happysalada Raphael Megzari

Compression AGSI middleware using brotli

Component orchestration engine

Random Name and Slug Generator

APScheduler for Django

SDK and CLI for parsing PDF, DOCX, HTML, and more, to a unified document representation for powering downstream workflows such as gen AI applications

PyTorch implementation of EfficientDet

Universal Tensor Operations in Einstein-Inspired Notation for Python

Functional programming for Python

PyTorch extensions for high performance and large scale training

Facebook AI Research Sequence-to-Sequence Toolkit

Fast, Accurate, Lightweight Python library to make State of the Art Embedding

More interactive weak supervision with FlyingSquid

Open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena

Collection of common code that's shared among different research projects in FAIR computer vision team

Channel Level Live Debug Information Service for gRPC

Standard Health Checking Service for gRPC

Standard Protobuf Reflection Service for gRPC

LLM orchestration framework to build customizable, production-ready LLM applications

Package acting as a wrapper around the headless mode of existing web browsers to generate images from URLs and from HTML+CSS strings or files

Python library that provides common I/O interface across different storage backends

Jinja2 extension to use humanize library inside jinja2 templates

Standalone module for Locality-Aware NMS

Unified toolkit for Deep Learning Based Document Image Analysis

Utility package to create lazy modules, deferring associated imports until attribute access

Use any LLM as a drop in replacement for gpt-3.5-turbo. Use Azure, OpenAI, Cohere, Anthropic, Ollama, VLLM, Sagemaker, HuggingFace, Replicate (100+ LLMs)

Lucene query parser generating ElasticSearch queries

Python module to read, parse and converting Microsoft Outlook MSG E-Mail files

Dead-simple tool for executing Jupyter notebooks from the command line

ONNXMLTools enables conversion of models to ONNX

Transformers Model Optimization Tool of ONNXRuntime

Paddle numpy bfloat16 package

Convert PDF to DOCX

Plumb a PDF for detailed information about each char, rectangle, line, et cetera — and easily extract text and tables

Pinecone Python SDK

Polygon is a python package that handles polygonal shapes in 2D

Module for interacting with PostHog

Simple client to fetch prompts from Prompt Hub using its REST API

Python 3 Bindings for the NVIDIA Management Library

Python SDK for Milvus

This module defines extensions to the standard “typing” module that are supported by the Pyre typechecker

Pytest plugin for parallel and concurrent testing

Create Open XML PowerPoint documents in Python

Python client for Qdrant vector search engine

Library for unit extraction - fork of quantulum for python3

Collection of BM25 Algorithms in Python

Hassle-free computation of shareable, comparable, and reproducible BLEU, chrF, and TER scores

System for quickly generating training data with weak supervision

Python implementations of various stemming algorithms

Fast BPE tokeniser for use with OpenAI's models

Easily trim 'messages' arrays for use with GPTs

Triton Python client

Tokenizer for Text to Speech (TTS) models

Semantic search and workflows powered by language models

Open source libraries and APIs to build custom preprocessing pipelines for labeling, training, or production machine learning pipelines

Hosted model inference code for layout parsing models

Python native client for easy interaction with a Weaviate instance

Sub-second search & analytics engine on cloud storage

Create full-fledged APIs for static datasets without writing a single line of code

TUS protocol implementation in Rust

Decentralized Operating System for Zero-Knowledge Applications

Awesome SurrealDB migration tool, with a user-friendly CLI and a versatile Rust library that enables seamless integration into any project

Tailwind CSS Language Server

Tus 1.0.0 client protocol implementation for bash

Open-source toolkit designed to make it easy to prepare unstructured data like PDFs, HTML and Word Documents for downstream data science tasks

Whisper command line client compatible with original OpenAI client based on CTranslate2

Open-source developer platform to turn scripts into workflows and UIs

Tool to be used as a custom build command for a Cloudflare Workers `workers-rs` project