From my view, performance is a big plus for MXNet and the reason why lots of
people adopted in MXNet.
I still think we need to have a top-level class for "performance".
Thanks,
--Patric
> -Original Message-
> From: Chen, Ciyong
> Sent: Monday, December 23, 2019 12:08 PM
> To:
Hi Aaron,
Thanks for your valuable feedback.
I'll prepare to contribute this change and PR soon, and update the contents as
suggested.
Regarding making "Performance" a Key Feature to replace with "Tools &
Libraries", anything I need to take care when removing "Tools & Libraries" part?
Thanks!
After some thoughts along the direction, I find a better and fun answer to the
above question: support tuple/ellipsis/slice in tvm ffi effectively.
I quickly hacked up a POC in https://github.com/tqchen/tvm/tree/pyffi that
supports the following benchmark script(disclaimer: it is only a POC so
Dear Administrator,
My name is Bac Trach Dam. I reside in Victoria, Australia and currently doing a
scholarship with Swinburne University of Technology on Machine Learning
Recommender System. I am currently using MXNet to build my neural network and
would greatly appreciate access to the Slack