Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Anirudh]

Thank you Anirudh for working on the fix!

Update on the current status of RC1 release:
Need to merge the following PRs: #10677
,#10661
,#10676
,#10666

Then we can proceed with running the nightlies and RAT checks before we cut
the release candidate rc1.


On Mon, Apr 23, 2018 at 10:11 PM, Anirudh Acharya 
wrote:

> Yes ideally, the fix should be -
> 1. take the padding tuple from onnx models and pass the appropriately sized
> tuple to mxnet. And if the padding in the onnx model is asymmetric we
> should throw an exception, saying it is not yet supported.
> 2. And also explore supporting asymmetric padding in mxnet.
>
> The model used in the example tutorial that is failing is actually using
> symmetric padding. And the code in onnx-mxnet can be fixed to handle this.
> Will get a PR out to pass the padding tuple of the right length and raise
> an exception when there is asymmetric padding.
>
>
>
> Regards,
> Anirudh
>
>
> On Mon, Apr 23, 2018 at 8:56 PM, Da Zheng  wrote:
>
> > Here is a temporary fix, so MKLDNN conv follows the behavior of MXNet
> > conv (always uses the first two elements in the tuple as padding).
> > https://github.com/apache/incubator-mxnet/pull/10666
> >
> > I also created an issue regarding this problem so that we'll fix it
> > properly in the future.
> > https://github.com/apache/incubator-mxnet/issues/10663
> >
> > Best,
> > Da
> >
> > On Mon, Apr 23, 2018 at 8:00 PM, Anirudh  wrote:
> > > Hi,
> > >
> > > Currently mkldnn support adds the check for tuple length for padding
> and
> > > raises the exception. Even though this is the correct approach, this is
> > not
> > > consistent with
> > > MXNet behavior without MKLDNN enabled.
> > >
> > > I agree with Da and Haibin that the right approach in the longer term
> > would
> > > be to:
> > > 1. raise an exception for MXNet conv operator when tuple size is 4.
> > > 2. Fix onnx converter to convert paddings correctly
> > >
> > > This would prevent silents failures with wrong results and also have a
> > > consistent behavior for MXNet with or without MKLDNN enabled.
> > >
> > > Having said that, in the shorter term I think it is okay to remove the
> > > check in MKLDNN, if we can notify the user somehow(maybe a warning) in
> > > ONNX-converter about (left, right, top, bottom) being interpreted as
> > > (height, width) for Conv operator.
> > >
> > > Anirudh
> > >
> > >
> > >
> > >
> > > On Mon, Apr 23, 2018 at 6:19 PM, Haibin Lin 
> > > wrote:
> > >
> > >> Hi Da,
> > >>
> > >> After looking at your detailed description in github issue
> > >> https://github.com/apache/incubator-mxnet/issues/10663, I would argue
> > that
> > >> the fix should go to mxnet-onnx instead of mxnet-mkldnn.
> > >>
> > >> In onnx padding params are in the form of (left, right, top, bottom),
> > which
> > >> supports asymmetric padding. In MXNet only symmetric padding is
> > supported
> > >> in the form of (height, width) tuple, which doesn't support asymmetric
> > >> padding.
> > >>
> > >> If you remove the check in mkldnn, that means onnx-mxnet user can pass
> > in
> > >> an asymmetric padding of (left, right, top, bottom) but mxnet would
> only
> > >> look at the first 2 numbers in the tuple and compute the wrong result
> > >> silently.
> > >>
> > >> One way to fix it is to support (left, right, top, bottom) in mxnet
> > padding
> > >> op (both with mkl-dnn and without mkl-dnn), which will take some time.
> > >> Another way is that mxnet-onnx checks the (left, right, top, bottom)
> > tuple,
> > >> and pass the arguments in the form of (height, width) to mxnet ops if
> > >> symmetric, or throw an exception for asymmetric paddings.
> > >>
> > >> I'm not an expert in onnx but I'm curious if this can be fixed in
> > >> mxnet-onnx quickly. Removing the correct check in mkldnn doesn't sound
> > >> reasonable...
> > >>
> > >> Best,
> > >> Haibin
> > >>
> > >> On Mon, Apr 23, 2018 at 5:12 PM, Zheng, Da  wrote:
> > >>
> > >> > I think I have found the root of the problem.
> > >> >
> > >> > The tutorial loads a model from onnx, which uses padding (left,
> right,
> > >> > top, bottom). But mxnet uses padding (height, width). Currently,
> when
> > an
> > >> > ONNX model is loaded to MXNet, the padding is converted correctly.
> > MXNet
> > >> > conv doesn't check the number of elements in the padding and ignores
> > the
> > >> > problem. However, mxnet-mkldnn checks it and fails.
> > >> >
> > >> > The correct way of fixing this issue is to check the number of
> > elements
> > >> in
> > >> > the padding tuple in mxnet conv. If the tuple size mismatches, it
> > should
> > >> > fail. When an ONNX model is loaded to MXNet, the padding should be
> > >> > 

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Anirudh Acharya]

Yes ideally, the fix should be -
1. take the padding tuple from onnx models and pass the appropriately sized
tuple to mxnet. And if the padding in the onnx model is asymmetric we
should throw an exception, saying it is not yet supported.
2. And also explore supporting asymmetric padding in mxnet.

The model used in the example tutorial that is failing is actually using
symmetric padding. And the code in onnx-mxnet can be fixed to handle this.
Will get a PR out to pass the padding tuple of the right length and raise
an exception when there is asymmetric padding.



Regards,
Anirudh


On Mon, Apr 23, 2018 at 8:56 PM, Da Zheng  wrote:

> Here is a temporary fix, so MKLDNN conv follows the behavior of MXNet
> conv (always uses the first two elements in the tuple as padding).
> https://github.com/apache/incubator-mxnet/pull/10666
>
> I also created an issue regarding this problem so that we'll fix it
> properly in the future.
> https://github.com/apache/incubator-mxnet/issues/10663
>
> Best,
> Da
>
> On Mon, Apr 23, 2018 at 8:00 PM, Anirudh  wrote:
> > Hi,
> >
> > Currently mkldnn support adds the check for tuple length for padding and
> > raises the exception. Even though this is the correct approach, this is
> not
> > consistent with
> > MXNet behavior without MKLDNN enabled.
> >
> > I agree with Da and Haibin that the right approach in the longer term
> would
> > be to:
> > 1. raise an exception for MXNet conv operator when tuple size is 4.
> > 2. Fix onnx converter to convert paddings correctly
> >
> > This would prevent silents failures with wrong results and also have a
> > consistent behavior for MXNet with or without MKLDNN enabled.
> >
> > Having said that, in the shorter term I think it is okay to remove the
> > check in MKLDNN, if we can notify the user somehow(maybe a warning) in
> > ONNX-converter about (left, right, top, bottom) being interpreted as
> > (height, width) for Conv operator.
> >
> > Anirudh
> >
> >
> >
> >
> > On Mon, Apr 23, 2018 at 6:19 PM, Haibin Lin 
> > wrote:
> >
> >> Hi Da,
> >>
> >> After looking at your detailed description in github issue
> >> https://github.com/apache/incubator-mxnet/issues/10663, I would argue
> that
> >> the fix should go to mxnet-onnx instead of mxnet-mkldnn.
> >>
> >> In onnx padding params are in the form of (left, right, top, bottom),
> which
> >> supports asymmetric padding. In MXNet only symmetric padding is
> supported
> >> in the form of (height, width) tuple, which doesn't support asymmetric
> >> padding.
> >>
> >> If you remove the check in mkldnn, that means onnx-mxnet user can pass
> in
> >> an asymmetric padding of (left, right, top, bottom) but mxnet would only
> >> look at the first 2 numbers in the tuple and compute the wrong result
> >> silently.
> >>
> >> One way to fix it is to support (left, right, top, bottom) in mxnet
> padding
> >> op (both with mkl-dnn and without mkl-dnn), which will take some time.
> >> Another way is that mxnet-onnx checks the (left, right, top, bottom)
> tuple,
> >> and pass the arguments in the form of (height, width) to mxnet ops if
> >> symmetric, or throw an exception for asymmetric paddings.
> >>
> >> I'm not an expert in onnx but I'm curious if this can be fixed in
> >> mxnet-onnx quickly. Removing the correct check in mkldnn doesn't sound
> >> reasonable...
> >>
> >> Best,
> >> Haibin
> >>
> >> On Mon, Apr 23, 2018 at 5:12 PM, Zheng, Da  wrote:
> >>
> >> > I think I have found the root of the problem.
> >> >
> >> > The tutorial loads a model from onnx, which uses padding (left, right,
> >> > top, bottom). But mxnet uses padding (height, width). Currently, when
> an
> >> > ONNX model is loaded to MXNet, the padding is converted correctly.
> MXNet
> >> > conv doesn't check the number of elements in the padding and ignores
> the
> >> > problem. However, mxnet-mkldnn checks it and fails.
> >> >
> >> > The correct way of fixing this issue is to check the number of
> elements
> >> in
> >> > the padding tuple in mxnet conv. If the tuple size mismatches, it
> should
> >> > fail. When an ONNX model is loaded to MXNet, the padding should be
> >> > converted correctly.
> >> >
> >> > For the time being, I'll just fix MKLDNN so it doesn't check the tuple
> >> > length of padding.
> >> >
> >> > Best,
> >> > Da
> >> >
> >> > On 4/23/18, 2:58 PM, "Zheng, Da"  wrote:
> >> >
> >> > I can reproduce the bug now. I'm working on a fix for the bug.
> >> >
> >> > Currently, there are a few more bug fixes for MKLDNN.
> >> > https://github.com/apache/incubator-mxnet/pull/10651
> >> > https://github.com/apache/incubator-mxnet/pull/10624
> >> > https://github.com/apache/incubator-mxnet/pull/10619/files
> >> > https://github.com/apache/incubator-mxnet/pull/10616
> >> > https://github.com/apache/incubator-mxnet/pull/10591/files
> >> >
> >> > They are ready for review.
> >> >
> >> > I just 

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Da Zheng]

Here is a temporary fix, so MKLDNN conv follows the behavior of MXNet
conv (always uses the first two elements in the tuple as padding).
https://github.com/apache/incubator-mxnet/pull/10666

I also created an issue regarding this problem so that we'll fix it
properly in the future.
https://github.com/apache/incubator-mxnet/issues/10663

Best,
Da

On Mon, Apr 23, 2018 at 8:00 PM, Anirudh  wrote:
> Hi,
>
> Currently mkldnn support adds the check for tuple length for padding and
> raises the exception. Even though this is the correct approach, this is not
> consistent with
> MXNet behavior without MKLDNN enabled.
>
> I agree with Da and Haibin that the right approach in the longer term would
> be to:
> 1. raise an exception for MXNet conv operator when tuple size is 4.
> 2. Fix onnx converter to convert paddings correctly
>
> This would prevent silents failures with wrong results and also have a
> consistent behavior for MXNet with or without MKLDNN enabled.
>
> Having said that, in the shorter term I think it is okay to remove the
> check in MKLDNN, if we can notify the user somehow(maybe a warning) in
> ONNX-converter about (left, right, top, bottom) being interpreted as
> (height, width) for Conv operator.
>
> Anirudh
>
>
>
>
> On Mon, Apr 23, 2018 at 6:19 PM, Haibin Lin 
> wrote:
>
>> Hi Da,
>>
>> After looking at your detailed description in github issue
>> https://github.com/apache/incubator-mxnet/issues/10663, I would argue that
>> the fix should go to mxnet-onnx instead of mxnet-mkldnn.
>>
>> In onnx padding params are in the form of (left, right, top, bottom), which
>> supports asymmetric padding. In MXNet only symmetric padding is supported
>> in the form of (height, width) tuple, which doesn't support asymmetric
>> padding.
>>
>> If you remove the check in mkldnn, that means onnx-mxnet user can pass in
>> an asymmetric padding of (left, right, top, bottom) but mxnet would only
>> look at the first 2 numbers in the tuple and compute the wrong result
>> silently.
>>
>> One way to fix it is to support (left, right, top, bottom) in mxnet padding
>> op (both with mkl-dnn and without mkl-dnn), which will take some time.
>> Another way is that mxnet-onnx checks the (left, right, top, bottom) tuple,
>> and pass the arguments in the form of (height, width) to mxnet ops if
>> symmetric, or throw an exception for asymmetric paddings.
>>
>> I'm not an expert in onnx but I'm curious if this can be fixed in
>> mxnet-onnx quickly. Removing the correct check in mkldnn doesn't sound
>> reasonable...
>>
>> Best,
>> Haibin
>>
>> On Mon, Apr 23, 2018 at 5:12 PM, Zheng, Da  wrote:
>>
>> > I think I have found the root of the problem.
>> >
>> > The tutorial loads a model from onnx, which uses padding (left, right,
>> > top, bottom). But mxnet uses padding (height, width). Currently, when an
>> > ONNX model is loaded to MXNet, the padding is converted correctly. MXNet
>> > conv doesn't check the number of elements in the padding and ignores the
>> > problem. However, mxnet-mkldnn checks it and fails.
>> >
>> > The correct way of fixing this issue is to check the number of elements
>> in
>> > the padding tuple in mxnet conv. If the tuple size mismatches, it should
>> > fail. When an ONNX model is loaded to MXNet, the padding should be
>> > converted correctly.
>> >
>> > For the time being, I'll just fix MKLDNN so it doesn't check the tuple
>> > length of padding.
>> >
>> > Best,
>> > Da
>> >
>> > On 4/23/18, 2:58 PM, "Zheng, Da"  wrote:
>> >
>> > I can reproduce the bug now. I'm working on a fix for the bug.
>> >
>> > Currently, there are a few more bug fixes for MKLDNN.
>> > https://github.com/apache/incubator-mxnet/pull/10651
>> > https://github.com/apache/incubator-mxnet/pull/10624
>> > https://github.com/apache/incubator-mxnet/pull/10619/files
>> > https://github.com/apache/incubator-mxnet/pull/10616
>> > https://github.com/apache/incubator-mxnet/pull/10591/files
>> >
>> > They are ready for review.
>> >
>> > I just discussed with @Anirudh. Maybe we should say in the release
>> > note that MKLDNN in MXNet is still experimental.
>> > What do you think?
>> >
>> > Best,
>> > Da
>> >
>> > On 4/21/18, 7:59 PM, "Zheng, Da"  wrote:
>> >
>> > It seems I have problems of compiling scala when running "make
>> > docs". Please see the error below.
>> >
>> > Are there any instructions of compiling these scala code? I guess
>> > I might miss some packages.
>> > I tried installing libslf4j-java and didn't help.
>> >
>> > Best,
>> > Da
>> >
>> > Execute "cd /home/ubuntu/apache-mxnet-src-
>> > 1.2.0.rc0-incubating/docs/../scala-package; scaladoc `find . -type f
>> > -name "*.scala" | egrep "\/core|\/infer" | egrep -v "Suite"`; exit 0"
>> > ./examples/src/main/scala/org/apache/mxnetexamples/infer/
>> > 

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Anirudh]

Hi,

Currently mkldnn support adds the check for tuple length for padding and
raises the exception. Even though this is the correct approach, this is not
consistent with
MXNet behavior without MKLDNN enabled.

I agree with Da and Haibin that the right approach in the longer term would
be to:
1. raise an exception for MXNet conv operator when tuple size is 4.
2. Fix onnx converter to convert paddings correctly

This would prevent silents failures with wrong results and also have a
consistent behavior for MXNet with or without MKLDNN enabled.

Having said that, in the shorter term I think it is okay to remove the
check in MKLDNN, if we can notify the user somehow(maybe a warning) in
ONNX-converter about (left, right, top, bottom) being interpreted as
(height, width) for Conv operator.

Anirudh




On Mon, Apr 23, 2018 at 6:19 PM, Haibin Lin 
wrote:

> Hi Da,
>
> After looking at your detailed description in github issue
> https://github.com/apache/incubator-mxnet/issues/10663, I would argue that
> the fix should go to mxnet-onnx instead of mxnet-mkldnn.
>
> In onnx padding params are in the form of (left, right, top, bottom), which
> supports asymmetric padding. In MXNet only symmetric padding is supported
> in the form of (height, width) tuple, which doesn't support asymmetric
> padding.
>
> If you remove the check in mkldnn, that means onnx-mxnet user can pass in
> an asymmetric padding of (left, right, top, bottom) but mxnet would only
> look at the first 2 numbers in the tuple and compute the wrong result
> silently.
>
> One way to fix it is to support (left, right, top, bottom) in mxnet padding
> op (both with mkl-dnn and without mkl-dnn), which will take some time.
> Another way is that mxnet-onnx checks the (left, right, top, bottom) tuple,
> and pass the arguments in the form of (height, width) to mxnet ops if
> symmetric, or throw an exception for asymmetric paddings.
>
> I'm not an expert in onnx but I'm curious if this can be fixed in
> mxnet-onnx quickly. Removing the correct check in mkldnn doesn't sound
> reasonable...
>
> Best,
> Haibin
>
> On Mon, Apr 23, 2018 at 5:12 PM, Zheng, Da  wrote:
>
> > I think I have found the root of the problem.
> >
> > The tutorial loads a model from onnx, which uses padding (left, right,
> > top, bottom). But mxnet uses padding (height, width). Currently, when an
> > ONNX model is loaded to MXNet, the padding is converted correctly. MXNet
> > conv doesn't check the number of elements in the padding and ignores the
> > problem. However, mxnet-mkldnn checks it and fails.
> >
> > The correct way of fixing this issue is to check the number of elements
> in
> > the padding tuple in mxnet conv. If the tuple size mismatches, it should
> > fail. When an ONNX model is loaded to MXNet, the padding should be
> > converted correctly.
> >
> > For the time being, I'll just fix MKLDNN so it doesn't check the tuple
> > length of padding.
> >
> > Best,
> > Da
> >
> > On 4/23/18, 2:58 PM, "Zheng, Da"  wrote:
> >
> > I can reproduce the bug now. I'm working on a fix for the bug.
> >
> > Currently, there are a few more bug fixes for MKLDNN.
> > https://github.com/apache/incubator-mxnet/pull/10651
> > https://github.com/apache/incubator-mxnet/pull/10624
> > https://github.com/apache/incubator-mxnet/pull/10619/files
> > https://github.com/apache/incubator-mxnet/pull/10616
> > https://github.com/apache/incubator-mxnet/pull/10591/files
> >
> > They are ready for review.
> >
> > I just discussed with @Anirudh. Maybe we should say in the release
> > note that MKLDNN in MXNet is still experimental.
> > What do you think?
> >
> > Best,
> > Da
> >
> > On 4/21/18, 7:59 PM, "Zheng, Da"  wrote:
> >
> > It seems I have problems of compiling scala when running "make
> > docs". Please see the error below.
> >
> > Are there any instructions of compiling these scala code? I guess
> > I might miss some packages.
> > I tried installing libslf4j-java and didn't help.
> >
> > Best,
> > Da
> >
> > Execute "cd /home/ubuntu/apache-mxnet-src-
> > 1.2.0.rc0-incubating/docs/../scala-package; scaladoc `find . -type f
> > -name "*.scala" | egrep "\/core|\/infer" | egrep -v "Suite"`; exit 0"
> > ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> > objectdetector/SSDClassifierExample.scala:24: error: object kohsuke is
> > not a member of package org
> > import org.kohsuke.args4j.{CmdLineParser, Option}
> >^
> > ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> > objectdetector/SSDClassifierExample.scala:25: error: object slf4j is not
> > a member of package org
> > import org.slf4j.LoggerFactory
> >^
> > ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> > objectdetector/SSDClassifierExample.scala:41: error: class Option is

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Haibin Lin]

Hi Da,

After looking at your detailed description in github issue
https://github.com/apache/incubator-mxnet/issues/10663, I would argue that
the fix should go to mxnet-onnx instead of mxnet-mkldnn.

In onnx padding params are in the form of (left, right, top, bottom), which
supports asymmetric padding. In MXNet only symmetric padding is supported
in the form of (height, width) tuple, which doesn't support asymmetric
padding.

If you remove the check in mkldnn, that means onnx-mxnet user can pass in
an asymmetric padding of (left, right, top, bottom) but mxnet would only
look at the first 2 numbers in the tuple and compute the wrong result
silently.

One way to fix it is to support (left, right, top, bottom) in mxnet padding
op (both with mkl-dnn and without mkl-dnn), which will take some time.
Another way is that mxnet-onnx checks the (left, right, top, bottom) tuple,
and pass the arguments in the form of (height, width) to mxnet ops if
symmetric, or throw an exception for asymmetric paddings.

I'm not an expert in onnx but I'm curious if this can be fixed in
mxnet-onnx quickly. Removing the correct check in mkldnn doesn't sound
reasonable...

Best,
Haibin

On Mon, Apr 23, 2018 at 5:12 PM, Zheng, Da  wrote:

> I think I have found the root of the problem.
>
> The tutorial loads a model from onnx, which uses padding (left, right,
> top, bottom). But mxnet uses padding (height, width). Currently, when an
> ONNX model is loaded to MXNet, the padding is converted correctly. MXNet
> conv doesn't check the number of elements in the padding and ignores the
> problem. However, mxnet-mkldnn checks it and fails.
>
> The correct way of fixing this issue is to check the number of elements in
> the padding tuple in mxnet conv. If the tuple size mismatches, it should
> fail. When an ONNX model is loaded to MXNet, the padding should be
> converted correctly.
>
> For the time being, I'll just fix MKLDNN so it doesn't check the tuple
> length of padding.
>
> Best,
> Da
>
> On 4/23/18, 2:58 PM, "Zheng, Da"  wrote:
>
> I can reproduce the bug now. I'm working on a fix for the bug.
>
> Currently, there are a few more bug fixes for MKLDNN.
> https://github.com/apache/incubator-mxnet/pull/10651
> https://github.com/apache/incubator-mxnet/pull/10624
> https://github.com/apache/incubator-mxnet/pull/10619/files
> https://github.com/apache/incubator-mxnet/pull/10616
> https://github.com/apache/incubator-mxnet/pull/10591/files
>
> They are ready for review.
>
> I just discussed with @Anirudh. Maybe we should say in the release
> note that MKLDNN in MXNet is still experimental.
> What do you think?
>
> Best,
> Da
>
> On 4/21/18, 7:59 PM, "Zheng, Da"  wrote:
>
> It seems I have problems of compiling scala when running "make
> docs". Please see the error below.
>
> Are there any instructions of compiling these scala code? I guess
> I might miss some packages.
> I tried installing libslf4j-java and didn't help.
>
> Best,
> Da
>
> Execute "cd /home/ubuntu/apache-mxnet-src-
> 1.2.0.rc0-incubating/docs/../scala-package; scaladoc `find . -type f
> -name "*.scala" | egrep "\/core|\/infer" | egrep -v "Suite"`; exit 0"
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> objectdetector/SSDClassifierExample.scala:24: error: object kohsuke is
> not a member of package org
> import org.kohsuke.args4j.{CmdLineParser, Option}
>^
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> objectdetector/SSDClassifierExample.scala:25: error: object slf4j is not
> a member of package org
> import org.slf4j.LoggerFactory
>^
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> objectdetector/SSDClassifierExample.scala:41: error: class Option is
> abstract; cannot be instantiated
>   @Option(name = "--model-path-prefix", usage = "the input model
> directory and prefix of the model")
>^
> warning: no valid targets for annotation on value modelPathPrefix
> - it is discarded unused. You may specify targets with meta-annotations,
> e.g. @( @getter)
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> objectdetector/SSDClassifierExample.scala:43: error: class Option is
> abstract; cannot be instantiated
>   @Option(name = "--input-image", usage = "the input image")
>^
> warning: no valid targets for annotation on value inputImagePath -
> it is discarded unused. You may specify targets with meta-annotations, e.g.
> @( @getter)
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> objectdetector/SSDClassifierExample.scala:45: error: class Option is
> abstract; cannot be instantiated
>   @Option(name = "--input-dir", usage = "the input batch of images
> directory")
>^
> warning: no valid 

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Zheng, Da]

I think I have found the root of the problem.

The tutorial loads a model from onnx, which uses padding (left, right, top, 
bottom). But mxnet uses padding (height, width). Currently, when an ONNX model 
is loaded to MXNet, the padding is converted correctly. MXNet conv doesn't 
check the number of elements in the padding and ignores the problem. However, 
mxnet-mkldnn checks it and fails.

The correct way of fixing this issue is to check the number of elements in the 
padding tuple in mxnet conv. If the tuple size mismatches, it should fail. When 
an ONNX model is loaded to MXNet, the padding should be converted correctly.

For the time being, I'll just fix MKLDNN so it doesn't check the tuple length 
of padding.

Best,
Da

On 4/23/18, 2:58 PM, "Zheng, Da"  wrote:

I can reproduce the bug now. I'm working on a fix for the bug.

Currently, there are a few more bug fixes for MKLDNN.
https://github.com/apache/incubator-mxnet/pull/10651
https://github.com/apache/incubator-mxnet/pull/10624
https://github.com/apache/incubator-mxnet/pull/10619/files
https://github.com/apache/incubator-mxnet/pull/10616
https://github.com/apache/incubator-mxnet/pull/10591/files

They are ready for review.

I just discussed with @Anirudh. Maybe we should say in the release note 
that MKLDNN in MXNet is still experimental.
What do you think?

Best,
Da

On 4/21/18, 7:59 PM, "Zheng, Da"  wrote:

It seems I have problems of compiling scala when running "make docs". 
Please see the error below.

Are there any instructions of compiling these scala code? I guess I 
might miss some packages.
I tried installing libslf4j-java and didn't help.

Best,
Da

Execute "cd 
/home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/docs/../scala-package; 
scaladoc `find . -type f -name "*.scala" | egrep "\/core|\/infer" | egrep -v 
"Suite"`; exit 0"

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:24:
 error: object kohsuke is not a member of package org
import org.kohsuke.args4j.{CmdLineParser, Option}
   ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:25:
 error: object slf4j is not a member of package org
import org.slf4j.LoggerFactory
   ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:41:
 error: class Option is abstract; cannot be instantiated
  @Option(name = "--model-path-prefix", usage = "the input model 
directory and prefix of the model")
   ^
warning: no valid targets for annotation on value modelPathPrefix - it 
is discarded unused. You may specify targets with meta-annotations, e.g. 
@( @getter)

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:43:
 error: class Option is abstract; cannot be instantiated
  @Option(name = "--input-image", usage = "the input image")
   ^
warning: no valid targets for annotation on value inputImagePath - it 
is discarded unused. You may specify targets with meta-annotations, e.g. 
@( @getter)

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:45:
 error: class Option is abstract; cannot be instantiated
  @Option(name = "--input-dir", usage = "the input batch of images 
directory")
   ^
warning: no valid targets for annotation on value inputImageDir - it is 
discarded unused. You may specify targets with meta-annotations, e.g. @( 
@getter)

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:51:
 error: not found: value LoggerFactory
  private val logger = 
LoggerFactory.getLogger(classOf[SSDClassifierExample])
   ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:109:
 error: not found: type CmdLineParser
val parser : CmdLineParser = new CmdLineParser(inst)
 ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:109:
 error: not found: type CmdLineParser
val parser : CmdLineParser = new CmdLineParser(inst)
 ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/imageclassifier/ImageClassifierExample.scala:21:
 error: object kohsuke is not a member of package org
import org.kohsuke.args4j.{CmdLineParser, Option}
   ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/imageclassifier/ImageClassifierExample.scala:22:
 error: object slf4j is not a member of package org

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Anirudh]

Hi all,

Thank you Thomas for reporting this issue, and thanks Da for working on the
fix!
We(Da, Steffen, Haibin and me) had a discussion offline about this.

Next steps are the following:
1. Da will be working on the fix for the tutorials(onnx/super_resolution)
and marking MKL-DNN as "Experimental" in the release notes. After this I
will cut RC1.
2. We can hold off on the other four fixes from going into the release
until there are more comprehensive tests in-place.

I am cancelling the RC0 vote. We will make the necessary changes, cut a RC1
and have a vote again.
If you have any concerns or disagreements, please let me know.

Anirudh

On Mon, Apr 23, 2018 at 2:58 PM, Zheng, Da  wrote:

> I can reproduce the bug now. I'm working on a fix for the bug.
>
> Currently, there are a few more bug fixes for MKLDNN.
> https://github.com/apache/incubator-mxnet/pull/10651
> https://github.com/apache/incubator-mxnet/pull/10624
> https://github.com/apache/incubator-mxnet/pull/10619/files
> https://github.com/apache/incubator-mxnet/pull/10616
> https://github.com/apache/incubator-mxnet/pull/10591/files
>
> They are ready for review.
>
> I just discussed with @Anirudh. Maybe we should say in the release note
> that MKLDNN in MXNet is still experimental.
> What do you think?
>
> Best,
> Da
>
> On 4/21/18, 7:59 PM, "Zheng, Da"  wrote:
>
> It seems I have problems of compiling scala when running "make docs".
> Please see the error below.
>
> Are there any instructions of compiling these scala code? I guess I
> might miss some packages.
> I tried installing libslf4j-java and didn't help.
>
> Best,
> Da
>
> Execute "cd /home/ubuntu/apache-mxnet-src-
> 1.2.0.rc0-incubating/docs/../scala-package; scaladoc `find . -type f
> -name "*.scala" | egrep "\/core|\/infer" | egrep -v "Suite"`; exit 0"
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> objectdetector/SSDClassifierExample.scala:24: error: object kohsuke is
> not a member of package org
> import org.kohsuke.args4j.{CmdLineParser, Option}
>^
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> objectdetector/SSDClassifierExample.scala:25: error: object slf4j is not
> a member of package org
> import org.slf4j.LoggerFactory
>^
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> objectdetector/SSDClassifierExample.scala:41: error: class Option is
> abstract; cannot be instantiated
>   @Option(name = "--model-path-prefix", usage = "the input model
> directory and prefix of the model")
>^
> warning: no valid targets for annotation on value modelPathPrefix - it
> is discarded unused. You may specify targets with meta-annotations, e.g.
> @( @getter)
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> objectdetector/SSDClassifierExample.scala:43: error: class Option is
> abstract; cannot be instantiated
>   @Option(name = "--input-image", usage = "the input image")
>^
> warning: no valid targets for annotation on value inputImagePath - it
> is discarded unused. You may specify targets with meta-annotations, e.g.
> @( @getter)
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> objectdetector/SSDClassifierExample.scala:45: error: class Option is
> abstract; cannot be instantiated
>   @Option(name = "--input-dir", usage = "the input batch of images
> directory")
>^
> warning: no valid targets for annotation on value inputImageDir - it
> is discarded unused. You may specify targets with meta-annotations, e.g.
> @( @getter)
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> objectdetector/SSDClassifierExample.scala:51: error: not found: value
> LoggerFactory
>   private val logger = LoggerFactory.getLogger(
> classOf[SSDClassifierExample])
>^
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> objectdetector/SSDClassifierExample.scala:109: error: not found: type
> CmdLineParser
> val parser : CmdLineParser = new CmdLineParser(inst)
>  ^
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> objectdetector/SSDClassifierExample.scala:109: error: not found: type
> CmdLineParser
> val parser : CmdLineParser = new CmdLineParser(inst)
>  ^
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> imageclassifier/ImageClassifierExample.scala:21: error: object kohsuke is
> not a member of package org
> import org.kohsuke.args4j.{CmdLineParser, Option}
>^
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> imageclassifier/ImageClassifierExample.scala:22: error: object slf4j is
> not a member of package org
> import org.slf4j.LoggerFactory
>^
> ./examples/src/main/scala/org/apache/mxnetexamples/infer/
> imageclassifier/ImageClassifierExample.scala:40: error: not 

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Zheng, Da]

I can reproduce the bug now. I'm working on a fix for the bug.

Currently, there are a few more bug fixes for MKLDNN.
https://github.com/apache/incubator-mxnet/pull/10651
https://github.com/apache/incubator-mxnet/pull/10624
https://github.com/apache/incubator-mxnet/pull/10619/files
https://github.com/apache/incubator-mxnet/pull/10616
https://github.com/apache/incubator-mxnet/pull/10591/files

They are ready for review.

I just discussed with @Anirudh. Maybe we should say in the release note that 
MKLDNN in MXNet is still experimental.
What do you think?

Best,
Da

On 4/21/18, 7:59 PM, "Zheng, Da"  wrote:

It seems I have problems of compiling scala when running "make docs". 
Please see the error below.

Are there any instructions of compiling these scala code? I guess I might 
miss some packages.
I tried installing libslf4j-java and didn't help.

Best,
Da

Execute "cd 
/home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/docs/../scala-package; 
scaladoc `find . -type f -name "*.scala" | egrep "\/core|\/infer" | egrep -v 
"Suite"`; exit 0"

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:24:
 error: object kohsuke is not a member of package org
import org.kohsuke.args4j.{CmdLineParser, Option}
   ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:25:
 error: object slf4j is not a member of package org
import org.slf4j.LoggerFactory
   ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:41:
 error: class Option is abstract; cannot be instantiated
  @Option(name = "--model-path-prefix", usage = "the input model directory 
and prefix of the model")
   ^
warning: no valid targets for annotation on value modelPathPrefix - it is 
discarded unused. You may specify targets with meta-annotations, e.g. @( 
@getter)

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:43:
 error: class Option is abstract; cannot be instantiated
  @Option(name = "--input-image", usage = "the input image")
   ^
warning: no valid targets for annotation on value inputImagePath - it is 
discarded unused. You may specify targets with meta-annotations, e.g. @( 
@getter)

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:45:
 error: class Option is abstract; cannot be instantiated
  @Option(name = "--input-dir", usage = "the input batch of images 
directory")
   ^
warning: no valid targets for annotation on value inputImageDir - it is 
discarded unused. You may specify targets with meta-annotations, e.g. @( 
@getter)

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:51:
 error: not found: value LoggerFactory
  private val logger = 
LoggerFactory.getLogger(classOf[SSDClassifierExample])
   ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:109:
 error: not found: type CmdLineParser
val parser : CmdLineParser = new CmdLineParser(inst)
 ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/objectdetector/SSDClassifierExample.scala:109:
 error: not found: type CmdLineParser
val parser : CmdLineParser = new CmdLineParser(inst)
 ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/imageclassifier/ImageClassifierExample.scala:21:
 error: object kohsuke is not a member of package org
import org.kohsuke.args4j.{CmdLineParser, Option}
   ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/imageclassifier/ImageClassifierExample.scala:22:
 error: object slf4j is not a member of package org
import org.slf4j.LoggerFactory
   ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/imageclassifier/ImageClassifierExample.scala:40:
 error: not found: value LoggerFactory
  private val logger = 
LoggerFactory.getLogger(classOf[ImageClassifierExample])
   ^

./examples/src/main/scala/org/apache/mxnetexamples/infer/imageclassifier/ImageClassifierExample.scala:108:
 error: not found: type CmdLineParser
val parser: CmdLineParser = new CmdLineParser(inst)
^

./examples/src/main/scala/org/apache/mxnetexamples/infer/imageclassifier/ImageClassifierExample.scala:108:
 error: not found: type CmdLineParser
val parser: CmdLineParser = new CmdLineParser(inst)
^

./examples/src/main/scala/org/apache/mxnetexamples/infer/imageclassifier/ImageClassifierExample.scala:157:
 error: class Option is abstract; cannot be instantiated
  @Option(name = "--model-path-prefix", usage = "the input model 

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Anirudh Acharya]

@ThomasDelteil Sorry for the late reply. Yes, it should indeed work when
compiled with mkldnn and cuda.

On Sat, Apr 21, 2018 at 5:15 PM, Thomas DELTEIL 
wrote:

> @Anirudh, thanks for looking into it! However I do not understand what you
> mean by 'set as CPU and not GPU'? MXNet compiled with mkldnn and cuda is
> supposed to be able to work with both context no? There are other tutorials
> that are running successfully on both CPU and GPU context.
>
> @Da to reproduce:
>
> Download the source of 1.2.0.rc0 and extract it, cd into it.
>
> make docs
> make clean
> make -j $(nproc) USE_OPENCV=1 USE_BLAS=openblas USE_CUDA=1
> USE_CUDA_PATH=/usr/local/cuda USE_CUDNN=1 USE_MKLDNN=1
> export PYTHONPATH=$(pwd)/python
> cd tests/nightly
> python test_tutorial.py --tutorial onnx/super_resolution
>
> you can also start a jupyter notebook server and try to run
> docs/_build/html/tutorials/onnx/super_resolution.ipynb
>
>
>
> 2018-04-21 15:08 GMT-07:00 Zheng, Da :
>
> > @ThomasDelteil could you show me how to reproduce the problem? I'll take
> > it a look as well.
> >
> > Best,
> > Da
> >
> > Sent from my iPhone
> >
> > On Apr 21, 2018, at 1:12 PM, Anirudh Acharya  > > wrote:
> >
> > @ThomasDelteil that might be due to the fact that in the example, the
> > context is being set as CPU and not GPU.
> > But I will still take a look as soon as possible.
> >
> >
> > Regards
> > Anirudh
> >
> > On Sat, Apr 21, 2018 at 11:10 AM, Thomas DELTEIL <
> > thomas.delte...@gmail.com>
> > wrote:
> >
> > *-0*
> >
> > compiled from source on GPU CUDA/CUDNN, tutorials run fine.
> >
> > However:
> > Compiled from source and adding USE_MKLDNN=1, the onnx/super_resolution
> > tutorial is crashing on this line:
> >
> > ```
> > from collections import namedtuple
> > Batch = namedtuple('Batch', ['data'])
> >
> > # forward on the provided data batch
> > mod.forward(Batch([mx.nd.array(test_image)]))
> > ```
> >
> > Stack trace returned 8 entries:
> > [bt] (0)
> > /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> > mxnet/../../lib/libmxnet.so(dmlc::StackTrace[abi:cxx11]()+0x5b)
> > [0x7feef615721b]
> > [bt] (1)
> > /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> > mxnet/../../lib/libmxnet.so(dmlc::LogMessageFatal::~
> > LogMessageFatal()+0x28)
> > [0x7feef6158258]
> > [bt] (2)
> > /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> > mxnet/../../lib/libmxnet.so(mxnet::engine::ThreadedEngine:
> > :ExecuteOprBlock(mxnet::RunContext,
> > mxnet::engine::OprBlock*)+0xfa9) [0x7feef8b1ad49]
> > [bt] (3)
> > /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> > mxnet/../../lib/libmxnet.so(std::_Function_handler > (std::shared_ptr),
> > mxnet::engine::ThreadedEnginePerDevice::PushToExecute(mxnet::engine::
> > OprBlock*,
> > bool)::{lambda()#1}::operator()()
> > const::{lambda(std::shared_ptr)#1}>::_
> > M_invoke(std::_Any_data
> > const&, std::shared_ptr&&)+0xe2) [0x7feef8b30d82]
> > [bt] (4)
> > /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> > mxnet/../../lib/libmxnet.so(std::thread::_Impl > simple (std::shared_ptr)> (std::shared_ptr > ManualEvent>)>
> > ::_M_run()+0x4a) [0x7feef8b2af1a]
> > [bt] (5) /home/ubuntu/anaconda3/bin/../lib/libstdc++.so.6(+0xafc5c)
> > [0x7fef7cc79c5c]
> > [bt] (6) /lib/x86_64-linux-gnu/libpthread.so.0(+0x76ba) [0x7fef7dec36ba]
> > [bt] (7) /lib/x86_64-linux-gnu/libc.so.6(clone+0x6d) [0x7fef7dbf941d]
> >
> > Depending on how experimental we consider MKLDNN, that could be a *-1
> *for
> > me.
> >
> > 2018-04-21 9:01 GMT-07:00 Jun Wu  > jun@gmail.com>>:
> >
> > +1
> >
> > Compiled from source. Ran the model quantization example. Both quantized
> > model generation and inference can run successfully.
> >
> > On Fri, Apr 20, 2018 at 5:14 PM, Indhu  indhubhara...@gmail.com>> wrote:
> >
> > +1
> >
> > Compiled from source on P3 instance. Tested the SSD example and some
> > Gluon
> > examples.
> >
> > On Wed, Apr 18, 2018, 7:40 PM Anirudh  anirudh2...@gmail.com>> wrote:
> >
> > Hi everyone,
> >
> > This is a vote to release Apache MXNet (incubating) version 1.2.0.
> > Voting
> > will start now (Wednesday, April 18th) and end at 7:40 PM PDT,
> > Saturday,
> > April 21st.
> >
> > Link to the release notes:
> >
> >
> > https://cwiki.apache.org/confluence/display/MXNET/
> > Apache+MXNet+%28incubating%29+1.2.0+Release+Notes
> >
> > Link to the release candidate 1.2.0.rc0:
> > https://github.com/apache/incubator-mxnet/releases/tag/1.2.0.rc0
> >
> > View this page, click on "Build from Source", and use the source code
> > obtained from the 1.2.0.rc0 tag:
> > https://mxnet.incubator.apache.org/install/index.html
> >
> > (Note: The README.md points to the 1.2.0 tag and does not work at the
> > moment.)
> >
> > Please remember to TEST first before voting 

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Thomas DELTEIL]

@Anirudh, thanks for looking into it! However I do not understand what you
mean by 'set as CPU and not GPU'? MXNet compiled with mkldnn and cuda is
supposed to be able to work with both context no? There are other tutorials
that are running successfully on both CPU and GPU context.

@Da to reproduce:

Download the source of 1.2.0.rc0 and extract it, cd into it.

make docs
make clean
make -j $(nproc) USE_OPENCV=1 USE_BLAS=openblas USE_CUDA=1
USE_CUDA_PATH=/usr/local/cuda USE_CUDNN=1 USE_MKLDNN=1
export PYTHONPATH=$(pwd)/python
cd tests/nightly
python test_tutorial.py --tutorial onnx/super_resolution

you can also start a jupyter notebook server and try to run
docs/_build/html/tutorials/onnx/super_resolution.ipynb



2018-04-21 15:08 GMT-07:00 Zheng, Da :

> @ThomasDelteil could you show me how to reproduce the problem? I'll take
> it a look as well.
>
> Best,
> Da
>
> Sent from my iPhone
>
> On Apr 21, 2018, at 1:12 PM, Anirudh Acharya  > wrote:
>
> @ThomasDelteil that might be due to the fact that in the example, the
> context is being set as CPU and not GPU.
> But I will still take a look as soon as possible.
>
>
> Regards
> Anirudh
>
> On Sat, Apr 21, 2018 at 11:10 AM, Thomas DELTEIL <
> thomas.delte...@gmail.com>
> wrote:
>
> *-0*
>
> compiled from source on GPU CUDA/CUDNN, tutorials run fine.
>
> However:
> Compiled from source and adding USE_MKLDNN=1, the onnx/super_resolution
> tutorial is crashing on this line:
>
> ```
> from collections import namedtuple
> Batch = namedtuple('Batch', ['data'])
>
> # forward on the provided data batch
> mod.forward(Batch([mx.nd.array(test_image)]))
> ```
>
> Stack trace returned 8 entries:
> [bt] (0)
> /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> mxnet/../../lib/libmxnet.so(dmlc::StackTrace[abi:cxx11]()+0x5b)
> [0x7feef615721b]
> [bt] (1)
> /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> mxnet/../../lib/libmxnet.so(dmlc::LogMessageFatal::~
> LogMessageFatal()+0x28)
> [0x7feef6158258]
> [bt] (2)
> /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> mxnet/../../lib/libmxnet.so(mxnet::engine::ThreadedEngine:
> :ExecuteOprBlock(mxnet::RunContext,
> mxnet::engine::OprBlock*)+0xfa9) [0x7feef8b1ad49]
> [bt] (3)
> /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> mxnet/../../lib/libmxnet.so(std::_Function_handler (std::shared_ptr),
> mxnet::engine::ThreadedEnginePerDevice::PushToExecute(mxnet::engine::
> OprBlock*,
> bool)::{lambda()#1}::operator()()
> const::{lambda(std::shared_ptr)#1}>::_
> M_invoke(std::_Any_data
> const&, std::shared_ptr&&)+0xe2) [0x7feef8b30d82]
> [bt] (4)
> /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> mxnet/../../lib/libmxnet.so(std::thread::_Impl simple (std::shared_ptr ManualEvent>)>
> ::_M_run()+0x4a) [0x7feef8b2af1a]
> [bt] (5) /home/ubuntu/anaconda3/bin/../lib/libstdc++.so.6(+0xafc5c)
> [0x7fef7cc79c5c]
> [bt] (6) /lib/x86_64-linux-gnu/libpthread.so.0(+0x76ba) [0x7fef7dec36ba]
> [bt] (7) /lib/x86_64-linux-gnu/libc.so.6(clone+0x6d) [0x7fef7dbf941d]
>
> Depending on how experimental we consider MKLDNN, that could be a *-1 *for
> me.
>
> 2018-04-21 9:01 GMT-07:00 Jun Wu  jun@gmail.com>>:
>
> +1
>
> Compiled from source. Ran the model quantization example. Both quantized
> model generation and inference can run successfully.
>
> On Fri, Apr 20, 2018 at 5:14 PM, Indhu > wrote:
>
> +1
>
> Compiled from source on P3 instance. Tested the SSD example and some
> Gluon
> examples.
>
> On Wed, Apr 18, 2018, 7:40 PM Anirudh > wrote:
>
> Hi everyone,
>
> This is a vote to release Apache MXNet (incubating) version 1.2.0.
> Voting
> will start now (Wednesday, April 18th) and end at 7:40 PM PDT,
> Saturday,
> April 21st.
>
> Link to the release notes:
>
>
> https://cwiki.apache.org/confluence/display/MXNET/
> Apache+MXNet+%28incubating%29+1.2.0+Release+Notes
>
> Link to the release candidate 1.2.0.rc0:
> https://github.com/apache/incubator-mxnet/releases/tag/1.2.0.rc0
>
> View this page, click on "Build from Source", and use the source code
> obtained from the 1.2.0.rc0 tag:
> https://mxnet.incubator.apache.org/install/index.html
>
> (Note: The README.md points to the 1.2.0 tag and does not work at the
> moment.)
>
> Please remember to TEST first before voting accordingly:
> +1 = approve
> +0 = no opinion
> -1 = disapprove (provide reason)
>
> Thanks,
>
> Anirudh
>
>
>
>
>

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Zheng, Da]

@ThomasDelteil could you show me how to reproduce the problem? I'll take it a 
look as well.

Best,
Da

Sent from my iPhone

On Apr 21, 2018, at 1:12 PM, Anirudh Acharya 
> wrote:

@ThomasDelteil that might be due to the fact that in the example, the
context is being set as CPU and not GPU.
But I will still take a look as soon as possible.


Regards
Anirudh

On Sat, Apr 21, 2018 at 11:10 AM, Thomas DELTEIL 
>
wrote:

*-0*

compiled from source on GPU CUDA/CUDNN, tutorials run fine.

However:
Compiled from source and adding USE_MKLDNN=1, the onnx/super_resolution
tutorial is crashing on this line:

```
from collections import namedtuple
Batch = namedtuple('Batch', ['data'])

# forward on the provided data batch
mod.forward(Batch([mx.nd.array(test_image)]))
```

Stack trace returned 8 entries:
[bt] (0)
/home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
mxnet/../../lib/libmxnet.so(dmlc::StackTrace[abi:cxx11]()+0x5b)
[0x7feef615721b]
[bt] (1)
/home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
mxnet/../../lib/libmxnet.so(dmlc::LogMessageFatal::~
LogMessageFatal()+0x28)
[0x7feef6158258]
[bt] (2)
/home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
mxnet/../../lib/libmxnet.so(mxnet::engine::ThreadedEngine:
:ExecuteOprBlock(mxnet::RunContext,
mxnet::engine::OprBlock*)+0xfa9) [0x7feef8b1ad49]
[bt] (3)
/home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
mxnet/../../lib/libmxnet.so(std::_Function_handler),
mxnet::engine::ThreadedEnginePerDevice::PushToExecute(mxnet::engine::
OprBlock*,
bool)::{lambda()#1}::operator()()
const::{lambda(std::shared_ptr)#1}>::_
M_invoke(std::_Any_data
const&, std::shared_ptr&&)+0xe2) [0x7feef8b30d82]
[bt] (4)
/home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
mxnet/../../lib/libmxnet.so(std::thread::_Impl)> (std::shared_ptr)>
::_M_run()+0x4a) [0x7feef8b2af1a]
[bt] (5) /home/ubuntu/anaconda3/bin/../lib/libstdc++.so.6(+0xafc5c)
[0x7fef7cc79c5c]
[bt] (6) /lib/x86_64-linux-gnu/libpthread.so.0(+0x76ba) [0x7fef7dec36ba]
[bt] (7) /lib/x86_64-linux-gnu/libc.so.6(clone+0x6d) [0x7fef7dbf941d]

Depending on how experimental we consider MKLDNN, that could be a *-1 *for
me.

2018-04-21 9:01 GMT-07:00 Jun Wu 
>:

+1

Compiled from source. Ran the model quantization example. Both quantized
model generation and inference can run successfully.

On Fri, Apr 20, 2018 at 5:14 PM, Indhu 
> wrote:

+1

Compiled from source on P3 instance. Tested the SSD example and some
Gluon
examples.

On Wed, Apr 18, 2018, 7:40 PM Anirudh 
> wrote:

Hi everyone,

This is a vote to release Apache MXNet (incubating) version 1.2.0.
Voting
will start now (Wednesday, April 18th) and end at 7:40 PM PDT,
Saturday,
April 21st.

Link to the release notes:


https://cwiki.apache.org/confluence/display/MXNET/
Apache+MXNet+%28incubating%29+1.2.0+Release+Notes

Link to the release candidate 1.2.0.rc0:
https://github.com/apache/incubator-mxnet/releases/tag/1.2.0.rc0

View this page, click on "Build from Source", and use the source code
obtained from the 1.2.0.rc0 tag:
https://mxnet.incubator.apache.org/install/index.html

(Note: The README.md points to the 1.2.0 tag and does not work at the
moment.)

Please remember to TEST first before voting accordingly:
+1 = approve
+0 = no opinion
-1 = disapprove (provide reason)

Thanks,

Anirudh

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Anirudh Acharya]

@ThomasDelteil that might be due to the fact that in the example, the
context is being set as CPU and not GPU.
But I will still take a look as soon as possible.


Regards
Anirudh

On Sat, Apr 21, 2018 at 11:10 AM, Thomas DELTEIL 
wrote:

> *-0*
>
> compiled from source on GPU CUDA/CUDNN, tutorials run fine.
>
> However:
> Compiled from source and adding USE_MKLDNN=1, the onnx/super_resolution
> tutorial is crashing on this line:
>
> ```
> from collections import namedtuple
> Batch = namedtuple('Batch', ['data'])
>
> # forward on the provided data batch
> mod.forward(Batch([mx.nd.array(test_image)]))
> ```
>
> Stack trace returned 8 entries:
> [bt] (0)
> /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> mxnet/../../lib/libmxnet.so(dmlc::StackTrace[abi:cxx11]()+0x5b)
> [0x7feef615721b]
> [bt] (1)
> /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> mxnet/../../lib/libmxnet.so(dmlc::LogMessageFatal::~
> LogMessageFatal()+0x28)
> [0x7feef6158258]
> [bt] (2)
> /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> mxnet/../../lib/libmxnet.so(mxnet::engine::ThreadedEngine:
> :ExecuteOprBlock(mxnet::RunContext,
> mxnet::engine::OprBlock*)+0xfa9) [0x7feef8b1ad49]
> [bt] (3)
> /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> mxnet/../../lib/libmxnet.so(std::_Function_handler (std::shared_ptr),
> mxnet::engine::ThreadedEnginePerDevice::PushToExecute(mxnet::engine::
> OprBlock*,
> bool)::{lambda()#1}::operator()()
> const::{lambda(std::shared_ptr)#1}>::_
> M_invoke(std::_Any_data
> const&, std::shared_ptr&&)+0xe2) [0x7feef8b30d82]
> [bt] (4)
> /home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/
> mxnet/../../lib/libmxnet.so(std::thread::_Impl simple (std::shared_ptr ManualEvent>)>
> >::_M_run()+0x4a) [0x7feef8b2af1a]
> [bt] (5) /home/ubuntu/anaconda3/bin/../lib/libstdc++.so.6(+0xafc5c)
> [0x7fef7cc79c5c]
> [bt] (6) /lib/x86_64-linux-gnu/libpthread.so.0(+0x76ba) [0x7fef7dec36ba]
> [bt] (7) /lib/x86_64-linux-gnu/libc.so.6(clone+0x6d) [0x7fef7dbf941d]
>
> Depending on how experimental we consider MKLDNN, that could be a *-1 *for
> me.
>
> 2018-04-21 9:01 GMT-07:00 Jun Wu :
>
> > +1
> >
> > Compiled from source. Ran the model quantization example. Both quantized
> > model generation and inference can run successfully.
> >
> > On Fri, Apr 20, 2018 at 5:14 PM, Indhu  wrote:
> >
> > > +1
> > >
> > > Compiled from source on P3 instance. Tested the SSD example and some
> > Gluon
> > > examples.
> > >
> > > On Wed, Apr 18, 2018, 7:40 PM Anirudh  wrote:
> > >
> > > > Hi everyone,
> > > >
> > > > This is a vote to release Apache MXNet (incubating) version 1.2.0.
> > Voting
> > > > will start now (Wednesday, April 18th) and end at 7:40 PM PDT,
> > Saturday,
> > > > April 21st.
> > > >
> > > > Link to the release notes:
> > > >
> > > >
> > > > https://cwiki.apache.org/confluence/display/MXNET/
> > > Apache+MXNet+%28incubating%29+1.2.0+Release+Notes
> > > >
> > > > Link to the release candidate 1.2.0.rc0:
> > > > https://github.com/apache/incubator-mxnet/releases/tag/1.2.0.rc0
> > > >
> > > > View this page, click on “Build from Source”, and use the source code
> > > > obtained from the 1.2.0.rc0 tag:
> > > > https://mxnet.incubator.apache.org/install/index.html
> > > >
> > > > (Note: The README.md points to the 1.2.0 tag and does not work at the
> > > > moment.)
> > > >
> > > > Please remember to TEST first before voting accordingly:
> > > > +1 = approve
> > > > +0 = no opinion
> > > > -1 = disapprove (provide reason)
> > > >
> > > > Thanks,
> > > >
> > > > Anirudh
> > > >
> > >
> >
>

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Thomas DELTEIL]

*-0*

compiled from source on GPU CUDA/CUDNN, tutorials run fine.

However:
Compiled from source and adding USE_MKLDNN=1, the onnx/super_resolution
tutorial is crashing on this line:

```
from collections import namedtuple
Batch = namedtuple('Batch', ['data'])

# forward on the provided data batch
mod.forward(Batch([mx.nd.array(test_image)]))
```

Stack trace returned 8 entries:
[bt] (0)
/home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/mxnet/../../lib/libmxnet.so(dmlc::StackTrace[abi:cxx11]()+0x5b)
[0x7feef615721b]
[bt] (1)
/home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/mxnet/../../lib/libmxnet.so(dmlc::LogMessageFatal::~LogMessageFatal()+0x28)
[0x7feef6158258]
[bt] (2)
/home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/mxnet/../../lib/libmxnet.so(mxnet::engine::ThreadedEngine::ExecuteOprBlock(mxnet::RunContext,
mxnet::engine::OprBlock*)+0xfa9) [0x7feef8b1ad49]
[bt] (3)
/home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/mxnet/../../lib/libmxnet.so(std::_Function_handler),
mxnet::engine::ThreadedEnginePerDevice::PushToExecute(mxnet::engine::OprBlock*,
bool)::{lambda()#1}::operator()()
const::{lambda(std::shared_ptr)#1}>::_M_invoke(std::_Any_data
const&, std::shared_ptr&&)+0xe2) [0x7feef8b30d82]
[bt] (4)
/home/ubuntu/apache-mxnet-src-1.2.0.rc0-incubating/python/mxnet/../../lib/libmxnet.so(std::thread::_Impl (std::shared_ptr)>
>::_M_run()+0x4a) [0x7feef8b2af1a]
[bt] (5) /home/ubuntu/anaconda3/bin/../lib/libstdc++.so.6(+0xafc5c)
[0x7fef7cc79c5c]
[bt] (6) /lib/x86_64-linux-gnu/libpthread.so.0(+0x76ba) [0x7fef7dec36ba]
[bt] (7) /lib/x86_64-linux-gnu/libc.so.6(clone+0x6d) [0x7fef7dbf941d]

Depending on how experimental we consider MKLDNN, that could be a *-1 *for
me.

2018-04-21 9:01 GMT-07:00 Jun Wu :

> +1
>
> Compiled from source. Ran the model quantization example. Both quantized
> model generation and inference can run successfully.
>
> On Fri, Apr 20, 2018 at 5:14 PM, Indhu  wrote:
>
> > +1
> >
> > Compiled from source on P3 instance. Tested the SSD example and some
> Gluon
> > examples.
> >
> > On Wed, Apr 18, 2018, 7:40 PM Anirudh  wrote:
> >
> > > Hi everyone,
> > >
> > > This is a vote to release Apache MXNet (incubating) version 1.2.0.
> Voting
> > > will start now (Wednesday, April 18th) and end at 7:40 PM PDT,
> Saturday,
> > > April 21st.
> > >
> > > Link to the release notes:
> > >
> > >
> > > https://cwiki.apache.org/confluence/display/MXNET/
> > Apache+MXNet+%28incubating%29+1.2.0+Release+Notes
> > >
> > > Link to the release candidate 1.2.0.rc0:
> > > https://github.com/apache/incubator-mxnet/releases/tag/1.2.0.rc0
> > >
> > > View this page, click on “Build from Source”, and use the source code
> > > obtained from the 1.2.0.rc0 tag:
> > > https://mxnet.incubator.apache.org/install/index.html
> > >
> > > (Note: The README.md points to the 1.2.0 tag and does not work at the
> > > moment.)
> > >
> > > Please remember to TEST first before voting accordingly:
> > > +1 = approve
> > > +0 = no opinion
> > > -1 = disapprove (provide reason)
> > >
> > > Thanks,
> > >
> > > Anirudh
> > >
> >
>

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Jun Wu]

+1

Compiled from source. Ran the model quantization example. Both quantized
model generation and inference can run successfully.

On Fri, Apr 20, 2018 at 5:14 PM, Indhu  wrote:

> +1
>
> Compiled from source on P3 instance. Tested the SSD example and some Gluon
> examples.
>
> On Wed, Apr 18, 2018, 7:40 PM Anirudh  wrote:
>
> > Hi everyone,
> >
> > This is a vote to release Apache MXNet (incubating) version 1.2.0. Voting
> > will start now (Wednesday, April 18th) and end at 7:40 PM PDT, Saturday,
> > April 21st.
> >
> > Link to the release notes:
> >
> >
> > https://cwiki.apache.org/confluence/display/MXNET/
> Apache+MXNet+%28incubating%29+1.2.0+Release+Notes
> >
> > Link to the release candidate 1.2.0.rc0:
> > https://github.com/apache/incubator-mxnet/releases/tag/1.2.0.rc0
> >
> > View this page, click on “Build from Source”, and use the source code
> > obtained from the 1.2.0.rc0 tag:
> > https://mxnet.incubator.apache.org/install/index.html
> >
> > (Note: The README.md points to the 1.2.0 tag and does not work at the
> > moment.)
> >
> > Please remember to TEST first before voting accordingly:
> > +1 = approve
> > +0 = no opinion
> > -1 = disapprove (provide reason)
> >
> > Thanks,
> >
> > Anirudh
> >
>

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Indhu]

+1

Compiled from source on P3 instance. Tested the SSD example and some Gluon
examples.

On Wed, Apr 18, 2018, 7:40 PM Anirudh  wrote:

> Hi everyone,
>
> This is a vote to release Apache MXNet (incubating) version 1.2.0. Voting
> will start now (Wednesday, April 18th) and end at 7:40 PM PDT, Saturday,
> April 21st.
>
> Link to the release notes:
>
>
> https://cwiki.apache.org/confluence/display/MXNET/Apache+MXNet+%28incubating%29+1.2.0+Release+Notes
>
> Link to the release candidate 1.2.0.rc0:
> https://github.com/apache/incubator-mxnet/releases/tag/1.2.0.rc0
>
> View this page, click on “Build from Source”, and use the source code
> obtained from the 1.2.0.rc0 tag:
> https://mxnet.incubator.apache.org/install/index.html
>
> (Note: The README.md points to the 1.2.0 tag and does not work at the
> moment.)
>
> Please remember to TEST first before voting accordingly:
> +1 = approve
> +0 = no opinion
> -1 = disapprove (provide reason)
>
> Thanks,
>
> Anirudh
>

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Rahul Huilgol]

+1

Compiled from source and verified distributed training with float32 and
float16

Regards,
Rahul

On Thu, Apr 19, 2018 at 4:30 PM, Anirudh Acharya 
wrote:

> +1
>
> Checked the following -
>
>- source compiles
>- the tests for the onnx import API passes.
>
>
> Regards
> Anirudh
>
>
> On Thu, Apr 19, 2018 at 1:11 PM, Meghna Baijal  >
> wrote:
>
> > +1 (non-binding)
> >
> >
> > I Checked the following:
> >
> > 1. Signatures are ok
> >
> > 2. Source compiles
> >
> > 3. mnist test passes
> >
> >
> > Regards,
> >
> > Meghna
> >
> > On Thu, Apr 19, 2018 at 10:12 AM, Anirudh  wrote:
> >
> > > Hi all,
> > >
> > > Given the weekend, I am extending the vote deadline to Sunday evening,
> > > April 22nd 7:40 PM PDT, considering Saturday and Sunday as half days(as
> > > done before).
> > >
> > > Anirudh
> > >
> > > On Wed, Apr 18, 2018 at 7:40 PM, Anirudh 
> wrote:
> > >
> > > > Hi everyone,
> > > >
> > > > This is a vote to release Apache MXNet (incubating) version 1.2.0.
> > Voting
> > > > will start now (Wednesday, April 18th) and end at 7:40 PM PDT,
> > Saturday,
> > > > April 21st.
> > > >
> > > > Link to the release notes:
> > > >
> > > > https://cwiki.apache.org/confluence/display/MXNET/
> > > > Apache+MXNet+%28incubating%29+1.2.0+Release+Notes
> > > >
> > > > Link to the release candidate 1.2.0.rc0:
> > > > https://github.com/apache/incubator-mxnet/releases/tag/1.2.0.rc0
> > > >
> > > > View this page, click on “Build from Source”, and use the source code
> > > > obtained from the 1.2.0.rc0 tag:
> > > > https://mxnet.incubator.apache.org/install/index.html
> > > >
> > > > (Note: The README.md points to the 1.2.0 tag and does not work at the
> > > > moment.)
> > > >
> > > > Please remember to TEST first before voting accordingly:
> > > > +1 = approve
> > > > +0 = no opinion
> > > > -1 = disapprove (provide reason)
> > > >
> > > > Thanks,
> > > >
> > > > Anirudh
> > > >
> > >
> >
>



-- 
Rahul Huilgol

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Anirudh Acharya]

+1

Checked the following -

   - source compiles
   - the tests for the onnx import API passes.


Regards
Anirudh


On Thu, Apr 19, 2018 at 1:11 PM, Meghna Baijal 
wrote:

> +1 (non-binding)
>
>
> I Checked the following:
>
> 1. Signatures are ok
>
> 2. Source compiles
>
> 3. mnist test passes
>
>
> Regards,
>
> Meghna
>
> On Thu, Apr 19, 2018 at 10:12 AM, Anirudh  wrote:
>
> > Hi all,
> >
> > Given the weekend, I am extending the vote deadline to Sunday evening,
> > April 22nd 7:40 PM PDT, considering Saturday and Sunday as half days(as
> > done before).
> >
> > Anirudh
> >
> > On Wed, Apr 18, 2018 at 7:40 PM, Anirudh  wrote:
> >
> > > Hi everyone,
> > >
> > > This is a vote to release Apache MXNet (incubating) version 1.2.0.
> Voting
> > > will start now (Wednesday, April 18th) and end at 7:40 PM PDT,
> Saturday,
> > > April 21st.
> > >
> > > Link to the release notes:
> > >
> > > https://cwiki.apache.org/confluence/display/MXNET/
> > > Apache+MXNet+%28incubating%29+1.2.0+Release+Notes
> > >
> > > Link to the release candidate 1.2.0.rc0:
> > > https://github.com/apache/incubator-mxnet/releases/tag/1.2.0.rc0
> > >
> > > View this page, click on “Build from Source”, and use the source code
> > > obtained from the 1.2.0.rc0 tag:
> > > https://mxnet.incubator.apache.org/install/index.html
> > >
> > > (Note: The README.md points to the 1.2.0 tag and does not work at the
> > > moment.)
> > >
> > > Please remember to TEST first before voting accordingly:
> > > +1 = approve
> > > +0 = no opinion
> > > -1 = disapprove (provide reason)
> > >
> > > Thanks,
> > >
> > > Anirudh
> > >
> >
>

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Meghna Baijal]

+1 (non-binding)


I Checked the following:

1. Signatures are ok

2. Source compiles

3. mnist test passes


Regards,

Meghna

On Thu, Apr 19, 2018 at 10:12 AM, Anirudh  wrote:

> Hi all,
>
> Given the weekend, I am extending the vote deadline to Sunday evening,
> April 22nd 7:40 PM PDT, considering Saturday and Sunday as half days(as
> done before).
>
> Anirudh
>
> On Wed, Apr 18, 2018 at 7:40 PM, Anirudh  wrote:
>
> > Hi everyone,
> >
> > This is a vote to release Apache MXNet (incubating) version 1.2.0. Voting
> > will start now (Wednesday, April 18th) and end at 7:40 PM PDT, Saturday,
> > April 21st.
> >
> > Link to the release notes:
> >
> > https://cwiki.apache.org/confluence/display/MXNET/
> > Apache+MXNet+%28incubating%29+1.2.0+Release+Notes
> >
> > Link to the release candidate 1.2.0.rc0:
> > https://github.com/apache/incubator-mxnet/releases/tag/1.2.0.rc0
> >
> > View this page, click on “Build from Source”, and use the source code
> > obtained from the 1.2.0.rc0 tag:
> > https://mxnet.incubator.apache.org/install/index.html
> >
> > (Note: The README.md points to the 1.2.0 tag and does not work at the
> > moment.)
> >
> > Please remember to TEST first before voting accordingly:
> > +1 = approve
> > +0 = no opinion
> > -1 = disapprove (provide reason)
> >
> > Thanks,
> >
> > Anirudh
> >
>

Re: [VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Anirudh]

Hi all,

Given the weekend, I am extending the vote deadline to Sunday evening,
April 22nd 7:40 PM PDT, considering Saturday and Sunday as half days(as
done before).

Anirudh

On Wed, Apr 18, 2018 at 7:40 PM, Anirudh  wrote:

> Hi everyone,
>
> This is a vote to release Apache MXNet (incubating) version 1.2.0. Voting
> will start now (Wednesday, April 18th) and end at 7:40 PM PDT, Saturday,
> April 21st.
>
> Link to the release notes:
>
> https://cwiki.apache.org/confluence/display/MXNET/
> Apache+MXNet+%28incubating%29+1.2.0+Release+Notes
>
> Link to the release candidate 1.2.0.rc0:
> https://github.com/apache/incubator-mxnet/releases/tag/1.2.0.rc0
>
> View this page, click on “Build from Source”, and use the source code
> obtained from the 1.2.0.rc0 tag:
> https://mxnet.incubator.apache.org/install/index.html
>
> (Note: The README.md points to the 1.2.0 tag and does not work at the
> moment.)
>
> Please remember to TEST first before voting accordingly:
> +1 = approve
> +0 = no opinion
> -1 = disapprove (provide reason)
>
> Thanks,
>
> Anirudh
>

[VOTE] Release Apache MXNet (incubating) version 1.2.0.RC0

[Anirudh]

Hi everyone,

This is a vote to release Apache MXNet (incubating) version 1.2.0. Voting
will start now (Wednesday, April 18th) and end at 7:40 PM PDT, Saturday,
April 21st.

Link to the release notes:

https://cwiki.apache.org/confluence/display/MXNET/Apache+MXNet+%28incubating%29+1.2.0+Release+Notes

Link to the release candidate 1.2.0.rc0:
https://github.com/apache/incubator-mxnet/releases/tag/1.2.0.rc0

View this page, click on “Build from Source”, and use the source code
obtained from the 1.2.0.rc0 tag:
https://mxnet.incubator.apache.org/install/index.html

(Note: The README.md points to the 1.2.0 tag and does not work at the
moment.)

Please remember to TEST first before voting accordingly:
+1 = approve
+0 = no opinion
-1 = disapprove (provide reason)

Thanks,

Anirudh