HNHacker News
TopNewBestAskShowJobs

komuher

322 karma · joined December 26, 2019

submissionscomments
komuher··on Nim 2.0 thoughts
Man nim is such a cool language it just need some more community and better docs to get some hype.
komuher··on Text to Image Generation
I love ur transformer repos, its always joy to re-implement stuff from ur code its easy to read and clean :D
komuher··on Julia 1.6 Highlights
If u cant even read code dont lie xD

    n = length(ARGS) > 0 ? parse(Int, ARGS[1]) : 100
    left = calc(101)  # <------- THIS IS COLD START JITTING CALL
    right = -18.67
    if abs(left - right) > 0.1
        println(stderr, "$(left) != $(right)")
        exit(1)
    end

    notify("Julia (no BLAS)\t$(getpid())")
    t = time()
    results = calc(n)
    elapsed = time() - t
    notify("stop")
komuher··on PyTorch 1.8, with AMD ROCm support
Wait wat, jax and also pytorch is used in a lot more areas then NN's. Jax is even consider to do better in that department in terms on performance then all of julia so wat are u talking about
komuher··on An Introduction to Crystal
nim-lang
komuher··on CDC website built by Deloitte at a cost of $44M is abandoned due to bugs
Sounds like poland and hungary today :)
komuher··on Rust for Windows
Correcting false statement?
komuher··on Rust for Windows
This is the same stuff like julia community is doing. Spam language in every Python/R related topic (D guys target Nim/Rust)
komuher··on Julia adoption keeps climbing
Flax/jax can :)
komuher··on Julia adoption keeps climbing
This is my favourite comment about julia for like last few years +1 on that.

Ecosystem is extremly poor outside very few niches and most of the Deep Learning stuff isn't even faster than python api (+C ofc.) so swaping is just usless if u dont have time to write your own GPU kernals for every new opertaion.

komuher··on Machine Learning: The Great Stagnation
Can u elaborate on that?
komuher··on Quick Overview of Julia language [pdf]
Still same as it was 2 years ago Flux + some other smaller repos.
komuher··on AI Update, Late 2020 – Dumpster Fire
+1 I can fully agree its so salty i cant read it
komuher··on What I wish someone had told me about tensor computation libraries
Not even cloese, jax.jit allow you to compute almost anything using lax.for_loops, lax.cond and other lax and jax contsturts pytorch jit does not allow that its just extra optimization for static pytorch functions.
komuher··on AWS Lambda – Functions with Up to 10 GB of Memory and 6 VCPUs
No GPU so ye its very bad :P
komuher··on AMD Radeon RX 6800 and RX 6800 XT Review
Can they just start investing in software so folks from ml/deep learning field can just get out from nvidia monopoly -.-
komuher··on AMD Radeon RX 6800 and RX 6800 XT Review
If u want to do anything related to more then very basic machine learning then NVIDIA gpu is must have
komuher··on RTX3090 TensorFlow, NAMD and HPCG Performance on Linux
Its without XLA and without cudnn and cuda 11.1 for ga102 lets wait for proper drivers to see full results :P
komuher··on Nim – version 1.2.6 and 1.0.8 released
It was hyped and probably are used in even more places (in my previous work i even wrote a full Julia based parsers etc.) But still i didn't see much advantage of writing in Julia now 2 years and even maybe 1 year ago there were few (before numba and other JIT python stuff was working fine). Right now for me its just l'art pour l'ar you almost never get performance like in C/Nim/RUST and you don't even have much good libs in the community (compared to python thier main rival).
komuher··on Nim – version 1.2.6 and 1.0.8 released
What purpose did you use Julia for? Julia was hyped like 2 years ago but right now i didn't see any new cool lib or even a blog posts and DL libs are lagging hard compared to python DL libs (torch, tf, jax etc.).
komuher··on Python 3.8 Makes me Sad Again
Meh another shit-post for some clout.
komuher··on Swift 5.3 Will Be Supported on Windows and Additional Linux Distributions
Yes they all did
komuher··on Progress on Plasma
arch
komuher··on Building an end-to-end Speech Recognition model in PyTorch
Dunno why (probably dataset) but open source Speech Recognition models are performing very poorly on real world data compared to google speech to text or azure cognitive.
komuher··on Swift: Google’s Bet on Differentiable Programming
Nope
komuher··on Swift: Google’s Bet on Differentiable Programming
Looking forward to S4TF 1.0 lets see how much progress they'll manage to make in next few months without Chris Lattner
komuher··on Jax – Composable transformations of Python and NumPy programs
Sure (dont have twitter yet) but will post it here on hacker news in next week probably :)
komuher··on Jax – Composable transformations of Python and NumPy programs
It is compiled to XLA so should be a lot faster then pure PyTorch but probably will be slower then TVM (https://tvm.apache.org/) i can prepare some benchmarks in next few days if u are interested :)
komuher··on Jax – Composable transformations of Python and NumPy programs
JAX is focusing more on performance and probably will be next big think for google machine learning teams like https://github.com/google/flax (most of the people hate tensorflow) Dunno about zygote future
komuher··on Julia 1.4
Sorry im not native :( I tried my best
← PreviousPage 2 of 3Next →