RT @Voxel51: In Part 1, @datascienceharp will walk through:. 👉 Why standard vision models fail catastrophically on GUI tasks.👉 The annotati….

Check out the dataset shown here: Here's the LeRobot dataset importer for FiftyOne: Listen to the podcast epiosode here:

I was literally inspired to action listening to.@Redpoint 's Unsupervised Learning podcast . They had the guys from @physical_int @hausman_k.talking about the need for tools to understand massive robotics datasets. They wanted to spot labeling errors across millions of samples,

RT @Voxel51: Here’s a peek at what’s on the agenda. ➡️ “Foundational models for generalist computer agents” – Raghav Kapoor, @Adobe . ➡️….

Labeling data has got to be the most time consuming task I’ve ever done in deep learning.

Oh yeah, here is the codebase:

Register here:. Session 1: Session 2: Session 3:

I'm teaching a series of workshops in August that go deep into Visual Agents (specifically GUI Agents). The last session is fine-tuning a VLM for this task. A huge challenge is building a dataset for fine-tuning. I want annotations in COCO Format, but nothing exists for this,

Older. 3.1 on a 386.

Will have to integrate this into FiftyOne asap!.

Had an awesome time visiting @huggingface office in Paris! Thank you @mervenoyann for the invite. Good you finally meet you and the legend @reach_vb in person. Looking forward to the next time. Cheers!

So pumped to hear @NielsRogge speak at the Brussels meetup!

Packed house for @tuanacelik’s talk about RAG with @llama_index at the Amsterdam meetup!

Does no one remember what happened with Inflection?.

RT @Voxel51: The era of single-purpose vision models is ending. Introducing agglomerative vision models: a fundamental shift from how we’ve….

RT @Voxel51: If you're in Germany, come out and join us for a hands-on computer vision workshop on Advanced Car Damage Detection with Fifty….

Resources:. ⭐️ the repo on GitHub: 👨🏽‍💻 Notebook to get started:

OS Atlas 7B is a solid vision model that will localize UI elements reliably, even when you deviate from their suggested prompts. Here's what I learned after two days of experimentation ⇣. 1) OS Atlas 7B reliably localizes UI elements even with prompt variations. • The model

Notebooks:. Natural images: OCR: Star the repo on GitHub:

Two days with Nemotron Nano VL taught me that it can spot a left leg in a crowd but can't find a button on a screen. It's surprisingly capable at natural images but completely breaks on UI tasks. Here are my main takeaways. 1. It's surprisingly good at natural images, despite