Group Abstract Group Abstract

Message Boards Message Boards

11
|
21.3K Views
|
15 Replies
|
51 Total Likes
View groups...
Share
Share this post:

About MxNet as Mathematica backend choice

Posted 4 years ago
POSTED BY: Yaroslav Bulatov
15 Replies
Posted 2 years ago
POSTED BY: Jack Hu

I agree with the original poster. Offering MXNet alternatives needs to be a very high priority. Perhaps doing so would help with the significant problem of Target->"GPU" that at least of version 13.3 did not work well on Macs when training neural nets.

POSTED BY: Seth Chandler

We are well aware of the MXNet situation and have been planning a backend switch. It is a very large project that's currently in its early research state, and at the moment we don't yet have a time estimate for its completion.

Posted 1 year ago

deleted

POSTED BY: Josh H

What machine learning framework (or frameworks) are of interest to you (as a possible complete replacement for MXNet)?

POSTED BY: Arnoud Buzing

Mathematica should be integrated with PyTorch instead of MxNet. It has by far the most community support.

The other candidates could be TensorFlow and Jax, but I would not recommend their integration in 2024. (I worked both on TensorFlow and PyTorch development teams)

POSTED BY: Yaroslav Bulatov

I personally find the JAX model compelling but I never seriously worked with it. I am courious: why you would not reccomend it's inclusion? Thanks!

POSTED BY: Yaroslav Bulatov
POSTED BY: Arnoud Buzing

Mathematica's flirting with Python and PyTorch, huh? I have to say, my past Python escapades felt like a wild goose chase through a maze of dependencies and deprecations. But hey, if Mathematica can tame that beast and keep the chaos at bay, I'm all for it! I'm so smitten with Mathematica that the thought of not having to play peek-a-boo with Python again fills me with glee. I hope this new project lets me cozy up even more in my Mathematica comfort zone – it’s like my favorite armchair that I never want to leave.

POSTED BY: James Linton
Posted 2 years ago
POSTED BY: Asim Ansari
POSTED BY: Yaroslav Bulatov
Posted 2 years ago

Regarding JAX, you may be aware of Aesara ( https://github.com/aesara-devs/aesara ).

POSTED BY: Asim Ansari
Posted 2 years ago

I recently found that the MXNET was retired as of Sep. 2023 and it is no longer actively developed as indicated from its website and Wikipedia. Mathematica uses MxNet as the backend for neural networks.

POSTED BY: Sangdon Lee

Building on Yaroslav Bulatov's post:

Closely related is support for things like NeuralODEs (backpropogating through NDSolve). This is another increasingly important area in scientific ML, and is now built into Julia (https://julialang.org/blog/2019/01/fluxdiffeq/) and PyTorch (https://towardsdatascience.com/neural-odes-with-pytorch-lightning-and-torchdyn-87ca4a7c6ffd) . One can kind of "hack it" by implementing ODE integrators as neural nets in Mathematica (I've done it, as have some others).

POSTED BY: Joshua Schrier
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard