Beast code in Giters

Guozhi Xu's repositories

alpaca_chinese_dataset

人工精调的中文对话数据集和一段chatglm的微调代码

Language:Jupyter NotebookMIT000

alter-sdk

Alter SDK lets you easily pipe avatars with motion capture into your game, app, or website. It just works.

Language:Objective-CNOASSERTION010

BLOOM-LORA

Due to restriction of LLaMA, we try to reimplement BLOOM-LoRA (much less restricted BLOOM license here https://huggingface.co/spaces/bigscience/license) using Alpaca-LoRA and Alpaca_data_cleaned.json

Apache-2.0000

CDBA

An Blender addon uses ROMP to extract human's 3D poses from image, video or webcam and drive your own 3D character.

Language:PythonMIT000

ChatGLM-chinese-insturct

探索中文instruct数据在ChatGLM, LLaMA上的微调表现

Language:Python000

ChatGLM-RLHF

对ChatGLM直接使用RLHF提升或降低目标输出概率|Modify ChatGLM output with only RLHF

Language:PythonAGPL-3.0000

chatglm_finetuning

chatglm 6b finetuning and alpaca finetuning

Language:Python000

Chatgpt-Custom

定制化chatgpt：多种超低成本定制化的chatgpt平替方案，带模型和数据且完整训练预测流程，可在线使用(仅限用于学术测试)[已有ChatGLM的LoRA与pt2方案]

Language:PythonMIT000

EasyMocap

Make human motion capture easier.

Language:PythonNOASSERTION000

FaceDiffuser

000

JoJoGAN-One-Shot-Face-Stylization-with-W-B-and-Gradio

JoJoGAN: One Shot Face Stylization

Language:Jupyter Notebook000

llm_finetuning

Large language Model fintuning

Language:Python000

lmft

Language Model Fine-Tuning, for ChatGLM, BELLE, LLaMA fine-tuning.

Apache-2.0000

mediapipe-prebuilt

Realtime iris tracking demo for iOS using MediaPipe

Language:Objective-C++MIT000

MediapipeFaceMeshIOSLibrary

Building of Mediapipe Facemesh into an easily usable iOS XCFramework

Apache-2.0000

MeFaMo

Language:PythonMIT000

menpo3d

Tools for manipulating 3D meshes within the Menpo project.

NOASSERTION000

metahuman-from-landmarks

Pipeline to convert landmarks (in json format) to metahuman animations in Unreal Engine.

GPL-3.0000

mobile-vision

Mobile vision models and code

Language:PythonNOASSERTION000

mocap4face

A free, multiplatform SDK for real-time facial motion capture using blendshapes, and rigid head pose in 3D space from any RGB camera, photo, or video.

Language:SwiftNOASSERTION000