[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Status tracking repo on salsa for deep learning team's work



Hi folks,

I realized that people are frequently asking or
continuously wondering the latest packaging status
on packages such as tensorflow (as well as its dependencies).
So I created a repo on salsa under deep learning team

https://salsa.debian.org/deeplearning-team/team-roadmap

The work of deep learning involves three parts: (1) deep
learning framework (tensorflow, pytorch, etc);
(2) hardware acceleration solutions (cuda, rocm, sycl, etc);
(3) debian infrastructure (no concrete plan yet).

I have already written down what I know there, including
the status about pytorch. My impression on bazel/tensorflow
packaging is definitely out-of-date so I shall leave
it to other deep learning team members and the bazel team.

I will update the repo with pytorch updates. How do you
think of this way to synchronize our recognition on the
status? I guess posting updates there will also be
acceptable to you?


Reply to: