I am excited to see something like this, but this website would inspire way more confidence if there was an "about us" page describing who/what/why.

On the other hand, some of the courses are written by very prominent people in the Julia community, so there is some minimal amount of legitimacy already present.

Yes, we'll put it up shortly. It is basically put together by Julia Computing, and we expect to recruit more content now that we have a basic set of materials.

Why Kent instead of flux?

I’ve never tried Julia, but looking at the sidelines it always seemed that flux is what makes Julia special for neural networks.

The creator of Knet is a professor and he teaches a class using Knet, so he must have tons of material available. If you look at the Knet docs compared to the Flux docs, the Knet docs are much more educational and explains things a lot better.

The first-class auto-differentiation seems to me like the special bit in Julia for machine learning (not only neural networks). Flux/Zygote is only one of the tools for that (but it is indeed what I view as most promising at the moment). Do check out cassette as well. However, for the restricted set of autodiff supported by libraries like Tensorflow, basically any of the Julia tools are good enough.

I'll probably get downvoted to oblivion as punishment for reporting but I got this error when going to https://www.juliaacademy.com/:

Error 1014 CNAME Cross-User Banned What happened? You've requested a page on a website that is part of the Cloudflare network. The host is configured as a CNAME across accounts on Cloudflare, which is prohibited by security policy.

Sorry about that. I've let the team know and they'll be taking a look shortly once folks are awake. Not sure what happened here, the site was up fine until now.

Not sure why you expect downvoting. Thanks for the warning.

Ive seen web site down reports get downvoted and criticised. Even basic questions relevant to a article can be downvoted or responded to in blatant hostility.

Just a set of patterns Ive seen. I rarely post now.

Thanks for reporting. JuliaAcademy is back up again.

Still see it.

I always think of ML https://en.m.wikipedia.org/wiki/ML_(programming_language) when people abbreviate Machine Learning to ML :(

I find that it's easiest to think "machine learning". It's easy to tell when people mean the programming language from context.

For one, at least IME, it's rare these days for someone to use ML to mean a specific programming language. They'll say SML or ocaml. Just ML ends up getting couched into phrases like "the ML family of languages".

To be fair, I only hear about ML the language when someone post in the comments complaining about a post about machine learning.

It's like when I see NLP I have to try and contextualise between natural language programming, natural language processing, neurolinguistic programming and network layer protocol.

I thought it was nonlinear programming.

I too used to, but have realized that almost never is the language referred to as “ML”. It would rather be SML or another dialect name.

The one context I've seen ML show up as such is when referring to the family as a whole, rather than a specific dialect.

I must admit I had the same reaction to this post, and thought that calling it Julia Academy was odd if they were going to teach other languages.

SML is a ML but ML is it's own pre-SML language too.

Right, but one that died before what I'm guessing is the average birth date of hacker news readers. It's probably at least as hard to find a pre-SML ML compiler that will run on a microcomputer as it is to find a SML compiler that can understand the code examples from any SML book written before 1997.

Probably, i've tried once or twice to find a pre-SML ML compiler, but not had any luck so far.

You'd be hard-pressed. Pre-SML, it mostly only saw the light of day as a the metalanguage of a theorem prover named LCF. Which I think ran on a PDP-11 or a VAX or something like that.

https://apps.dtic.mil/dtic/tr/fulltext/u2/785072.pdf

I always think of Maximum Likelihood.

Same

Please don’t call it “free” if I have to “enroll” and hand over information in order to watch the videos. They’re not free. They cost me information/privacy.

Yeah, next up they will be requiring an iris and fingerprint scan to authenticate you into the system for watching the videos...read Viral’s book about it.

uggg... if you are so concerned just give them fake info...

These look like great introductions, will there be more advanced courses available at some point? I especially love Flux but I've already written some more advanced networks in it like ResNets to test drive Julia.

Open source by any chance? :) I've been hunting for some examples of more modern detection networks written in Flux - particularly CenterNet. Seems like the only built-in examples they have are toys though.

I'm facing Error 1014 CNAME Cross-User Banned error.

It's fixed now.

Are these certified courses or audit only?

At the moment they are audit only. We would love to explore possibilities around certification and what people would find helpful.

you can get to it by https://juliaacademy.com/