S4TF on Archive

123 views
Skip to first unread message

m.b.schret...@gmail.com

unread,
Feb 19, 2021, 9:53:10 AMFeb 19
to Ewa Matejska, Swift for TensorFlow
Hi Ewa,

Thanks a lot for the pleasure of the S4TF ride and all the efforts you
and your team have accomplished. S4TF-Team (and all the contributors)
have shown the potential of the approach. I'm sure that you are facing
a lot of great challenges and bright opportunities in your
environment.

From the beginning on it was made clear, that an own fork will not be
the goal of the project- so the decision, go to archive came not out of
the blue. All the more as the consolidation of the project started last
year in November.

Of course, GOOGLE and APPLE (and f. i. NVIDIA) could have decided
otherwise- but why should they have done it? On the contrary, the
political and technological leadership as well as first tiers and
customer organizations around the world keep a wary eye on any kind of
cooperation or collaboration between the leading giants. In addition
the work you have done, offered a clear route to go to silicon in order
to exploit (and protect) the benefits in embedded systems along a
(potentially huge) product pipeline.

I wish you (and S4TF contributors) all the best in your next projects.

Regards
Manfred

liewan yan

unread,
May 20, 2021, 3:40:39 AMMay 20
to Swift for TensorFlow, m.b.schret...@gmail.com, Swift for TensorFlow

Hi,ALL:
Why did the S4TF project stop? Did it encounter insurmountable difficulties, or did it find its value in doubt?
Reply all
Reply to author
Forward
0 new messages