模型训练与微调
模型微调与训练工具
共 50 个项目
huggingface / diffusers
Hugging Face 推出的 PyTorch 扩散模型库,提供 SOTA 模型,轻松实现图像、视频、音频的生成与转换。
apache / airflow
Apache Airflow:以编程方式编写、调度和监控工作流的平台。
deepspeedai / DeepSpeed
DeepSpeed 是深度学习优化库,让分布式训练和推理简单高效。
huggingface / transformers
统一框架,轻松加载、微调和部署文本、视觉、音频等多模态SOTA预训练模型。
ultralytics / yolov5
YOLOv5 是一个基于 PyTorch 的快速、准确的目标检测模型,支持导出至多种移动端和嵌入式格式。
modelscope / FunASR
工业级语音识别工具包,支持超实时识别、50多种语言、说话人分离、情绪检测及流式服务。
NVIDIA / cutlass
NVIDIA CUTLASS 提供 CUDA 模板和 Python DSL,用于高性能线性代数计算。
tracel-ai / burn
Burn 是一个用 Rust 编写的新一代张量库与深度学习框架,兼顾灵活性、效率和可移植性。
sktime / sktime
统一的时间序列机器学习框架,集成预测、分类、异常检测等功能。
microsoft / onnxruntime
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
optuna / optuna
Optuna 是一个开源的超参数优化框架,能自动化、高效地搜索机器学习模型的最佳参数配置。
Project-MONAI / MONAI
面向医疗影像的深度学习AI工具包
Automattic / mongoose
Mongoose 是 MongoDB 的 Node.js 对象建模工具,提供异步环境下的数据验证与封装。
lance-format / lance
面向多模态AI的开放湖仓格式,性能提升百倍,兼容主流数据工具。
nautechsystems / nautilus_trader
用Rust构建的生产级确定性事件驱动交易引擎,支持多资产与机器学习。
ggml-org / whisper.cpp
Port of OpenAI's Whisper model in C/C++
paperless-ngx / paperless-ngx
一个社区驱动的文档管理系统,支持扫描、索引和归档所有文档。
explosion / spaCy
💫 Industrial-strength Natural Language Processing (NLP) in Python
pyg-team / pytorch_geometric
PyTorch Geometric 是一个基于 PyTorch 的图神经网络库,提供高效的数据处理、丰富的模型和工具。
NVIDIA / Megatron-LM
NVIDIA 开源的用于大规模训练 Transformer 模型的研究框架。
cvat-ai / cvat
Computer Vision Annotation Tool (CVAT) is a leading platform for building high-quality visual datasets for vision AI. It offers open-source, cloud, and enterprise products, as well as labeling services, for image, video, and 3D annotation with AI-assisted labeling, quality assurance, team collaboration, analytics, and developer APIs.
deepchem / deepchem
Democratizing Deep-Learning for Drug Discovery, Quantum Chemistry, Materials Science and Biology
AUTOMATIC1111 / stable-diffusion-webui
Stable Diffusion 的 Web 界面,支持文生图、图生图等多种图像生成功能。
online-ml / river
Python中的在线机器学习库,支持流式数据增量学习与概念漂移检测。
qdrant / qdrant
高性能、大规模向量数据库与搜索引擎,支撑下一代AI应用。
invoke-ai / InvokeAI
InvokeAI是面向专业创作者的稳定扩散模型引擎,提供领先WebUI,支持图像生成与编辑。
BlinkDL / RWKV-LM
RWKV 是一个兼具 RNN 高效与 Transformer 高性能的线性注意力语言模型架构,支持并行训练和无限长上下文。
run-llama / llama_index
LlamaIndex is the leading document agent and OCR platform
EleutherAI / lm-evaluation-harness
A framework for few-shot evaluation of language models.
intel / neural-compressor
英特尔开源神经网络压缩工具,支持SOTA低比特量化与稀疏化,提升大模型部署效率。
openvinotoolkit / nncf
神经网络压缩框架,用于提升OpenVINO推理性能,支持量化、剪枝等压缩技术。
civitai / civitai
Stable Diffusion 模型与资源的社交分享平台
speechbrain / speechbrain
A PyTorch-based Speech Toolkit
volcano-sh / volcano
云原生批处理系统(CNCF 孵化项目),专为 AI、大数据、HPC 等场景设计。
feast-dev / feast
The Open Source Feature Store for AI/ML
snakers4 / silero-vad
Silero VAD: pre-trained enterprise-grade Voice Activity Detector
neuraloperator / neuraloperator
Learning in infinite dimension with neural operators.
OpenBB-finance / OpenBB
Open Data Platform for analysts, quants and AI agents.
apache / casbin
Apache Casbin: an authorization library that supports access control models like ACL, RBAC, ABAC.
microsoft / qlib
微软开源的AI量化投资平台,集成多种机器学习模型和自动化研发工具。
naptha / tesseract.js
纯 JavaScript 实现的 OCR 引擎,支持 100+ 语言,基于 WebAssembly 在浏览器和 Node.js 中运行。
huggingface / pytorch-image-models
The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more
skypilot-org / skypilot
The AI Compute Platform for frontier teams. SkyPilot turns fragmented AI compute into one AI supercomputer, so frontier AI teams build custom intelligence faster.
SharpAI / DeepCamera
Open-Source AI Camera Skills Platform, AI NVR & CCTV Surveillance. Local VLM video analysis with Qwen, DeepSeek, SmolVLM, LLaVA, YOLO26. LLM-powered agentic security camera agent — watches, understands, remembers & guards your home via Telegram, Discord or Slack. Pluggable AI skills. OpenAI, Google, Anthropic or local AI. Runs on Mac Mini & AI PC.
voxel51 / fiftyone
一个用于可视化、分析和优化计算机视觉数据集的开源工具。
torchgeo / torchgeo
TorchGeo: datasets, samplers, transforms, and pre-trained models for geospatial data
NVIDIA / warp
A Python framework for GPU-accelerated simulation, robotics, and machine learning.
sonos / tract
Tiny, no-nonsense, self-contained, Tensorflow and ONNX inference
PINTO0309 / PINTO_model_zoo
A repository for storing models that have been inter-converted between various frameworks. Supported frameworks are TensorFlow, PyTorch, ONNX, OpenVINO, TFJS, TFTRT, TensorFlowLite (Float32/16/INT8), EdgeTPU, CoreML.
vwxyzjn / cleanrl
一个将多种深度强化学习算法以单文件形式清晰实现的开源项目,便于研究与学习。