amkkma

11 ms·

61.
by amkkma 5y ago
cool!
62.
The GPGPU developer experience has a long way to go(youtube.com)
10 points by amkkma 5y ago|1 comments
63.
by amkkma 5y ago
in what is your phd?
64.
MIT Introduction to Computational Thinking(computationalthinking.mit.edu)
2 points by amkkma 5y ago|0 comments
65.
by amkkma 5y ago
They wrecked their lives in the course of a crime, as criminals do when they get caught.
66.
by amkkma 5y ago
I want to validate the complaints about startup and error messages as a Julia user / fanboy ;) They don't really bother me anymore personally as I've gotten used to them and make frequent use of packagcompiler.jl However, I&#
67.
Julia PR: Separate codegen/LLVM from runtime for static compilation(github.com)
3 points by amkkma 5y ago|0 comments
68.
Decision Making Under Uncertainty with POMDPs.jl(juliaacademy.com)
87 points by amkkma 5y ago|9 comments
69.
by amkkma 5y ago
That's because it's keeping things fairly simple. Implementing the more involved/advanced techniques would quickly outpace what's possible in python, now and for the foreseeable future (see https://www.stochas
70.
Julia Computing funded by DARPA to Build AI-Based Digital Phased Arrays w/ GPUs(juliacomputing.com)
17 points by amkkma 5y ago|3 comments
71.
A brief History of Julia's C++ REPL and interop [pdf](compiler-research.org)
3 points by amkkma 5y ago|3 comments
72.
AbstractDifferentiation.jl for AD-back end agnostic code(frankschae.github.io)
1 points by amkkma 5y ago|0 comments
73.
by amkkma 5y ago
As a julia user, thanks for this! Inspiring and packed with pearls. There's a lot we can learn from the python community
74.
by amkkma 5y ago
yes Flux doesn't ship GPU drivers. It ships everything else (like CUDA toolkit etc) as needed, using the artifact / pkg system, for all mainstream OSes. Doesn't interfere with system libraries. https://julialang.
75.
by amkkma 5y ago
Sure! NASA: https://www.youtube.com/watch?v=tQpqsmwlfY0 Moderna: https://pumas.ai/ https://discourse.julialang.org/t/has-moderna-used-pumas-ai-... There are many many more. These uniq
76.
by amkkma 5y ago
Some specific steps that will push it past jax/pytorch for chunky array heavy GPU code (can already beat or meet openblas/MKL for kernels written in scalar form). 1. Better compile time memory management ( https://github
77.
by amkkma 5y ago
>it has severe performance problems It had. It's now around parity with pytorch. And no, it wasn't about a usability tradeoff. It was about being more general- More general compiler, more general code, more composable code. Th
78.
Bpfnative.jl: eBPF/Linux kernel programming in Julia(youtube.com)
2 points by amkkma 5y ago|0 comments
79.
State of Julia and Roadmap(youtube.com)
4 points by amkkma 5y ago|0 comments
80.
by amkkma 5y ago
Sounds like expanding this into a blog post would be helpful
81.
by amkkma 5y ago
Thanks
82.
by amkkma 5y ago
Only quibble i have is that Numba is actually an Just AOT method at a time JIT, last I checked in 2018...just much more rudimentary and limited than Julia.
83.
by amkkma 5y ago
Can someone from the core team chime in on the general thinking around the type system/trait stuff? Most of the rest of the complaints are actively being addressed or have been thoroughly discussed, which leaves me unconcerned. Haven
84.
by amkkma 5y ago
They've been pretty mum on the type system stuff though. Do you have any insight there?
85.
by amkkma 5y ago
Point 4 is going to get explicit dev time after the initial stages of the new general purpose compiler plugin infra is completed. Then, proper debugging and introspection tools will be easier to build.
86.
by amkkma 5y ago
>There's a complete and total disregard to any notion of a smaller Julia or a Julia more suited for general purpose That's false. General purpose Julia is a stated goal of the core team, it's just that numerical domination
87.
by amkkma 5y ago
I disagree that it cannot be obviated in Julia. Can be solved with one or more of static tooling (ie proving there's no overwriting), tiered compilation with a faster interpreter and sealed modules. All things jeff et al have been dis
88.
by amkkma 5y ago
I don't have a comprehensive suite of numbers at the moment (which is why I qualified it with "I believe"). My tentative conclusion is based on experience by a flux maintainer benchmarking forwards and backwards passes compa
89.
by amkkma 5y ago
"By not being generic the builtin CL numeric functions run very fast on the most common number types; that was the tradeoff benefit of not designing them be generic." Julia shows this tradeoff isn't necessary.
90.
by amkkma 5y ago
Thanks so much for taking the time to outline your thoughts...I share the same goals and input from industry experience like this very valuable. This has spawned some discussion on the Julia slack about how best to target your usecase. Can

More ›