mxnet-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Naveen Swamy <mnnav...@gmail.com>
Subject Re: [VOTE] Release MXNet version 0.11.0.rc2
Date Wed, 23 Aug 2017 00:17:10 GMT
Suneel/Joern,

Yes, we are addressing the Licensing issue, we are not relocating the
licenses but linking them in the top level license as specified by.

   1. Apache license guideline
   <http://www.apache.org/dev/licensing-howto.html#bundle-asf-product>.
   2. the licenses are in other submodules(dmlc-core, etc.,) that is not a
   part of mxnet repo and we cannot move them.
   3. Also if we want to release a subpackage, it's best to keep their
   respective licenses in their folder. example if we want to release mxnet-r,
   the associated license should live in the r-package sub-folder.


The changes are merge to the branch here:
https://github.com/apache/incubator-mxnet/pull/7565

Let us know if you have concerns, we will be cutting an RC in the next few
hours.

Thanks, Naveen

On Mon, Aug 21, 2017 at 7:46 PM, Suneel Marthi <smarthi@apache.org> wrote:

> I would also address the issue with multi-license files that Joern's
> pointed out in the next RC.
>
> You don't have to wait for 3 days to pass the next RC and the VOTE can be
> closed once we have atleast 3 binding +1s.
>
> On Mon, Aug 21, 2017 at 8:59 PM, Naveen Swamy <mnnaveen@gmail.com> wrote:
>
> > Joern and Chris, Thanks for the taking the time to test and providing
> your
> > input, appreciate it.
> >
> > The bug Chris found is in examples, this should not impact the product as
> > such. However, we will include the fix that Chris has, add automated
> tests
> > and create a new RC. We will also append to the license and notice files
> > the other licenses used in this repo.
> >
> > Question to Suneel/Mentors.
> > Since the changes are very specific that will be included in the new RC,
> Do
> > we need to run another VOTE for 3 days? I am wondering if we can run the
> > VOTE on the new RC for 24 hours instead. Please let us know.
> >
> > this RC(0.11.RC2) will be rolled back and a new RC will be published
> > tomorrow.
> >
> > Thanks, Naveen
> >
> >
> > On Mon, Aug 21, 2017 at 5:21 PM, Chris Olivier <cjolivier01@gmail.com>
> > wrote:
> >
> > > We can also discuss the option of adding to "known bugs" in release?
> > >
> > > On Mon, Aug 21, 2017 at 3:07 PM Suneel Marthi <smarthi@apache.org>
> > wrote:
> > >
> > > > In light of the previous -1, suggest that we cancel this vote and
> > > rollback
> > > > this Release Candidate.
> > > >
> > > > On Mon, Aug 21, 2017 at 6:03 PM, Chris Olivier <
> cjolivier01@gmail.com>
> > > > wrote:
> > > >
> > > > > Added fix in this PR:
> > > > https://github.com/apache/incubator-mxnet/pull/7545
> > > > >
> > > > > On Mon, Aug 21, 2017 at 2:29 PM, Chris Olivier <
> > cjolivier01@gmail.com>
> > > > > wrote:
> > > > >
> > > > > > If we delay to fix this before the release, I also recommend
> > changing
> > > > the
> > > > > > '=' to ':=' in the Makefile line below so that the script isn't
> run
> > > for
> > > > > > every make spawn (or a large number of times, for that matter).
I
> > > > imagine
> > > > > > this is slowing down the compile quite a bit:
> > > > > >
> > > > > > RETURN_STRING = $(shell ./prepare_mkl.sh $(MKLML_ROOT))
> > > > > >
> > > > > >
> > > > > > On Mon, Aug 21, 2017 at 2:19 PM, Chris Olivier <
> > > cjolivier01@gmail.com>
> > > > > > wrote:
> > > > > >
> > > > > >> -1
> > > > > >>
> > > > > >> Simple mnist script fails if adagrad optimizer selected
> > :(unexpected
> > > > > >> keyword 'multi_precision'). I expect other optimizers may
be
> > > > similarly?
> > > > > >>
> > > > > >> python example/image-classification/train_mnist.py --gpu
-1
> > > > > >> --optimizer=adagrad
> > > > > >>
> > > > > >> Connected to pydev debugger (build 172.3544.40)
> > > > > >> INFO:root:start with arguments Namespace(add_stn=False,
> > > batch_size=64,
> > > > > >> disp_batches=100, dtype='float32', gpus='-1', kv_store='device',
> > > > > >> load_epoch=None, lr=0.05, lr_factor=0.1, lr_step_epochs='10',
> > > > > >> model_prefix=None, mom=0.9, monitor=0, network='mlp',
> > > num_classes=10,
> > > > > >> num_epochs=20, num_examples=60000, num_layers=None,
> > > > optimizer='adagrad',
> > > > > >> test_io=0, top_k=0, wd=0.0001)
> > > > > >> Traceback (most recent call last):
> > > > > >>   File "/mnt/Terrace/Apps/clion-2017.
> 1.2/plugins/python/helpers/
> > > > > pydev/pydevd.py",
> > > > > >> line 1599, in <module>
> > > > > >>     globals = debugger.run(setup['file'], None, None, is_module)
> > > > > >>   File "/mnt/Terrace/Apps/clion-2017.
> 1.2/plugins/python/helpers/
> > > > > pydev/pydevd.py",
> > > > > >> line 1026, in run
> > > > > >>     pydev_imports.execfile(file, globals, locals)  # execute
the
> > > > script
> > > > > >>   File "example/image-classification/train_mnist.py", line
96,
> in
> > > > > >> <module>
> > > > > >>     fit.fit(args, sym, get_mnist_iter)
> > > > > >>   File "example/image-classification/common/fit.py", line
207,
> in
> > > fit
> > > > > >>     monitor            = monitor)
> > > > > >>   File "/home/local/ANT/coolivie/src/DeepLearning/mxnet/python/
> > > > > mxnet/module/base_module.py",
> > > > > >> line 466, in fit
> > > > > >>     optimizer_params=optimizer_params)
> > > > > >>   File "/home/local/ANT/coolivie/src/DeepLearning/mxnet/python/
> > > > > mxnet/module/module.py",
> > > > > >> line 507, in init_optimizer
> > > > > >>     **optimizer_params)
> > > > > >>   File "/home/local/ANT/coolivie/src/DeepLearning/mxnet/python/
> > > > > mxnet/optimizer.py",
> > > > > >> line 150, in create_optimizer
> > > > > >>     return Optimizer.opt_registry[name.lower()](**kwargs)
> > > > > >>   File "/home/local/ANT/coolivie/src/DeepLearning/mxnet/python/
> > > > > mxnet/optimizer.py",
> > > > > >> line 607, in __init__
> > > > > >>     super(AdaGrad, self).__init__(**kwargs)
> > > > > >> TypeError: __init__() got an unexpected keyword argument
> > > > > 'multi_precision'
> > > > > >>
> > > > > >>
> > > > > >> On Thu, Aug 17, 2017 at 1:57 AM, Joern Kottmann <
> > kottmann@gmail.com
> > > >
> > > > > >> wrote:
> > > > > >>
> > > > > >>> I downloaded the src distribution file and noticed the
> following
> > > > > things:
> > > > > >>> - There are many LICENSE files, and the top-level file
doesn't
> > > > contain
> > > > > >>> all licenses,  you should consolidate all those LICENSE
files
> and
> > > > > >>> place only one at the top level of the source tree,
see here
> for
> > > > > >>> instructions [1].
> > > > > >>> - The distribution contains a .DS_Store file
> > > > > >>> - The distribution contains the .git folder
> > > > > >>>
> > > > > >>> Jörn
> > > > > >>>
> > > > > >>> [1] http://www.apache.org/dev/licensing-howto.html#
> assembling-li
> > > > > >>> cense-and-notice
> > > > > >>>
> > > > > >>> On Thu, Aug 17, 2017 at 10:11 AM, Joern Kottmann <
> > > kottmann@gmail.com
> > > > >
> > > > > >>> wrote:
> > > > > >>> > The release tag can be found here (and should be
included in
> > the
> > > > vote
> > > > > >>> mail):
> > > > > >>> > https://github.com/apache/incubator-mxnet/tree/0.11.0.rc2
> > > > > >>> >
> > > > > >>> > Jörn
> > > > > >>> >
> > > > > >>> > On Thu, Aug 17, 2017 at 2:38 AM, Meghna Baijal
> > > > > >>> > <meghnabaijal2017@gmail.com> wrote:
> > > > > >>> >> This is the vote to release Apache MXNet (incubating)
> version
> > > > > 0.11.0.
> > > > > >>> >> Voting will start now (Thursday, August 17,
2017 12:37 AM
> UTC)
> > > and
> > > > > >>> >> close Monday, August 21, 2017 12:37 AM UTC.
> > > > > >>> >>
> > > > > >>> >> Link to release notes:
> > > > > >>> >> https://cwiki.apache.org/confluence/display/MXNET/v0.
> 11.0+Re
> > > > > >>> lease+Notes <https://cwiki.apache.org/
> > confluence/display/MXNET/v0.
> > > > > 11.0+R
> > > > > >>> elease+Notes>
> > > > > >>> >>
> > > > > >>> >> Link to release candidate 0.11.0.rc2:
> > > > > >>> >>
> > > > https://dist.apache.org/repos/dist/dev/incubator/mxnet/0.11.0.rc2/
> > > > > <
> > > > > >>> https://dist.apache.org/repos/dist/dev/incubator/mxnet/0.11.
> > 0.rc2/
> > > >
> > > > > >>> >>
> > > > > >>> >> View this page and scroll down to “Build
from Source” to
> build
> > > > this
> > > > > >>> project:
> > > > > >>> >> http://mxnet.incubator.apache.org/get_started/install.html
> <
> > > > > >>> http://mxnet.incubator.apache.org/get_started/install.html>
> > > > > >>> >>
> > > > > >>> >> Changes between rc1 and rc2:
> > > > > >>> >> 1. Remove WaitToRead in dist-kvstore
> > > > > >>> >>
> > > > > >>> >> Major Features in v0.11:
> > > > > >>> >> 1. CoreML Converter: https://github.com/apache/incu
> > > > > >>> bator-mxnet/blob/master/tools/coreml/README.md <
> > > > > >>> https://github.com/apache/incubator-mxnet/blob/master/tools
> > > > > >>> /coreml/README.md>
> > > > > >>> >> 2. Keras 1.2.2 Support: https://github.com/dmlc/keras/
> > > > > >>> wiki/Installation <https://github.com/dmlc/
> > keras/wiki/Installation
> > > >
> > > > > >>> >>
> > > > > >>> >>
> > > > > >>> >>
> > > > > >>> >> Please make sure you TEST before you vote accordingly:
> > > > > >>> >>
> > > > > >>> >> +1 = approve
> > > > > >>> >>
> > > > > >>> >> +0 = no opinion
> > > > > >>> >>
> > > > > >>> >> -1 = disapprove (provide reason)
> > > > > >>>
> > > > > >>
> > > > > >>
> > > > > >
> > > > >
> > > >
> > >
> >
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message