szha commented on a change in pull request #8642: ctypes speed improvement
URL: https://github.com/apache/incubator-mxnet/pull/8642#discussion_r150756308
##
File path: python/mxnet/base.py
##
@@ -213,7 +248,55 @@ def c_array(ctype, values):
>>> x[1]
2.0
"""
szha commented on issue #8642: ctypes speed improvement
URL: https://github.com/apache/incubator-mxnet/pull/8642#issuecomment-344171754
Some explanation on the slowness of the array init constructor can be found
in https://bugs.python.org/issue27926
szha commented on issue #8642: ctypes speed improvement
URL: https://github.com/apache/incubator-mxnet/pull/8642#issuecomment-344171754
Some explanation on the slowness of the array constructor can be found in
https://bugs.python.org/issue27926
piiswrong commented on a change in pull request #8642: ctypes speed improvement
URL: https://github.com/apache/incubator-mxnet/pull/8642#discussion_r150751615
##
File path: python/mxnet/base.py
##
@@ -213,7 +248,55 @@ def c_array(ctype, values):
>>> x[1]
2.0
piiswrong commented on issue #8642: ctypes speed improvement
URL: https://github.com/apache/incubator-mxnet/pull/8642#issuecomment-344165379
Are you sure this really improves performance? Why would assignment be
faster than constructor?
anxiaoxi45 opened a new issue #8646: Bug in FullyConnected???
URL: https://github.com/apache/incubator-mxnet/issues/8646
I built a simple network like this:
` data = mx.symbol.Variable(name="data")
net = mx.symbol.Pooling(data=data, global_pool=True, kernel=(2, 2),
eric-haibin-lin opened a new issue #8645:
gluon.utils.split_and_load(even_split=True) is much slower than even_split=False
URL: https://github.com/apache/incubator-mxnet/issues/8645
Note: Providing complete information in the most concise form is the best
way to get help. This issue
cypw commented on issue #8644: fix bug in TryCrop()
URL: https://github.com/apache/incubator-mxnet/pull/8644#issuecomment-344153970
@zhreshold
If all conditions must be true, then why did you add the `if` in [Line
zhreshold commented on issue #8644: fix bug in TryCrop()
URL: https://github.com/apache/incubator-mxnet/pull/8644#issuecomment-344151444
All conditions must be true.
This is an automated message from the Apache Git Service.
cypw opened a new pull request #8644: fix bug in TryCrop()
URL: https://github.com/apache/incubator-mxnet/pull/8644
The conditions for the TryCrop() is correct.
This is an automated message from the Apache Git Service.
To
cjolivier01 opened a new pull request #8643: independent test-related changes
pulled from broadcast-tuning branch
URL: https://github.com/apache/incubator-mxnet/pull/8643
## Description ##
(Brief description on what this PR is about)
## Checklist ##
### Essentials ###
- [ ]
szha opened a new pull request #8642: ctypes speed improvement
URL: https://github.com/apache/incubator-mxnet/pull/8642
## Description ##
Improvements on ctypes calls for creating arrays. The changes are based on
the following micro benchmarks.
```python
import timeit
# ints
eric-haibin-lin opened a new pull request #8641: Doc updates for sparse
operators
URL: https://github.com/apache/incubator-mxnet/pull/8641
## Description ##
This PR adds many missing operators to the sparse documentation page.
It also add arithmetic functions such as
eric-haibin-lin closed pull request #8641: Doc updates for sparse operators
URL: https://github.com/apache/incubator-mxnet/pull/8641
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 a
asmushetzel commented on issue #8338: master branch cannot build on centos 7
with cuda-8.0
URL:
https://github.com/apache/incubator-mxnet/issues/8338#issuecomment-344131973
So who will drive it to a final resolution. @sxjscience ?
eric-haibin-lin opened a new pull request #8641: Doc update
URL: https://github.com/apache/incubator-mxnet/pull/8641
## Description ##
This PR adds many missing operators to the sparse documentation page.
It also add arithmetic functions such as `nd.sparse.add`, which is almost
analog-cbarber opened a new issue #8640: CustomOp with more than one auxiliary
state segfaults
URL: https://github.com/apache/incubator-mxnet/issues/8640
## Description
A trivial CustomOp in python that uses more than one auxiliary state
variable often causes
a segfault,
zhreshold opened a new pull request #8639: [WIP] Gluon object detection
URL: https://github.com/apache/incubator-mxnet/pull/8639
## Description ##
Pure gluon object detection baselines.
### Targeted object algorithms
By stage
- Single stage: SSD,YOLO, RetinaNet
- Two
thirdwing commented on a change in pull request #8118: [R] various small
changes for efficiency and robustness in R package
URL: https://github.com/apache/incubator-mxnet/pull/8118#discussion_r150715182
##
File path: R-package/R/zzz.R
##
@@ -16,7 +16,10 @@ NULL
.onLoad
bfgray3 commented on a change in pull request #8118: [R] various small changes
for efficiency and robustness in R package
URL: https://github.com/apache/incubator-mxnet/pull/8118#discussion_r150714195
##
File path: R-package/R/model.R
##
@@ -589,27 +587,20 @@
bfgray3 commented on a change in pull request #8118: [R] various small changes
for efficiency and robustness in R package
URL: https://github.com/apache/incubator-mxnet/pull/8118#discussion_r150712851
##
File path: R-package/R/zzz.R
##
@@ -16,7 +16,10 @@ NULL
.onLoad
piiswrong closed pull request #8633: [WIP]Image Augmenter
URL: https://github.com/apache/incubator-mxnet/pull/8633
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 a foreign pull request
This is an automated email from the ASF dual-hosted git repository.
jxie pushed a commit to branch vision
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git
The following commit(s) were added to refs/heads/vision by this push:
new c1de00a [WIP]Image Augmenter (#8633)
szha commented on issue #7458: Return to the upstream CUB
URL: https://github.com/apache/incubator-mxnet/pull/7458#issuecomment-344104789
This issue is closed due to lack of activity in the last 90 days. Feel free
to ping me to reopen if this is still an active issue. Thanks!
Also, do
szha closed issue #7462: [gluon] nomenclature for hybrid computation
URL: https://github.com/apache/incubator-mxnet/issues/7462
This is an automated message from the Apache Git Service.
To respond to the message, please log
szha closed pull request #7458: Return to the upstream CUB
URL: https://github.com/apache/incubator-mxnet/pull/7458
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 a foreign pull
szha commented on issue #7462: [gluon] nomenclature for hybrid computation
URL:
https://github.com/apache/incubator-mxnet/issues/7462#issuecomment-344104792
This issue is closed due to lack of activity in the last 90 days. Feel free
to ping me to reopen if this is still an active issue.
szha closed issue #7454: Question about networks for cifar100.
URL: https://github.com/apache/incubator-mxnet/issues/7454
This is an automated message from the Apache Git Service.
To respond to the message, please log on
vatsal2020 opened a new issue #8638: allow_extra parameter in Line 652 in
incubator-mxnet/python/mxnet/module/base_module.py needs to be removed to make
things work
URL: https://github.com/apache/incubator-mxnet/issues/8638
The function that was working perfectly in mxnet 0.11.0 only
piiswrong closed pull request #8543: fix custom op error when using auxiliary
states
URL: https://github.com/apache/incubator-mxnet/pull/8543
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 an automated email from the ASF dual-hosted git repository.
cjolivier01 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 399ac03 [EXPERIMENT] increasing
cjolivier01 closed pull request #8613: [EXPERIMENT] increasing build timeout to
24hrs
URL: https://github.com/apache/incubator-mxnet/pull/8613
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
cjolivier01 commented on issue #8617: Failed to detect tcmalloc or jemalloc for
Fedora 26
URL:
https://github.com/apache/incubator-mxnet/issues/8617#issuecomment-344084740
Of course you may submit a PR.
How do you handle libtcmalloc.so.3 and libtcmalloc.so.4 in the same
directory?
cjolivier01 commented on issue #8617: Failed to detect tcmalloc or jemalloc for
Fedora 26
URL:
https://github.com/apache/incubator-mxnet/issues/8617#issuecomment-344084740
Of course you may submit a CR.
How do you handle libtcmalloc.so.3 and libtcmalloc.so.4 in the same
directory?
ptrendx commented on issue #8636: What differences between ImageRecordIter and
ImageRecordIter_v1?
URL:
https://github.com/apache/incubator-mxnet/issues/8636#issuecomment-344082649
ImageRecordIter_v1 is a much slower, deprecated version of ImageRecordIter.
I will make a PR to
eric-haibin-lin commented on issue #8617: Failed to detect tcmalloc or jemalloc
for Fedora 26
URL:
https://github.com/apache/incubator-mxnet/issues/8617#issuecomment-344081860
@cjolivier01 do you know how to resolve this issue?
eric-haibin-lin commented on issue #8628: [CoreML Converter] Fully connected
layers w/o bias
URL:
https://github.com/apache/incubator-mxnet/issues/8628#issuecomment-344081572
@pracheer Is there any plan to fix it?
This is
eric-haibin-lin commented on issue #8636: What differences between
ImageRecordIter and ImageRecordIter_v1?
URL:
https://github.com/apache/incubator-mxnet/issues/8636#issuecomment-344081266
Could you help clarify and improve the documentation if necessary? @ptrendx
rahul003 commented on issue #5460: very low training accuracies with alexnet or
vgg but good with resnet
URL:
https://github.com/apache/incubator-mxnet/issues/5460#issuecomment-344081096
@anijain2305 @Luo-Liang @xingzhaoo were you able to train alexnet
successfully? If so, do you
This is an automated email from the ASF dual-hosted git repository.
jxie 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 d8d2c39 Continued Work on Advanced
piiswrong closed pull request #8246: Continued Work on Advanced Indexing
URL: https://github.com/apache/incubator-mxnet/pull/8246
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 a
piiswrong commented on issue #8334: Bugfix: Python 3 compatiblity during
optimizer serialization.
URL: https://github.com/apache/incubator-mxnet/pull/8334#issuecomment-344015742
Looks like test is stuck. Please rebase.
This
eric-haibin-lin commented on a change in pull request #8632: a user friendly
way to use g2c in module
URL: https://github.com/apache/incubator-mxnet/pull/8632#discussion_r150620300
##
File path: tests/python/unittest/test_module.py
##
@@ -78,15 +78,15 @@ def
reminisce commented on issue #8246: Continued Work on Advanced Indexing
URL: https://github.com/apache/incubator-mxnet/pull/8246#issuecomment-343993705
@meanmee This is a feature for 1.0 which is coming soon. Thanks for your
interest.
thirdwing commented on issue #8545: Incorrect results from R 3.4.2 in MNIST
URL:
https://github.com/apache/incubator-mxnet/issues/8545#issuecomment-343954000
#8121 has been merged. @ironfrown Can you try the latest code?
thirdwing commented on a change in pull request #8118: [R] various small
changes for efficiency and robustness in R package
URL: https://github.com/apache/incubator-mxnet/pull/8118#discussion_r150567434
##
File path: R-package/R/zzz.R
##
@@ -16,7 +16,10 @@ NULL
.onLoad
thirdwing commented on a change in pull request #8118: [R] various small
changes for efficiency and robustness in R package
URL: https://github.com/apache/incubator-mxnet/pull/8118#discussion_r150569936
##
File path: R-package/R/mx.io.bucket.iter.R
##
@@ -36,54 +36,50 @@
thirdwing commented on a change in pull request #8118: [R] various small
changes for efficiency and robustness in R package
URL: https://github.com/apache/incubator-mxnet/pull/8118#discussion_r150568073
##
File path: R-package/R/viz.graph.R
##
@@ -133,22 +135,23 @@
thirdwing commented on a change in pull request #8118: [R] various small
changes for efficiency and robustness in R package
URL: https://github.com/apache/incubator-mxnet/pull/8118#discussion_r150569510
##
File path: R-package/R/model.R
##
@@ -589,27 +587,20 @@
gooaah commented on issue #8577: support for lapack functions with mkl
URL: https://github.com/apache/incubator-mxnet/pull/8577#issuecomment-343923465
@asmushetzel thx, the matrix has some tiny negtive eigenvalues. After adding
a small diagonal matrix, it works.
asmushetzel commented on issue #8577: support for lapack functions with mkl
URL: https://github.com/apache/incubator-mxnet/pull/8577#issuecomment-343915058
That means that the call failed internally inside the MKL-function. Did you
check whether your input is valid (i.e. positive definite
shipeng-uestc commented on issue #1905: Random scaling in image augmentation
URL:
https://github.com/apache/incubator-mxnet/issues/1905#issuecomment-343911047
Hi, how do you make your own augmenter.Can you share it with me ?
shipeng-uestc opened a new issue #8636: What differences between
ImageRecordIter and ImageRecordIter_v1?
URL: https://github.com/apache/incubator-mxnet/issues/8636
I read the ImageRecordIter_v1 API ,it mentions "Read images batches from
RecordIO files with a rich of data augmentation
szha commented on issue #7451: Not appending to end of linked list correctly?
URL:
https://github.com/apache/incubator-mxnet/issues/7451#issuecomment-343903974
This issue is closed due to lack of activity in the last 90 days. Feel free
to ping me to reopen if this is still an active
szha commented on issue #7361: training speed of batch-norm is less than
batch-norm-v1
URL:
https://github.com/apache/incubator-mxnet/issues/7361#issuecomment-343903984
This issue is closed due to lack of activity in the last 90 days. Feel free
to ping me to reopen if this is still an
szha commented on issue #7406: inconsistent accuracy: ImageRecordIter vs
ImageIter
URL:
https://github.com/apache/incubator-mxnet/issues/7406#issuecomment-343903988
This issue is closed due to lack of activity in the last 90 days. Feel free
to ping me to reopen if this is still an active
szha commented on issue #7453: Pb in import of mxnet
URL:
https://github.com/apache/incubator-mxnet/issues/7453#issuecomment-343903978
This issue is closed due to lack of activity in the last 90 days. Feel free
to ping me to reopen if this is still an active issue. Thanks!
Also, do
szha commented on issue #7448: out of memory when training imagenet with .rec
file.
URL:
https://github.com/apache/incubator-mxnet/issues/7448#issuecomment-343903969
This issue is closed due to lack of activity in the last 90 days. Feel free
to ping me to reopen if this is still an
szha closed issue #7361: training speed of batch-norm is less than batch-norm-v1
URL: https://github.com/apache/incubator-mxnet/issues/7361
This is an automated message from the Apache Git Service.
To respond to the message,
szha closed issue #7453: Pb in import of mxnet
URL: https://github.com/apache/incubator-mxnet/issues/7453
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
szha closed issue #7448: out of memory when training imagenet with .rec file.
URL: https://github.com/apache/incubator-mxnet/issues/7448
This is an automated message from the Apache Git Service.
To respond to the message,
szha closed issue #7451: Not appending to end of linked list correctly?
URL: https://github.com/apache/incubator-mxnet/issues/7451
This is an automated message from the Apache Git Service.
To respond to the message, please
szha closed issue #7406: inconsistent accuracy: ImageRecordIter vs ImageIter
URL: https://github.com/apache/incubator-mxnet/issues/7406
This is an automated message from the Apache Git Service.
To respond to the message,
gooaah commented on issue #8577: support for lapack functions with mkl
URL: https://github.com/apache/incubator-mxnet/pull/8577#issuecomment-343877450
I have compiled mxnet in lapack_mkl version, and I got an error"spotrf
failed in lapack on cpu" when using linalg_potrf. But other linalg
bfgray3 commented on issue #8118: [R] various small changes for efficiency and
robustness in R package
URL: https://github.com/apache/incubator-mxnet/pull/8118#issuecomment-343900238
@thirdwing I have rectified the problems and the tests are green. Please let
me know your thoughts.
ZiyueHuang commented on issue #8631: linear classification bug fix: kv may be
None
URL: https://github.com/apache/incubator-mxnet/pull/8631#issuecomment-343890208
Thanks anyway :)
This is an automated message from the Apache
kobenaxie opened a new issue #8634: How to use label smoothing and set
'ignore_label' with gluon like mx.symSoftmaxOutput ?
URL: https://github.com/apache/incubator-mxnet/issues/8634
In mx.sym.SoftmaxOutput(), we can set `ignore_label` and `smooth_alpha` ,
how to use these in gluon ?
gooaah commented on issue #8577: support for lapack functions with mkl
URL: https://github.com/apache/incubator-mxnet/pull/8577#issuecomment-343877450
I have compiled mxnet in lapack_mkl version, and I got an error"spotrf
failed in lapack on cpu" when using linalg_potrf. But other linalg
yajiedesign opened a new pull request #8633: ]WIP]Image Augmenter
URL: https://github.com/apache/incubator-mxnet/pull/8633
## Description ##
(Brief description on what this PR is about)
## Checklist ##
### Essentials ###
- [ ] Passed code style checking (`make lint`)
- [
meanmee commented on issue #8246: Continued Work on Advanced Indexing
URL: https://github.com/apache/incubator-mxnet/pull/8246#issuecomment-343854746
could this feature be used in version 0.12?
This is an automated message
ZiyueHuang opened a new pull request #8632: a user friendly way to use g2c in
module
URL: https://github.com/apache/incubator-mxnet/pull/8632
## Description ##
use the interface stated in
https://github.com/apache/incubator-mxnet/pull/8539
@eric-haibin-lin
## Checklist ##
chinakook commented on issue #8629: c_predict_api BUG in new mxnet version.
URL:
https://github.com/apache/incubator-mxnet/issues/8629#issuecomment-343854055
It's the BUG with MKL2018 when set USE_MKL_IF_AVALIABLE is on.
chinakook closed issue #8629: c_predict_api BUG in new mxnet version.
URL: https://github.com/apache/incubator-mxnet/issues/8629
This is an automated message from the Apache Git Service.
To respond to the message, please log
formath commented on issue #8631: linear classification bug fix: kv may be None
URL: https://github.com/apache/incubator-mxnet/pull/8631#issuecomment-343849343
being fixed in another pr
This is an automated message from the
formath closed pull request #8631: linear classification bug fix: kv may be None
URL: https://github.com/apache/incubator-mxnet/pull/8631
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
formath opened a new pull request #8631: linear classification bug fix: kv may
be None
URL: https://github.com/apache/incubator-mxnet/pull/8631
@eric-haibin-lin @piiswrong
This is an automated message from the Apache Git
76 matches
Mail list logo