My vote is -1 due to an issue that will impact majority of MXNet users. I like to see the fixing PR <https://github.com/apache/incubator-mxnet/pull/13980> cherry picked to this release to fix the issue.
Problem: Training resnet50_v1 using MXNet Symbolic API and kvstore=device on a single p3.16xlarge instance hit 10~20% drop in training throughput. It also applies to Gluon API and other kvstore settings. Detailed performance numbers can be found in the fixing PR <https://github.com/apache/incubator-mxnet/pull/13980>. Impact: MXNet users that conduct distributed training on GPUs using ImageRecordIter to load data will experience this regression. Solution and Status: A fixing PR <https://github.com/apache/incubator-mxnet/pull/13980> is created and under review on master. Thanks, Yuxi On Thu, Jan 24, 2019 at 10:06 AM Steffen Rochel <steffenroc...@gmail.com> wrote: > Dear MXNet community, please test v1.4.0.rc1 and provide your feedback and > vote. > > Regards, > Steffen > > On Tue, Jan 22, 2019 at 10:40 AM Steffen Rochel <steffenroc...@gmail.com> > wrote: > > > Dear MXNet community, > > > > This is the vote to release Apache MXNet (incubating) version v1.4.0. > > Voting will start today, January 22nd 2019 noon PST and will close on > > Friday, January 25th 9pm PST. > > > > Link to release notes: > > > > > > > https://cwiki.apache.org/confluence/display/MXNET/Apache+MXNet+%28incubating%29+ > > 1.4.0+Release+Notes > > > > Link to release candidate: > > https://github.com/apache/incubator-mxnet/releases/tag/ > > <https://github.com/apache/incubator-mxnet/releases/tag/1.4.0.rc1> > > 1.4.0.rc1 > > <https://dist.apache.org/repos/dist/dev/incubator/mxnet/1.4.0.rc1> > > > > Link to source and signatures on apache dist server: > > https://dist.apache.org/repos/dist/dev/incubator/mxnet/1.4.0.rc1 > > > > > > Please remember to TEST first before voting accordingly: > > +1 = approve > > +0 = no opinion > > -1 = disapprove (provide reason) > > > > > > Best regards, > > Steffen > > > -- Yuxi(Darren) Hu, Ph.D. Software Development Engineer Amazon Web Services
