After the Twitter space Q&A @logankilpatrick hosted yesterday on “The future of machine learning and why it looks a lot like Julia,” I thought it would be useful to accumulate some community responses to a few questions about the current state of machine learning in Julia: Where does ML in Julia r…

[image] patrick-kidger: (F) No built-in/ubiquitous way to catch passing arrays of the wrong shape; a very common error. (At least that I know of.) Perhaps SizedArray from StaticArrays.jl may be useful here. On one hand, you are encoding the size of the array in the type. Thus array size error…

This thread proves to be an interesting read so far. I’m more of a silent observer here and don’t have any insightful comments on the technical details… but I’d like to point out certain implicit assumptions which have framed the discussion so far, and add my two cents. The discussion so far has la…

This thread inspired me to try getting my hands dirty with ML in Julia and stream myself implementing MNIST classification from scratch in Julia as a noob. Intended as a sort of user experience report a la “Don’t Make Me Think”. Video: shitty ML livestream 1: MNIST classification in Julia - YouTube …

There are some very interesting points brought up in this thread, and several points we are aware of as developers of the ecosystem. It is good that we are separating the concerns for “conventional” DL and SciML, since the challenges in the two fields are very different and can sometimes be at odds …

We ( metalenz.com ) use Julia for differentiable physics, which we take the step further to providing an engineer-friendly interface to model specifications for solving and optimizing PDEs. Think COMSOL or Simulink, with gradients for physical model parameters applied to user-defined objectives and co…

I found a similar thread from 2019: [image] State of deep learning in Julia Machine Learning Although Julia is promoted as an excellent language for deep learning, I still don’t see any framework I could use in production or even in long-term research. Here ar…

There are tons of pretrained models with PyTorch. It makes it very easy to deploy models. I really love Julia. But it is easier to think that I will use Julia for analysis on a Notebook or some simulation, than a NLP Model with BERT that I may have later to deploy on Mobile. On a recent project, I…

[image] Thibaut: But it is easier to think that I will use Julia for analysis on a Notebook or some simulation, than a NLP Model with BERT that I may have later to deploy on Mobile. You might want to track the progress on ONNX.jl - when it’s ready, you should be able to load any of the thousa…

To me that’s just a reframing of the same problem. If you’re restricted to a smaller set of centralized blessed ops, then you might as well use pytorch. I ideally there would be a way to typecheck functions or have an API that enforces semantics which are more easily optimizable so that an ecosyste…

[image] Akatz: If you’re restricted to a smaller set of centralized blessed ops, then you might as well use pytorch. Having a list of operators with certain guarantees doesn’t restrict you in any way. Take mutation for example. Most AD frameworks either don’t favor or explicitly don’t support…

State of machine learning in Julia

Specific Domains Machine Learning

jzr January 13, 2022, 9:58pm 39

I found a similar thread from 2019:

Topic		Replies	Views
Where does Julia provide the biggest benefits over other ML frameworks for research? Machine Learning	34	10792	September 16, 2019
Deep learning in Julia Machine Learning	35	12885	April 22, 2024
Is it a good time for a PyTorch developer to move to Julia? If so, Flux? Knet? Machine Learning	52	25619	January 11, 2021
Why is Python, not Julia, still used for most state-of-the-art AI research? Offtopic knet , flux , machine-learning , mlj , sciml	65	11849	August 14, 2021
On Machine Learning and Programming Languages Machine Learning	48	9009	January 25, 2018

State of machine learning in Julia

Related topics