865: [numpy][Do Not Review]add
op insert
URL: https://github.com/apache/incubator-mxnet/pull/16865#discussion_r379681197
##
File path: src/operator/numpy/np_insert_op_tensor.cc
##
@@ -0,0 +1,152 @@
+/*
+ * Licensed to the Ap
haojin2 commented on issue #17590: [WIP][numpy] Implement Weibull backward
URL: https://github.com/apache/incubator-mxnet/pull/17590#issuecomment-586505702
@D-Roberts Thanks for your contribution! Please let @xidulu and/or me know
when you think this PR is ready for a review.
-
This is an automated email from the ASF dual-hosted git repository.
haoj pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 8d887ca add isposinf isneginf isfinite (#17563)
add 39b158f quantile_scalar (#17572)
No new revisions
haojin2 merged pull request #17572: [Numpy] Quantile/Percentile-scalar support
URL: https://github.com/apache/incubator-mxnet/pull/17572
This is an automated message from the Apache Git Service.
To respond to the message, pl
This is an automated email from the ASF dual-hosted git repository.
haoj pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 1c07771 fix CD and remove leftover from #15990 (#17551)
add 8d887ca add isposinf isneginf isfinite (#17
haojin2 merged pull request #17563: [Numpy] add isposinf isneginf isfinite
URL: https://github.com/apache/incubator-mxnet/pull/17563
This is an automated message from the Apache Git Service.
To respond to the message, please
leezu closed pull request #17598: Add OS X Staticbuild CI based on Github
Actions
URL: https://github.com/apache/incubator-mxnet/pull/17598
This is an automated message from the Apache Git Service.
To respond to the message,
leezu commented on issue #17598: Add OS X Staticbuild CI based on Github Actions
URL: https://github.com/apache/incubator-mxnet/pull/17598#issuecomment-586497521
Closing to avoid trigger Jenkins on every push. You can track status at
above link
-
leezu commented on issue #17598: Add OS X Staticbuild CI based on Github Actions
URL: https://github.com/apache/incubator-mxnet/pull/17598#issuecomment-586496421
WIP: See build status at https://github.com/leezu/mxnet/actions
leezu opened a new pull request #17598: Add OS X Staticbuild CI based on Github
Actions
URL: https://github.com/apache/incubator-mxnet/pull/17598
## Description ##
Add OS X Staticbuild CI based on Github Actions
This is an
ChaiBapchya commented on issue #17597: move stacktrance print to console rather
than system.err
URL: https://github.com/apache/incubator-mxnet/pull/17597#issuecomment-586492046
@anirudh2290 @access2rohit @marcoabreu
Since you reviewed the previous PR, thought you could take a look at th
ChaiBapchya commented on issue #17597: move stacktrance print to console rather
than system.err
URL: https://github.com/apache/incubator-mxnet/pull/17597#issuecomment-586491889
@mxnet-label-bot add [pr-awaiting-review]
This i
ChaiBapchya opened a new pull request #17597: move stacktrance print to console
rather than system.err
URL: https://github.com/apache/incubator-mxnet/pull/17597
## Description ##
Right now with #17465 if the build fails, we print the stacktrace using
`error.printStackTrace()` function
access2rohit edited a comment on issue #17595: MKLDNN incompatibility with
large tensor (dim >= 2^32) data
URL:
https://github.com/apache/incubator-mxnet/issues/17595#issuecomment-586451995
@connorgoggins thanks for bringing this up
@PatricZhao @TaoLv looks like blas=MKL/openblas/
leezu commented on issue #16899: Enable MKL-DNN in pip packages
URL: https://github.com/apache/incubator-mxnet/pull/16899#issuecomment-586477909
@TaoLv Sheng just merged fixes for the CD pipeline. Let's wait 24 hours to
verify CD works on master branch. Then we can merge this PR.
-
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 0f35489 use py3 for kvstore tests (#17593)
add 1c07771 fix CD and remove leftover from #15990 (#175
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 0f35489 use py3 for kvstore tests (#17593)
add 1c07771 fix CD and remove leftover from #15990 (#175
szha merged pull request #17551: [CD] fix CD and remove leftover from #15990
URL: https://github.com/apache/incubator-mxnet/pull/17551
This is an automated message from the Apache Git Service.
To respond to the message, pleas
leezu opened a new pull request #17596: Fix transformer.cu interleaved matmul
for cuda arch < 5
URL: https://github.com/apache/incubator-mxnet/pull/17596
## Description ##
`cublasGemmBatchedEx` is only supported for GPU with architecture
capabilities equal or greater than 5.0.
Fi
eric-haibin-lin commented on issue #1137: cublas error when using like mx.nd.***
URL:
https://github.com/apache/incubator-mxnet/issues/1137#issuecomment-586459620
or just put a `mx.nd.waitall()`
This is an automated message f
access2rohit edited a comment on issue #17595: MKLDNN incompatibility with
large tensor (dim >= 2^32) data
URL:
https://github.com/apache/incubator-mxnet/issues/17595#issuecomment-586451995
@connorgoggins thanks for bringing this up
@PatricZhao @TaoLv looks like blas=MKL/openblas/
access2rohit commented on issue #17595: MKLDNN incompatibility with large
tensor (dim >= 2^32) data
URL:
https://github.com/apache/incubator-mxnet/issues/17595#issuecomment-586451995
@PatricZhao @TaoLv looks like blas=MKL/openblas/none(mnative mxnet) and
MKLDNN=OFF is supporting gemm on
apeforest commented on issue #17595: MKLDNN incompatibility with large tensor
(dim >= 2^32) data
URL:
https://github.com/apache/incubator-mxnet/issues/17595#issuecomment-586443098
@PatricZhao Could your team please take a look at this? Thanks!
-
access2rohit commented on issue #17595: MKLDNN incompatibility with large
tensor (dim >= 2^32) data
URL:
https://github.com/apache/incubator-mxnet/issues/17595#issuecomment-586439922
@mxnet-label-bot add [MKLDNN]
This is an
connorgoggins opened a new issue #17595: MKLDNN incompatibility with large
tensor (dim >= 2^32) data
URL: https://github.com/apache/incubator-mxnet/issues/17595
## Description
While testing individual ops for large tensor (dimension >= 2^32) input
functionality, I found an error in MKLD
zhreshold commented on a change in pull request #17566: [WIP] Discussion on
merging BMXNet 2 contributions
URL: https://github.com/apache/incubator-mxnet/pull/17566#discussion_r379594847
##
File path: .gitlab-ci.yml
##
@@ -0,0 +1,71 @@
+stages:
Review comment:
this f
zhreshold commented on a change in pull request #17566: [WIP] Discussion on
merging BMXNet 2 contributions
URL: https://github.com/apache/incubator-mxnet/pull/17566#discussion_r379595075
##
File path: .gitmodules
##
@@ -29,3 +29,6 @@
[submodule "3rdparty/nvidia_cub"]
zhreshold commented on a change in pull request #17566: [WIP] Discussion on
merging BMXNet 2 contributions
URL: https://github.com/apache/incubator-mxnet/pull/17566#discussion_r379595336
##
File path: README.md
##
@@ -1,116 +1,136 @@
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
aaronmarkham opened a new pull request #17594: upgrade sphinx and autodocsumm
URL: https://github.com/apache/incubator-mxnet/pull/17594
## Description ##
Rollback of #17561
A new version of autodocsumm came out. It supports the most recent version
of Sphinx.
This reverts th
mjdenkowski commented on issue #17559: [MXNET-1446] Quantization: intgemm
matrix multiply wrappers
URL: https://github.com/apache/incubator-mxnet/pull/17559#issuecomment-586425627
Hi @TaoLv, thanks for taking a look at this! We understand your comments
about MXNet already having multiple
apeforest edited a comment on issue #16735: Use single-bit for mask in dropout
operator
URL: https://github.com/apache/incubator-mxnet/pull/16735#issuecomment-586062120
@roywei Using the test script in
https://github.com/apache/incubator-mxnet/pull/13896
Build | runtime (before) | r
leezu edited a comment on issue #16408: Add MXNet Ops for fast multihead
attention
URL: https://github.com/apache/incubator-mxnet/pull/16408#issuecomment-586418793
There are two separate problems. `cublasGemmStridedBatchedEx` is buggy and
has fixes in 10.2. But `cublasGemmStridedBatchedEx`
leezu commented on issue #16408: Add MXNet Ops for fast multihead attention
URL: https://github.com/apache/incubator-mxnet/pull/16408#issuecomment-586418793
No. My comment was wrong
This is an automated message from the Apache
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 978b5ea Bump the publis
larroy commented on issue #16408: Add MXNet Ops for fast multihead attention
URL: https://github.com/apache/incubator-mxnet/pull/16408#issuecomment-586415593
So it's related to cuda and not the arch?
This is an automated messa
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 8438d98 Implement all miscellaneous ops (#17511)
add 0f35489 use py3 for kvstore tests (#17593)
No
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 8438d98 Implement all miscellaneous ops (#17511)
add 0f35489 use py3 for kvstore tests (#17593)
No
szha merged pull request #17593: [CI] use py3 for kvstore tests
URL: https://github.com/apache/incubator-mxnet/pull/17593
This is an automated message from the Apache Git Service.
To respond to the message, please log on to G
szha commented on a change in pull request #17265: Add bfloat16 floating-point
format support based on AMP
URL: https://github.com/apache/incubator-mxnet/pull/17265#discussion_r379517986
##
File path: 3rdparty/mshadow/mshadow/bfloat.h
##
@@ -0,0 +1,167 @@
+/*!
Review c
TaoLv commented on a change in pull request #17265: Add bfloat16 floating-point
format support based on AMP
URL: https://github.com/apache/incubator-mxnet/pull/17265#discussion_r379484516
##
File path: 3rdparty/mshadow/mshadow/bfloat.h
##
@@ -0,0 +1,167 @@
+/*!
Review
TaoLv commented on issue #17265: Add bfloat16 floating-point format support
based on AMP
URL: https://github.com/apache/incubator-mxnet/pull/17265#issuecomment-586329605
Hi @leezu, @larroy, this PR can pass the builds for ARM but always hit the
time out of the test.
http://jenkins.mx
TaoLv commented on a change in pull request #17318: Enable MKL-DNN
FullyConnected backward
URL: https://github.com/apache/incubator-mxnet/pull/17318#discussion_r379465641
##
File path: tests/cpp/include/test_mkldnn.h
##
@@ -330,7 +330,7 @@ inline void PrintVerifyMsg(const
TaoLv commented on a change in pull request #17318: Enable MKL-DNN
FullyConnected backward
URL: https://github.com/apache/incubator-mxnet/pull/17318#discussion_r379465354
##
File path: tests/cpp/include/test_mkldnn.h
##
@@ -63,24 +63,24 @@ struct TestArrayShapes {
};
/
sl1pkn07 commented on issue #17514: C++-Package examples broken
URL:
https://github.com/apache/incubator-mxnet/issues/17514#issuecomment-586274460
try with
~~~
-DCUDA_HOST_COMPILER=/usr/bin/cc-8 \
-DCMAKE_C_COMPILER=/usr/bin/cc-8 \
-DCMAKE_C_COMPILER_AR=/usr/bin/
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 3fca7d8 Bump the publis
ciyongch commented on a change in pull request #17318: Enable MKL-DNN
FullyConnected backward
URL: https://github.com/apache/incubator-mxnet/pull/17318#discussion_r379387996
##
File path: tests/cpp/include/test_mkldnn.h
##
@@ -63,24 +63,24 @@ struct TestArrayShapes {
};
ciyongch commented on a change in pull request #17318: Enable MKL-DNN
FullyConnected backward
URL: https://github.com/apache/incubator-mxnet/pull/17318#discussion_r379389375
##
File path: tests/cpp/include/test_mkldnn.h
##
@@ -63,24 +63,24 @@ struct TestArrayShapes {
};
ChiaraXian commented on issue #8148: How do I interpret the JSON compute graph?
URL:
https://github.com/apache/incubator-mxnet/issues/8148#issuecomment-586195655
@aidan-plenert-macdonald Thanks a lot!!! Thank you for your help!!!
pengzhao-intel commented on a change in pull request #17318: Enable MKL-DNN
FullyConnected backward
URL: https://github.com/apache/incubator-mxnet/pull/17318#discussion_r379349568
##
File path: tests/cpp/include/test_mkldnn.h
##
@@ -330,7 +330,7 @@ inline void PrintVerifyM
1 - 100 of 101 matches
Mail list logo