10 Chinese cities · 16 years of data · 5-year forecasts · Real public data
An open-source urban economic observatory. Auto-collects public data, runs forecasts, generates reports. Updates annually on January 15th — for as long as the internet exists.
Started as a graduation project. Built to last forever.
# Docker (recommended)
docker compose up
# → http://localhost:8000/dashboard
# → http://localhost:8000/docs
# Native Python
pip install -r requirements.txt
uvicorn backend.api.main:app --reload- Competitiveness Index — Entropy-weighted scoring across 9 dimensions (capital, technology, institutions, etc.) with radar charts and heatmaps
- Time-Series Forecasting — ARIMA / GARCH / Theta forecasts with confidence intervals
- City Comparison — Side-by-side analysis across 19+ economic indicators
- Enterprise Analysis — Cost, supply chain, and policy environment scoring
- Dynamic Dashboard — Full ECharts SPA, zero build step
Data Collection ──→ Forecast Engine ──→ FastAPI ──→ ECharts Dashboard
│
GitHub Actions
(annual auto-update)
Beijing · Shanghai · Shenzhen · Guangzhou · Chengdu · Wuhan · Hangzhou · Nanjing · Suzhou · Xi'an
Core indicators: GDP / GDP growth / Population / Fiscal revenue / R&D intensity
City data stored in data/cities/ as JSON files, organized by year.
| Layer | Tools |
|---|---|
| Backend | FastAPI (Python 3.11+) |
| Forecasting | statsforecast / statsmodels / arch |
| Data Science | pandas / numpy / scikit-learn |
| Frontend | ECharts 5 (vanilla JS SPA, zero build) |
| CI/CD | GitHub Actions + Pages |
| Container | Docker / docker compose |
urban-pulse/
├── backend/
│ ├── api/ # REST API routes
│ ├── core/ # Forecast engine, risk engine, multi-city
│ ├── analysis/ # Enterprise, government, economic analysis
│ ├── analytics/ # Competitiveness index engine
│ ├── data/ # Hardcoded city data snapshots
│ ├── data_collection/ # Public data scrapers
│ └── data_processing/ # Cleaner, transformer, validator
├── frontend/ # ECharts dashboard (standalone SPA)
├── site/ # GitHub Pages showcase page
├── docs/ # Architecture, data pipeline, plugin docs
├── config/ # Application config, per-city/industry profiles
├── tests/ # 116+ tests
├── data/cities/ # City data (JSON)
├── scripts/ # Build & utility scripts
├── Dockerfile
└── pyproject.toml
| Endpoint | Description |
|---|---|
POST /api/v1/index/compute |
Compute competitiveness index (entropy weighting) |
GET /api/v1/index/rankings |
Get rankings by dimension |
POST /api/v1/index/report/{city} |
City competitiveness report |
POST /api/v1/cities/time-series |
Multi-city time-series comparison |
POST /api/v1/forecast/gdp/{city} |
GDP forecast with confidence intervals |
Full API docs at /docs when running.
- Expand to 15-20 cities
- Auto-scrapers for education, environment, infrastructure data
- Multi-city VAR forecasting (economic ripple effects)
- Forecast accuracy tracking vs actuals
- Dataset snapshots for reproducible research
GPL-3.0-or-later
🇨🇳 中文版说明
10 座中国城市 · 16 年数据 · 5 年预测 · 真实公开数据
一个开源的城市经济观测站。自动采集公开数据、跑预测、出报表。每年 1 月 15 号自动更新一次，重新跑一遍全量预测。长期跑着，年复一年。
docker compose up
# → http://localhost:8000/dashboard
# → http://localhost:8000/docs基于熵权法的城市竞争力评价，覆盖成本力、资本力、产业链力、制度力、科技力、规模力、聚集力、区位力、生命健康力 9 个维度，28 个二级指标。
| 端点 | 说明 |
|---|---|
POST /api/v1/index/compute |
计算竞争力指数（熵权法） |
GET /api/v1/index/rankings |
按维度查看排名 |
POST /api/v1/index/report/{city} |
单城市分析报告 |
POST /api/v1/cities/time-series |
多城市时间序列对比 |
POST /api/v1/forecast/gdp/{city} |
GDP 预测（含置信区间） |
FastAPI + ECharts 5 + Docker + GitHub Actions
GPL-3.0-or-later
Built by @badhope. Started as a graduation project, built to last forever.