Sign in

Repository Index

47,133 repositories indexed

Tags: All Tags
Tags

A repository shows when it has any selected Tag.

Filters (1)

Filters

Clear all filters

625–648 of 1,466 · Most stars

Clean PyTorch implementations of imitation and reward learning algorithms

  • reward-learning
  • inverse-reinforcement-learning
  • imitation-learning
  • gymnasium

Interpretability and explainability of data and machine learning models

  • explainable-ai
  • explainable-ml
  • trusted-ai
  • trusted-ml

Official SGLang x Datawhale course on LLM inference: understand inference, build a mini-sglang from scratch, then read the real SGLang source and land your first PR. Available in English and Chinese.

A set of documented functions for simulating the performance of photovoltaic energy systems.

  • solar-energy
  • python
  • renewable-energy
  • renewables

A deep learning toolkit for healthcare AI: EHR (MIMIC, eICU, OMOP), clinical text, signals and imaging. 30+ models, reproducible pipelines.

1M param OCR model - runs on CPU, 15x faster than tesseract, close to 1B+ param models in accuracy

TranslatedDSH × Blender direct-connect plugin v3.0.6: 15 model tools that let AI drive the GUI Blender (viewport / custom viewpoints / headless worker / job layer / transaction rollback / recipe library), with a contract layer of 34 families / 228 ops; installed via npm + addon zip, GUI first, headless as backup; dsh-TUI selected chip / @ mention / /blender, a modeling skill bundled with the package, supports WSL / Windows / macOS.

This repository will be used to develop a rain generator, actually it use white noises to simulate wind and have an integrated thunderstorm simulator.

Turn any LLM into a Jev-style decision model: typed decisions, real probabilities, no training. (continue updating, welcome any issue and PR request)

  • calibration
  • decision-model
  • jev
  • jev-model

Source-available agent platform for Global × China enterprises — wire every system through one agent core. Self-hosted, any LLM.

TranslatedFail-Closed account environment manager: 11-item environment fingerprint audit (Block-on-Fail), a WebRTC leak probe with a hand-written STUN protocol stack, Kill-Switch emergency network cutoff, multi-account snapshot isolation, configuration drift self-healing — making it physically impossible for the real IP to leak