Python SDK for Cortexa: download & parse datasets (JSON/YOLO/COCO), list datasets with template-annotation filter, and structured EvalSample output for evaluation pipelines.
Project description
audience: [engineering, business] edit_policy: factual_direct source_of_truth:
- backend/server/settings.py
- backend/server/urls.py
- backend/requirements.txt
- frontend/package.json
- pyproject.toml
- .docsmith/features.yml
- {fact: "仓库结构与开发者入口"}
Cortexa
把原始图片和视频变成可训练数据集:上传素材、标注(人工或 AI 预标注)、审查、打包导出,一条线走完。
- 多种标注类型:矩形框、多边形、3D 立方体、折线、自定义答题模板,覆盖检测/分割/3D/质检场景
- AI 预标注加速:接入 PaddleX 小目标检测、SAM2/SAM3 语义分割、3D-MOOD 立方体检测模型,标注员只需修正
- 一行代码拉数据集:Python SDK 支持 JSON/YOLO/COCO 格式导出,按标签过滤、按模板分组打包
架构
graph TD
Browser["UmiJS + Ant Design 前端<br/>:8001"]
API["Django REST Framework<br/>:8000"]
Worker["Celery Worker"]
Beat["Celery Beat"]
Mongo[(MongoDB)]
MinIO[(MinIO 对象存储)]
RabbitMQ[(RabbitMQ)]
AI["AI 模型服务<br/>(PaddleX / SAM2 / SAM3 / 3D-MOOD)"]
SDK["cortexa-sdk<br/>Python CLI + API"]
Browser -->|HTTP /api/v1/| API
SDK -->|HTTP /api/v1/| API
API --> Mongo
API --> MinIO
API --> RabbitMQ
RabbitMQ --> Worker
Beat -->|定时任务| RabbitMQ
Worker --> Mongo
Worker --> MinIO
Worker --> AI
前置条件
| 依赖 | 版本 | 用途 |
|---|---|---|
| Python | >= 3.10 | 后端运行时 |
| Node.js | >= 18 | 前端构建 |
| pnpm | >= 8 | 前端包管理 |
| MongoDB | >= 6 | 主数据库 |
| MinIO | 最新 | 对象存储(图片/视频/zip) |
| RabbitMQ | >= 3.12 | Celery 消息队列 |
快速启动
# 1. 克隆仓库
git clone <repo-url> && cd cortexa
# 2. 启动后端
cd backend
cp .env.example .env # 填写 MongoDB / MinIO / RabbitMQ 地址
python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
python manage.py runserver 0.0.0.0:8000 # API 服务
# 3. 启动 Celery(另开两个终端)
celery -A server worker -l info # 异步任务处理
celery -A server beat -l info # 定时任务调度
# 4. 启动前端
cd ../frontend
pnpm install && pnpm dev # 开发模式,访问 http://localhost:8001
# 5. 安装 SDK(可选,用于脚本化下载数据集)
pip install cortexa-sdk
cortexa-sdk --init-config # 生成 ~/.cortexa/config.json
仓库结构
| 目录 | 说明 |
|---|---|
backend/ |
Django REST Framework API 服务(Python 3.10 + MongoDB) |
frontend/ |
UmiJS + Ant Design 标注 Web 界面(TypeScript + React) |
cortexa_sdk/ |
用户侧 Python SDK(pip install cortexa-sdk) |
文档导航
| 文档 | 说明 |
|---|---|
PRODUCT.md |
产品能力总览 |
docs/product/feature-map.md |
完整能力矩阵 |
docs/reference/api.md |
HTTP 接口清单 |
docs/reference/config.md |
环境变量配置参考 |
docs/reference/data-model.md |
数据模型 |
docs/reference/glossary.md |
术语表 |
docs/how-to/local-development.md |
本地开发 |
docs/how-to/deploy.md |
部署 |
docs/how-to/troubleshooting.md |
排障 |
docs/explanation/architecture.md |
架构设计 |
Contributing
- 从
main拉分支,命名feat/xxx或fix/xxx - 后端改完跑
pytest,前端改完跑pnpm build - 提 PR 并填写
.github/pull_request_template.md模板
License
Copyright (c) 2025 Vortek Inc. and Tuanliu (Hainan Special Economic Zone) Technology Co., Ltd. All rights reserved.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file cortexa_sdk-0.3.0.tar.gz.
File metadata
- Download URL: cortexa_sdk-0.3.0.tar.gz
- Upload date:
- Size: 16.7 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
8f7e054f0d39b1175a4e695d8f0f98f1f9af7854d1e9766f6ce5b1adb1ea5de4
|
|
| MD5 |
0a5ff6f279089d749b64653deb529b0e
|
|
| BLAKE2b-256 |
25fbb800071b6dc05de4a44d5f61291fe700c2adad11dc50ddc88dd384cd353d
|
File details
Details for the file cortexa_sdk-0.3.0-py3-none-any.whl.
File metadata
- Download URL: cortexa_sdk-0.3.0-py3-none-any.whl
- Upload date:
- Size: 16.8 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
18f6d190d1df6dd2b8032477eec16526e50e3cf617b9e5ee1149217f17fd2499
|
|
| MD5 |
6d9a8835df3c44ce4ec0fc3449789745
|
|
| BLAKE2b-256 |
adeb8f95dbf8b76908a698797648a5ffadc60d4b059e0cfa2039f4bb048b30dc
|