Re: Issue with releases / feedback from ASF board
Hi Justin, Here's an update on the progress of addressing the license issues: Done: - Add disclaimer to repo.mxnet.io and dist.mxnet.io to clarify that the nightly releases there are not intended for public consumption. (changed. pending cache refresh) - Remove non-Apache releases in github release page. Ongoing: - Delete problematic Maven releases. dev@ is notified and per discussion it's pending lazy consensus [1]. - Review with Apache Trademark on the current third-party binary distributions of mxnet and make necessary correction. Review initiated with trademarks@. - Review with Apache Legal on the appropriate license for convenience binary distribution and display it prominently. Currently waiting for reply [2] To do: - Add source distribution to PyPI package `apache-mxnet`. This is intended to be the official source release that is compliant with incubator distribution guidelines [3]. - Non-official third-party Maven binary releases. We need input on the requirements for observing the proper trademark usage first. As we proceed, there may be more work to do, and we are tracking the progress in https://github.com/apache/incubator-mxnet/issues/18397. Let us know if you have any question. Regards, Sheng [1] https://lists.apache.org/thread.html/ra4e9572ac74857a80c64a31e8bf292d353e74cfa87bf457f47450303%40%3Cdev.mxnet.apache.org%3E [2] https://issues.apache.org/jira/browse/LEGAL-515 [3] https://cwiki.apache.org/confluence/display/INCUBATOR/DistributionGuidelines On Sun, May 31, 2020 at 2:01 AM Justin Mclean wrote: > Hi, > > I'm writing my board report in the next couple of days and just wondering > what progress has been made on this. > > Thanks, > Justin > > - > To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org > For additional commands, e-mail: general-h...@incubator.apache.org > >
Re: [NOTIFICATION]: Jenkins CI restarted
Thank you Joe for getting the CI back into a working state. Do you have any insights into what went wrong to get into the dysfunctional state? On Mon, 2020-06-01 at 12:17 -0700, Joe Evans wrote: > CAUTION: This email originated from outside of the organization. Do not click > links or open attachments unless you can confirm the sender and know the > content is safe. > > > > Hi dev community, > > > The Jenkins CI server had to be restarted due to a high number of jobs > being hung in the queue. If you have open PRs waiting for status checks, > you may have to re-trigger CI checks using mxnet-ci bot. > > > To re-trigger CI checks that have failed, comment on your PR using the > following syntax: > > > @mxnet-bot run ci [] > > > Example: > > @mxnet-bot run ci [unix-gpu] > > > Sorry for the inconvenience. > > > Joe
[NOTIFICATION]: Jenkins CI restarted
Hi dev community, The Jenkins CI server had to be restarted due to a high number of jobs being hung in the queue. If you have open PRs waiting for status checks, you may have to re-trigger CI checks using mxnet-ci bot. To re-trigger CI checks that have failed, comment on your PR using the following syntax: @mxnet-bot run ci [] Example: @mxnet-bot run ci [unix-gpu] Sorry for the inconvenience. Joe
Re: Profiler Broken?
FYI, It works with 1.6.0, the output json file was too large for chrome to handle and it silently failed. I reduced the number of iterations to profile to create a smaller file, anything above 500 MB seems to fail. Thanks Sheng for the pointer and Anirudh for the help. On Thu, May 28, 2020 at 1:07 PM Pedro Larroy wrote: > Yes the profiler seems to be broken / has some concurrency issues. I have > seen corrupted profile results. > > On Thu, May 28, 2020 at 12:30 PM Naveen Swamy wrote: > > > I am attempting to profile one of our models, I used the profiler.state > to > > run/stop in code and also used the environment variables to autostart the > > profiler. It creates a 600MB json file, however when I view in chrome > > tracing it comes out to be blank screen (loading seems to be fine, didn't > > get any errors) > > > > Wondering if anyone has recently tried or if aware of profiler being > > broken? > > > > ENVIRON: Ubuntu 18.04 > > MXNet : mxnet-cu101mkl > > Deep Learning AMI (Ubuntu 18.04) Version 29.0 (ami-043f9aeaf108ebc37) > > > > Thanks, Naveen > > >
Re: [apache/incubator-mxnet] [RFC] Custom Operator Part 2 (#17006)
Speed. All those `std::string` and `std::unordered_map` objects don't come cheaply. I compared an integrated fork with a custom operator. https://github.com/kpuatamazon/incubator-mxnet/tree/intgemm integrated version end-to-end Sockeye performance (based on 1.6.0): ``` real2m57.962s user7m3.986s sys 0m6.724s ``` Custom operator version (based on 1.7.x. because it had to be for custom operators): ``` real3m16.879s user7m43.727s sys 0m8.273s ``` Conditions: `unset MXNET_ENGINE_TYPE; export OMP_NUM_THREADS=2; numactl -C 0-7 translate.sh` Both were compiled with the MKL backend hack for the remaining fp32 operations. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/apache/incubator-mxnet/issues/17006#issuecomment-636870342