Bharat Soni

bharatxoni

AI & ML interests

None yet

Recent Activity

upvoted a collection 27 days ago

InternVL2.5

liked a Space about 1 month ago

showlab/ShowUI

liked a model about 1 month ago

jadechoghari/Ferret-UI-Llama8b

View all activity

Organizations

None yet

bharatxoni's activity

upvoted a collection 27 days ago

InternVL2.5

Collection

Better than InternVL 2.0 • 18 items • Updated 6 days ago • 78

liked a Space about 1 month ago

Running on Zero

190

💻

jadechoghari/Ferret-UI-Llama8b

Image-Text-to-Text • Updated Oct 18, 2024 • 444 • 57

liked a Space about 2 months ago

Running on Zero

🏃

JanusFlow 1.3B

Huggingface space for JanusFlow-1.3B

liked a model about 2 months ago

netrunner-exe/Insight-Swap-models-onnx

Updated Oct 18, 2024 • 23

liked a model 3 months ago

allenai/Molmo-72B-0924

Image-Text-to-Text • Updated Oct 10, 2024 • 2.79k • 275

liked 2 Spaces 4 months ago

Running

👕👔👚

IDM VTON PLUS

High-fidelity Virtual Try-on

Running on Zero

1.72k

👕👔👚

IDM VTON

High-fidelity Virtual Try-on

liked a Space 5 months ago

Running on T4

309

🤲

sarvamai/shuka_v1

Updated Oct 16, 2024 • 179 • 45

Sao10K/L3-70B-Euryale-v2.1

Text Generation • Updated Jun 14, 2024 • 365 • 124

liked a dataset 6 months ago

lmsys/lmsys-chat-1m

Viewer • Updated Jul 27, 2024 • 1M • 2.87k • 624

liked a model 6 months ago

Groq/Llama-3-Groq-8B-Tool-Use

Text Generation • Updated Aug 27, 2024 • 676 • 269

reacted to merve's post with 🤗 7 months ago

Post

6054

Fine-tune Florence-2 on any task 🔥

Today we release a notebook and a walkthrough blog on fine-tuning Florence-2 on DocVQA dataset @andito @SkalskiP

Blog: https://huggingface.co/blog 📕
Notebook: https://colab.research.google.com/drive/1hKDrJ5AH_o7I95PtZ9__VlCTNAo1Gjpf?usp=sharing 📖
Florence-2 is a great vision-language model thanks to it's massive dataset and small size!

This model requires conditioning through task prefixes and it's not as generalist, requiring fine-tuning on a new task, such as DocVQA 📝

We have fine-tuned the model on A100 (and one can also use a smaller GPU with smaller batch size) and saw that model picks up new tasks 🥹

See below how it looks like before and after FT 🤩
Play with the demo here andito/Florence-2-DocVQA 🏄‍♀️