I do ML. Write open-source. occasionally win data-science competitions. Ukrainian. Kaggle, Signate and dark sarcasm grandmaster. #armukrainenow, #cancelrussia
I improved NanoGPT Speedrun from 39.9s → 6.7s (-83.2%), more than every record since 2025 combined (-81.4%)!
The web is full of rare, repeated content that the LM can’t memorize in time. I solve this with retrieval-augmented n-gram hints🧵
@dr_hb_ai Yea. My teammates used it. For it was the biggest surprise that zero-shot prompt allowed astra to work in agentic mode and “ask” for special slices to confirm a condition. I gave it a fixed set of slices abd instruction that it can ask for extra slices and it worked really well
@devellloper HM Mirra 2
12 років політ нормальний. Сітка ціла, нічо ніде не лізе. Хоча може бути трохи завузька по стегнам якщо пан не пропускає день ніг в залі.
@JFPuget@gonedarragh@kaggle I’m very impatient and while I may run one set of hyperparameter sweep i always tempted to shove one or two extra changes (tune two knobs at the same time). With CC and autoresearch I found it’s easier to plot a path where autoresearch show go and forget about it for a few days.
@wightmanr Not sure how data IO intensive runs you are going to have, but I went for RAID 0 configuration for a pair of NVMEs on my 4x3090 workstation (double capacity, double speed; reliability solved with prayers)
Great of all sorts of 3d scans/video processing
Hi, I'm Ivan - a totally-not-suspicious aerospace prodigy.
I just happen to have a charming Russian accent, a past internship at Sukhoi (no big deal), and a mild obsession with drone jet propulsion systems.
@ducha_aiki Any chance you have dynamic size of your inputs/targets? The way pin memory works is allocating a buffer on device memory where it would be copied. As long you have another size a new buffer must be allocated which over time leads to fragmentation of memory and even OOM
Just secured 🥈2nd place in the BirdCLEF+ 2025 Kaggle competition with Fernando Gonçalves!
2,025 teams, tons of experiments, and a lot of fun.
Huge thanks to the organizers and community!
Write-up: https://t.co/7CPNnL6Py9
#kaggle#deeplearning#ecoacoustics#audio