kohillyang opened a new issue #8614: Does mxnet.symbol.SoftmaxOutput support
matrix label?
URL: https://github.com/apache/incubator-mxnet/issues/8614
Assume I have several Matrix classes, they are mutually exclusive, which
means that the sum of them must be a matrix whose all items is 1.
cjolivier01 commented on issue #8599: Pull more optimize and simplification
changes from tuner branch
URL: https://github.com/apache/incubator-mxnet/pull/8599#issuecomment-343376929
I am doing rebase, no idea why it is showing baaafc5
stoneyang commented on issue #1217: Resume training failed
URL:
https://github.com/apache/incubator-mxnet/issues/1217#issuecomment-343637129
Thanks for sharing your experience! @fengjunlv
This is an automated message from
ZiyueHuang commented on issue #8259: check_format of sparse ndrray
URL: https://github.com/apache/incubator-mxnet/pull/8259#issuecomment-343634820
The result err number will be returned to `SyncCheckFormat` in ndarray.cc
and do the check there. `CheckFormatImpl` can be used in other OPs in
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 35ceea7 use first class cuda
cjolivier01 closed pull request #8572: use first class cuda with cmake 3.9 and
cuda9.0 support
URL: https://github.com/apache/incubator-mxnet/pull/8572
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
ZiyueHuang commented on issue #8259: check_format of sparse ndrray
URL: https://github.com/apache/incubator-mxnet/pull/8259#issuecomment-343634820
The result err number will be returned to SyncCheckFormat in ndarray.cc and
do the check there. CheckFormatImpl can be used in other OPs in
ZiyueHuang commented on a change in pull request #8259: check_format of sparse
ndrray
URL: https://github.com/apache/incubator-mxnet/pull/8259#discussion_r150372061
##
File path: src/common/utils.h
##
@@ -43,9 +43,177 @@
#include
#include
+#include
ZiyueHuang commented on a change in pull request #8611: optimization for
dot(csr.T, dense) = rsp
URL: https://github.com/apache/incubator-mxnet/pull/8611#discussion_r150371782
##
File path: src/operator/tensor/dot-inl.h
##
@@ -598,40 +614,52 @@ inline void
thirdwing closed pull request #8121: [R] Initializer fix and adjustments to RNN
API
URL: https://github.com/apache/incubator-mxnet/pull/8121
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.
qkou 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 5c8981a [R] Initializer fix and
yajiedesign commented on issue #8572: use first class cuda with cmake 3.9 and
cuda9.0 support
URL: https://github.com/apache/incubator-mxnet/pull/8572#issuecomment-343631038
yes
This is an automated message from the Apache
rahul003 commented on issue #8578: Add development instructions and update
installation instructions
URL: https://github.com/apache/incubator-mxnet/pull/8578#issuecomment-343625788
How much better is it to use cmake+ninja vs make on ubuntu?
In terms of speed
sxjscience 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-343627954
@asmushetzel This makes sense to me.
This is an
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-343625693
I suppose this is still about this one:
rahul003 commented on issue #8578: Add development instructions and update
installation instructions
URL: https://github.com/apache/incubator-mxnet/pull/8578#issuecomment-343625788
How much better is it to use cmake+ninja vs make on ubuntu?
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-343625693
I suppose this is still about this one:
mbaijal opened a new pull request #8613: [EXPERIMENT] increasing build timeout
to 24hrs
URL: https://github.com/apache/incubator-mxnet/pull/8613
## Description ##
All builds are timing out in the build queue since the number of ubuntu
slaves is less.
Ideally this timeout is meant
zhreshold commented on a change in pull request #8603: Contrib operators for
object-detection bounding box related stuffs
URL: https://github.com/apache/incubator-mxnet/pull/8603#discussion_r150365483
##
File path: src/operator/contrib/bounding_box.cc
##
@@ -0,0 +1,194 @@
szha commented on issue #7431: AttributeError: 'module' object has no attribute
'stack'
URL:
https://github.com/apache/incubator-mxnet/issues/7431#issuecomment-343621970
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 #4696: run errors
URL:
https://github.com/apache/incubator-mxnet/issues/4696#issuecomment-343621976
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 please check
szha closed issue #4696: run errors
URL: https://github.com/apache/incubator-mxnet/issues/4696
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above
szha commented on issue #7384: Parameter 'dtype' in the function
'ImageRecordIter' doesn't work
URL:
https://github.com/apache/incubator-mxnet/issues/7384#issuecomment-343621973
This issue is closed due to lack of activity in the last 90 days. Feel free
to ping me to reopen if this is
szha closed issue #7431: AttributeError: 'module' object has no attribute
'stack'
URL: https://github.com/apache/incubator-mxnet/issues/7431
This is an automated message from the Apache Git Service.
To respond to the
szha closed issue #7384: Parameter 'dtype' in the function 'ImageRecordIter'
doesn't work
URL: https://github.com/apache/incubator-mxnet/issues/7384
This is an automated message from the Apache Git Service.
To respond to
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 582e14c Fix for "name epochs is not
piiswrong closed pull request #8399: Fix for "name epochs is not defined"
URL: https://github.com/apache/incubator-mxnet/pull/8399
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 #8610: How to resize the image to same shape in
ImageRecordIter?
URL:
https://github.com/apache/incubator-mxnet/issues/8610#issuecomment-343618547
imagerecorditer has a resize parameter. Use it to resize the image to a
bigger size then crop size first
sxjscience commented on a change in pull request #8603: Contrib operators for
object-detection bounding box related stuffs
URL: https://github.com/apache/incubator-mxnet/pull/8603#discussion_r150362562
##
File path: src/operator/contrib/bounding_box-inl.h
##
@@ -0,0
zhreshold commented on a change in pull request #8603: Contrib operators for
object-detection bounding box related stuffs
URL: https://github.com/apache/incubator-mxnet/pull/8603#discussion_r150362100
##
File path: src/operator/contrib/bounding_box-inl.h
##
@@ -0,0 +1,730
piiswrong commented on issue #8259: check_format of sparse ndrray
URL: https://github.com/apache/incubator-mxnet/pull/8259#issuecomment-343615517
How is the check result returned? Is the behavior consistent with scipy?
This
piiswrong commented on a change in pull request #8259: check_format of sparse
ndrray
URL: https://github.com/apache/incubator-mxnet/pull/8259#discussion_r150360872
##
File path: src/common/utils.h
##
@@ -43,9 +43,177 @@
#include
#include
+#include
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 00ddcf4 support for lapack functions
piiswrong commented on a change in pull request #8577: support for lapack
functions with mkl
URL: https://github.com/apache/incubator-mxnet/pull/8577#discussion_r150360534
##
File path: src/operator/c_lapack_api.h
##
@@ -160,7 +162,75 @@ inline void flip(int
piiswrong closed pull request #8577: support for lapack functions with mkl
URL: https://github.com/apache/incubator-mxnet/pull/8577
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
jeremiedb commented on issue #8545: Incorrect results from R 3.4.2 in MNIST
URL:
https://github.com/apache/incubator-mxnet/issues/8545#issuecomment-343611371
Yes #8121 will fix this problem.
This is an automated message from
jeremiedb commented on a change in pull request #8121: [R] Initializer fix and
adjustments to RNN API
URL: https://github.com/apache/incubator-mxnet/pull/8121#discussion_r150357674
##
File path: R-package/R/initializer.R
##
@@ -4,11 +4,11 @@
#' @param shape the shape of
sxjscience commented on a change in pull request #8603: Contrib operators for
object-detection bounding box related stuffs
URL: https://github.com/apache/incubator-mxnet/pull/8603#discussion_r150357652
##
File path: src/operator/contrib/bounding_box-inl.h
##
@@ -0,0
eric-haibin-lin commented on issue #8611: optimization for dot(csr.T, dense) =
rsp
URL: https://github.com/apache/incubator-mxnet/pull/8611#issuecomment-343603981
Yes, the original dot(csr.T, dense) performance is not good. The same for
scipy - it's not always faster than the dense dot
eric-haibin-lin commented on a change in pull request #8611: optimization for
dot(csr.T, dense) = rsp
URL: https://github.com/apache/incubator-mxnet/pull/8611#discussion_r150351688
##
File path: src/operator/tensor/dot-inl.h
##
@@ -598,40 +614,52 @@ inline void
eric-haibin-lin commented on a change in pull request #8611: optimization for
dot(csr.T, dense) = rsp
URL: https://github.com/apache/incubator-mxnet/pull/8611#discussion_r150351336
##
File path: src/operator/tensor/dot-inl.h
##
@@ -573,6 +577,18 @@ inline void
thirdwing commented on issue #8545: Incorrect results from R 3.4.2 in MNIST
URL:
https://github.com/apache/incubator-mxnet/issues/8545#issuecomment-343598024
https://github.com/apache/incubator-mxnet/pull/8121 will also fix this,
right? @jeremiedb
thirdwing commented on issue #8575: mxnet multicore on LInux in R
URL:
https://github.com/apache/incubator-mxnet/issues/8575#issuecomment-343597753
Please provide a reproducible example.
This is an automated message from the
thirdwing commented on issue #8121: [R] Initializer fix and adjustments to RNN
API
URL: https://github.com/apache/incubator-mxnet/pull/8121#issuecomment-343597558
@jeremiedb Sorry for the long delay. Just a minor question.
thirdwing commented on a change in pull request #8121: [R] Initializer fix and
adjustments to RNN API
URL: https://github.com/apache/incubator-mxnet/pull/8121#discussion_r150346131
##
File path: R-package/R/initializer.R
##
@@ -4,11 +4,11 @@
#' @param shape the shape of
piiswrong opened a new pull request #8612: generalize array dataset
URL: https://github.com/apache/incubator-mxnet/pull/8612
## Description ##
(Brief description on what this PR is about)
## Checklist ##
### Essentials ###
- [ ] Passed code style checking (`make lint`)
-
cjolivier01 commented on a change in pull request #8302: Refactor operators
URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r150338088
##
File path: src/operator/nn/batch_norm.cc
##
@@ -313,45 +314,76 @@ void BatchNormOp
piiswrong closed pull request #8539: expose group2ctx to module
URL: https://github.com/apache/incubator-mxnet/pull/8539
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
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 c9bde1b expose group2ctx to module
piiswrong commented on a change in pull request #8566: optimize broadcast
URL: https://github.com/apache/incubator-mxnet/pull/8566#discussion_r150332277
##
File path: src/operator/mxnet_op.h
##
@@ -345,6 +394,13 @@ __global__ void mxnet_generic_kernel(int N, Args... args)
zheng-da commented on a change in pull request #8302: Refactor operators
URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r150330094
##
File path: src/operator/nn/fully_connected-inl.h
##
@@ -176,124 +162,80 @@ class FullyConnectedOp : public Operator {
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 97e7015 Update PR & Issue Template
piiswrong closed pull request #8555: Update PR & Issue Template
URL: https://github.com/apache/incubator-mxnet/pull/8555
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
reminisce commented on issue #8246: Continued Work on Advanced Indexing
URL: https://github.com/apache/incubator-mxnet/pull/8246#issuecomment-343549343
@piiswrong @eric-haibin-lin @szha I added support for `__setitem__` and
simplified basic and advanced indexing dispatching logic. Please
zhreshold commented on issue #8603: Contrib operators for object-detection
bounding box related stuffs
URL: https://github.com/apache/incubator-mxnet/pull/8603#issuecomment-343542184
@sbodenstein Functionally, yes, but not identical api. This one provide more
options.
Do you have a
fungtion opened a new issue #4159: How to bind 3 inputs using
mxnet.io.NDArrayIter?
URL: https://github.com/apache/incubator-mxnet/issues/4159
For bugs or installation issues, please provide the following information.
The more information you provide, the more likely people will be able
ptrendx commented on issue #8600: How many new data points does data
augmentation with mx.io.ImageRecordIter produce?
URL:
https://github.com/apache/incubator-mxnet/issues/8600#issuecomment-343538346
Augmentations do not increase the size of your dataset explicitly. How it
works is that
ZiyueHuang opened a new pull request #8611: optimization for dot(csr.T, dense)
= rsp
URL: https://github.com/apache/incubator-mxnet/pull/8611
## Description ##
Use prefix sum to compute `nnr` in order to allocate the row_sparse output.
Currently `dot(csr.T, dense) = rsp` will
aidan-plenert-macdonald commented on issue #8591: How do I make a siamese
network with pretrained models (esp. keeping the weights the same?)
URL:
https://github.com/apache/incubator-mxnet/issues/8591#issuecomment-343536911
If you want to share weights across the network, why not just use
kaylimekay commented on issue #7698: Second order gradient and Subgraph
execution
URL: https://github.com/apache/incubator-mxnet/pull/7698#issuecomment-343535842
Is there a list of operators that currently support the higher order
gradients from grad()?
cjolivier01 commented on issue #8572: use first class cuda with cmake 3.9 and
cuda9.0 support
URL: https://github.com/apache/incubator-mxnet/pull/8572#issuecomment-343533256
Is this ready to merge?
This is an automated
jeremiedb commented on issue #8121: [R] Initializer fix and adjustments to RNN
API
URL: https://github.com/apache/incubator-mxnet/pull/8121#issuecomment-343527120
@thirdwing Anytime remaining to fix? I thin this PR is important as no
training process is currently broken by the initializer
anjishnu commented on issue #4159: How to bind 3 inputs using
mxnet.io.NDArrayIter?
URL:
https://github.com/apache/incubator-mxnet/issues/4159#issuecomment-343505909
I am also using a network architecture like OP which takes in 2 streams as
input.
anjishnu commented on issue #4159: How to bind 3 inputs using
mxnet.io.NDArrayIter?
URL:
https://github.com/apache/incubator-mxnet/issues/4159#issuecomment-343505315
I just got this issue -
> TypeError: Invalid type '' for x_2, should be
NDArray, numpy.ndarray or h5py.Dataset
mstokes42 commented on issue #8601: Why is the accuracy overestimated?
URL:
https://github.com/apache/incubator-mxnet/issues/8601#issuecomment-343499688
I figured out the problem was with my batch_size on the val_iter. The sample
size was not a multiple of the batch size, so some samples
lilhope opened a new issue #8610: How to resize the image to same shape in
ImageRecordIter?
URL: https://github.com/apache/incubator-mxnet/issues/8610
Hi,
I have a set of images with various shapes and I write it to `.rec`
file,and using `mx.io.ImageRecordIter` to load it. But I
mwunderlich opened a new issue #8609: git clone command wrong for Scala setup
URL: https://github.com/apache/incubator-mxnet/issues/8609
The git clone command provided for the scala setup on OSX (on page
https://mxnet.incubator.apache.org/install/osx_setup.html) seems to be wrong.
When I
sbodenstein commented on issue #8582: Yolo2 operator
URL: https://github.com/apache/incubator-mxnet/pull/8582#issuecomment-343462534
One more comment on this PR: is there anyway of verifying that the gradient
is correctly calculated?
szha closed issue #7420: simple_bind failed error
URL: https://github.com/apache/incubator-mxnet/issues/7420
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use
szha closed issue #7422: Data provided by data_shapes don't match names
specified by data_names
URL: https://github.com/apache/incubator-mxnet/issues/7422
This is an automated message from the Apache Git Service.
To respond
szha closed issue #7414: why rnn train speed is not stable,sometims very slow?
URL: https://github.com/apache/incubator-mxnet/issues/7414
This is an automated message from the Apache Git Service.
To respond to the message,
szha commented on issue #7420: simple_bind failed error
URL:
https://github.com/apache/incubator-mxnet/issues/7420#issuecomment-343460770
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 #7414: why rnn train speed is not stable,sometims very
slow?
URL:
https://github.com/apache/incubator-mxnet/issues/7414#issuecomment-343460773
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 #7422: Data provided by data_shapes don't match names
specified by data_names
URL:
https://github.com/apache/incubator-mxnet/issues/7422#issuecomment-343460775
This issue is closed due to lack of activity in the last 90 days. Feel free
to ping me to reopen if this
wkcn commented on issue #8569: The conflict between MXNet and OpenCV
URL:
https://github.com/apache/incubator-mxnet/issues/8569#issuecomment-343458075
@KellenSunderland Thank you!
This is an automated message from the Apache
KellenSunderland commented on issue #8569: The conflict between MXNet and OpenCV
URL:
https://github.com/apache/incubator-mxnet/issues/8569#issuecomment-343442891
@wkcn Thanks, for the info. I think your description on the relation
between the USE_GPERFTOOLS flag and the crash is clear.
tz-hmc commented on issue #8591: How do I make a siamese network with
pretrained models (esp. keeping the weights the same?)
URL:
https://github.com/apache/incubator-mxnet/issues/8591#issuecomment-343438692
Wow. Thank you so much. Alright, this gives me a lot to think about.
I'm really
asmushetzel commented on a change in pull request #8577: support for lapack
functions with mkl
URL: https://github.com/apache/incubator-mxnet/pull/8577#discussion_r150193519
##
File path: src/operator/c_lapack_api.h
##
@@ -160,7 +162,75 @@ inline void flip
KellenSunderland commented on issue #8573: /usr/bin/ld: cannot find -lgtest ?
URL:
https://github.com/apache/incubator-mxnet/issues/8573#issuecomment-343424577
Hello @eLvErDe , looks like you're missing the gtest library on your system.
The good news is it's quite easy to install.
wkcn commented on issue #8569: The conflict between MXNet and OpenCV
URL:
https://github.com/apache/incubator-mxnet/issues/8569#issuecomment-343424178
@KellenSunderland The crash will occur too if importing cv2 after mxnet with
`gperftools`.
I set `USE_GPERFTOOLS = 0` and `USE_JEMALLOC
KellenSunderland commented on issue #8569: The conflict between MXNet and OpenCV
URL:
https://github.com/apache/incubator-mxnet/issues/8569#issuecomment-343421365
Out of curiosity, does the crash occur if you import cv2 after mxnet?
sbodenstein commented on a change in pull request #8603: Contrib operators for
object-detection bounding box related stuffs
URL: https://github.com/apache/incubator-mxnet/pull/8603#discussion_r150181423
##
File path: src/operator/contrib/bounding_box.cc
##
@@ -0,0 +1,194
sbodenstein commented on issue #8603: Contrib operators for object-detection
bounding box related stuffs
URL: https://github.com/apache/incubator-mxnet/pull/8603#issuecomment-343414068
Is the NMS op consistent with TensorFlows
mseeger 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-343414021
@asmushetzel Hi Asmus, some people continue to have problems building on
some CentOS setup that I cannot work on.
mseeger 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-343413652
Hello,
I'd like to help, but I just cannot reproduce the error (given it is still
the same as the one described
liumilan closed issue #8574: how to print debug log in gpu mode?
URL: https://github.com/apache/incubator-mxnet/issues/8574
This is an automated message from the Apache Git Service.
To respond to the message, please log on
86 matches
Mail list logo