myrkur (Amir Masoud Ahmadi)

BiRefNet State Of The Art Newest Very Best Background Batch Remover APP

Official repo : https://github.com/ZhengPeng7/BiRefNet

Download APP and installers from : https://www.patreon.com/posts/109913645

Hugging Face Demo : ZhengPeng7/BiRefNet_demo

I have developed a very advanced Gradio APP for this with full proper file saving and batch processing. Also my version removes BG and saves as transparent background.

The APP uses huge VRAM for high resolution images. However it is still working uber fast even though using shared VRAM. So make sure that you have high RAM or set virtual RAM.

Click below to see how to set virtual RAM on Windows.
https://www.windowscentral.com/how-change-virtual-memory-size-windows-10

On Massed Compute A6000 GPU (31 cents per hour) you can very fast remove even very high res images backgrounds.

Currently we have 1 click installers for RunPod, Massed Compute, Kaggle and Windows.

Windows Requirements
Python 3.10, FFmpeg, Cuda 11.8, C++ tools and Git

If it doesn't work make sure to below tutorial and install everything exactly as shown in this below tutorial

https://youtu.be/-NjNy7afOQ0

How To Use On Windows
Just extract files into like c:/BiRefNet_v1

Double click Windows_Install.bat file and it will generate a isolated virtual environment and install requirements

It will automatically download models into your Hugging Face cache (best model under 1 GB)

Then start and use the Gradio APP with Windows_Start_App.bat

Cloud How To Use
Massed Compute, RunPod has instructions txt files. Follow them

Kaggle has all the instructions 1 by 1

On Kaggle set resolution 1024x1024 or you will get out of memory error

Reacted to louisbrulenaudet's post with 👍 4 months ago

Post

2757

🚀 RAGoon is now available on PyPI, GitHub, and as a Space on Hugging Face for batched embeddings generation 🤗

RAGoon is a set of NLP utilities for multi-model embedding production, high-dimensional vector visualization, and aims to improve language model performance by providing contextually relevant information through search-based querying, web scraping and data augmentation techniques.

At this stage, 5 major classes are available via RAGoon to facilitate:
- the production of chain embeddings for several models to simplify a continuous deployment process;
- production of LLM requests for web querying and content retrieval via the Google API;
- recursive chunking via tokens;
- data visualization and the function to load embeddings from a FAISS index, reduce their dimensionality using PCA and/or t-SNE, and visualize them in an interactive 3D graph;
- the creation of binary indexes for search with scalar (int8) rescoring.

Link to GitHub: https://github.com/louisbrulenaudet/ragoon
Link to the 🤗 Space: louisbrulenaudet/ragoon

Reacted to singhsidhukuldeep's post with 👍 4 months ago

Post

754

Looks like @Google is still not satisfied with Gemini 1.5 Pro! 😲

Good folks at @GoogleDeepMind quietly updated the already good Gemini 1.5 Pro to Gemini-1.5-Pro-Experiment-0801 🚀

Unremarkable naming aside, the model itself outperforms GPT-4o, Claude-3.5, and LLama 3.1 on LMSYS and the Vision Leaderboard. 🌟

Gemini-1.5-Pro-Experiment-0801 is great at almost everything, multi-lingual tasks, Maths, understanding, and coding. 🌐📚💻

Although in my testing, I felt Claude-3.5 was slightly better at coding! 👨‍💻🤔

Also, still cannot find an LLM that can solve the "Strawberry prompt"! 🍓❓

"""
How many R's are there in Strawberry?
Also, write Strawberry with all r's in brackets
"""

Try here: https://aistudio.google.com/app/prompts/new_chat