really like whoever's idea was to make this a challenge like this was, viral potential and being actually useful :)
2,304 karma · joined April 6, 2025
kuber.studio x.com/kuberwastaken linkedin.com/in/kubermehta/
Feel free to reach out for a chat!
really like whoever's idea was to make this a challenge like this was, viral potential and being actually useful :)
Also, I do remember seeing you on the leaderboard, cool stuff!!!
Thanks for the comment! Means a lot :)
Yes, it is a big JSON blob atm, haha and t's definitely still a POC, but the idea is to avoid having a separate JSON file that adds to the complexity. While EXIF data works pretty well for most basic stuff, it's not enough for everything one might need for AI specific stuff, especially for things like attention maps and saliency regions.
I'm currently working on redundancy and error correction to deal with the resizing problem. Having a separate file format, even if it's a headache and adds another one to the list (well, another cute-sounding one at least), gives more customization options and makes it easier to associate the properties directly.
There's definitely a ton of work left to do, but I see a lot of potential in something like this (also, nice username)
I'm working on redundancy and error correction to make it better!
> Maybe your format could have some use, but I don't find your motivation convincing.
That's respectable and it very much still is a POC, I hope to keep working on it to be actually great :)
It’s definitely not finished, more like a poc right now for storing richer, AI-relevant metadata in a portable way. Appreciate you for taking the time to check it out.
>what's an "attention map"? with maximum charity, if we had some sort of attention-as-in-LLM-like structure precached, how would it apply beyond one model? By “attention map” I meant a visual representation of where a model can focuses its “attention” when analyzing an image — basically, a heatmap highlighting important regions that influence the model’s output. It isn't something that is very useful now but might be.
> You reinvented EXIF/JPEG metadata with naivete Partly true (at least for now) the core idea was to experiment with alternative metadata or feature embedding, not to replace well-established standards. It's not where I NEED it to be yet but as far as metadata usecases go, it's pretty cool.
> Your setup instructions use UNIX commands up until they require running a Windows batch file It's easier to set windows up to directly open other file formats, it's just a thing (and I'm on windows - so)
also, the idea is to integrate the conversion processes/ pipelines with other data that'll help with customized workflows.
TREAT could only become what it has due to Hugging Face Spaces and this was indeed a satisfying adventure going back and trying to capture at least a spirit of TREAT and content moderation through open models - Llama 3.2 and Flan T-5 Large
That’s what I faced already when an upcoming first year AI and Data Science student at Indraprastha University. Every single exam season, students frustrated by juggling between ad-filled websites, Google Drive windows, and WhatsApp groups just to find decent notes before exams, I honestly had enough and decided to change that.
I’m building ENGRAM, a Free to use, Open Source, ad-free platform that brings together study materials from across the community into one clean (some might say- AWESOME LOOKING), easy-to-use space. It already pulls from major existing platforms (with due credit) but there’s room for so much more.
The real issue isn’t just access, it’s sustainability. Most student-made sites break down once their creators graduate or get busy. Notes become outdated. Interfaces get clunky. Nobody’s around to fix things. That’s why I chose open source from the start. Anyone can contribute, update materials, suggest features, or improve the codebase directly on GitHub. It’s a platform that grows with its community.
This started as a side project out of necessity. But it’s also a case study in how open-source thinking can solve overlooked, everyday problems even outside traditional dev spaces.
If you're from IPU, ENGRAM is made for you. If you're from anywhere else, it’s made for you too, it's important to recreate it's spirit so that it's not just great for you but also the students who come after you.
GitHub Repo: https://github.com/Kuberwastaken/engram
Yeah, no
Update: Fixed Uncompletable maps too