https://github.com/lihangalex/geofeaturekit
GeoFeatureKit: From coordinates to geospatial features and travel-time accessibility for ML, urban analytics, and location intelligence.
https://github.com/lihangalex/geofeaturekit
accessibility data-science embedding feature-engineering geospatial gis isochrone machine-learning openstreetmap osmnx python urban-analytics urban-planning
Last synced: about 15 hours ago
JSON representation
GeoFeatureKit: From coordinates to geospatial features and travel-time accessibility for ML, urban analytics, and location intelligence.
- Host: GitHub
- URL: https://github.com/lihangalex/geofeaturekit
- Owner: lihangalex
- License: mit
- Created: 2025-07-04T20:56:36.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2025-07-08T17:23:19.000Z (3 months ago)
- Last Synced: 2025-09-20T01:11:19.323Z (15 days ago)
- Topics: accessibility, data-science, embedding, feature-engineering, geospatial, gis, isochrone, machine-learning, openstreetmap, osmnx, python, urban-analytics, urban-planning
- Language: Python
- Homepage: https://pypi.org/project/geofeaturekit/
- Size: 220 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# GeoFeatureKit
[](https://www.python.org/downloads/)
[](https://pypi.org/project/geofeaturekit/)
[](https://opensource.org/licenses/MIT)
**Turn any location into rich urban data. Get POIs, street networks, and accessibility metrics from just coordinates.**
## 🚀 Quick Start
```bash
pip install geofeaturekit
```
### Basic Analysis
```python
from geofeaturekit import extract_features
# Analyze any location (lat, lon, radius_meters)
features = extract_features(40.7580, -73.9855, 500)
# Access the structured data
print(features['poi_metrics']['absolute_counts']['total_points_of_interest']) # 1076
print(features['poi_metrics']['density_metrics']['points_of_interest_per_sqkm']) # 1370.0
print(features['network_metrics']['basic_metrics']['total_intersections']) # 731
```
**Sample Output Structure:**
```json
{
"poi_metrics": {
"absolute_counts": {
"total_points_of_interest": 1076,
"counts_by_category": {
"total_dining_places": {"count": 400, "percentage": 37.17},
"total_retail_places": {"count": 126, "percentage": 11.71},
"total_public_transit_places": {"count": 96, "percentage": 8.92},
"total_healthcare_places": {"count": 13, "percentage": 1.21}
}
},
"density_metrics": {
"points_of_interest_per_sqkm": 1370.0
},
"distribution_metrics": {
"diversity_metrics": {"shannon_diversity_index": 2.147},
"spatial_distribution": {"pattern_interpretation": "random"}
}
},
"network_metrics": {
"basic_metrics": {
"total_intersections": 731,
"total_street_length_meters": 80044.7
},
"connectivity_metrics": {
"average_connections_per_node": {
"value": 5.954,
"confidence_interval_95": {"lower": 5.837, "upper": 6.071}
}
}
}
}
```
### Multi-Modal Accessibility
```python
from geofeaturekit import extract_multimodal_features
# Compare walking vs biking reach
features = extract_multimodal_features(
40.7580, -73.9855,
walk_time_minutes=10,
bike_time_minutes=10
)
walk_pois = features['walk_features']['poi_metrics']['absolute_counts']['total_points_of_interest']
bike_pois = features['bike_features']['poi_metrics']['absolute_counts']['total_points_of_interest']
print(f"🚶 10-min walk: {walk_pois} POIs reachable") # 2104 POIs
print(f"🚴 10-min bike: {bike_pois} POIs reachable") # 12200 POIs
print(f"🎯 Bike advantage: {bike_pois/walk_pois:.1f}x more places") # 5.8x
```
### Batch Processing
```python
from geofeaturekit import extract_features_batch
# Analyze multiple locations at once
locations = [
(40.7580, -73.9855, 500), # Times Square
(40.7829, -73.9654, 500), # Central Park
(40.7527, -73.9772, 500), # Grand Central
]
results = extract_features_batch(locations)
for i, result in enumerate(results):
pois = result['poi_metrics']['absolute_counts']['total_points_of_interest']
restaurants = result['poi_metrics']['absolute_counts']['counts_by_category']['total_dining_places']['count']
print(f"Location {i+1}: {pois} POIs, {restaurants} restaurants")
# Location 1: 1076 POIs, 400 restaurants
# Location 2: 185 POIs, 12 restaurants
# Location 3: 1131 POIs, 323 restaurants
```
### Progress Tracking
```python
def progress_handler(message, progress):
print(f"[{progress:.0%}] {message}")
# Add progress callback for long-running analysis
features = extract_features(
40.7580, -73.9855, 1000,
verbose=True,
progress_callback=progress_handler
)
```
## 📊 What You Get
**POI Analysis:**
- `poi_metrics['absolute_counts']` - Raw counts by category (23 categories)
- `poi_metrics['density_metrics']` - POIs per km², density by category
- `poi_metrics['distribution_metrics']` - Diversity indices, spatial patterns
**Street Network:**
- `network_metrics['basic_metrics']` - Nodes, intersections, street length
- `network_metrics['connectivity_metrics']` - Connections per node, ratios
- `network_metrics['street_pattern_metrics']` - Bearing entropy, grid patterns
**Multi-Modal Features:**
- `walk_features`, `bike_features`, `drive_features` - Same structure as above
- `radius_features` - Circular analysis results
## 🎯 Use Cases
- **Real estate analysis**: Compare neighborhood walkability
- **Urban planning**: Assess transit accessibility
- **Machine learning**: Generate location features for models
- **Market research**: Analyze competitor density
## 🗺️ Roadmap
- **Temporal pattern mining** - Discover how neighborhoods change across time scales
- **Gentrification prediction** - ML models to forecast neighborhood change
- **Causal inference engine** - Identify what actually drives urban outcomes
- **Anomaly detection** - Identify unusual urban patterns and opportunities
*For feature requests, please contact: lihangalex@pm.me*
## 📚 API Reference
```python
# Basic analysis
extract_features(lat, lon, radius_meters, verbose=False, cache=True, progress_callback=None)
# Multi-modal accessibility
extract_multimodal_features(lat, lon, walk_time_minutes=None, bike_time_minutes=None, drive_time_minutes=None, verbose=False, cache=True, progress_callback=None)
# Batch processing
extract_features_batch(locations, verbose=False, cache=True, progress_callback=None)
```
## 🤝 Contributing
Contributions welcome! See [Contributing Guide](CONTRIBUTING.md).
## ⚖️ License
MIT License - see [LICENSE](LICENSE) file for details.
## 🙏 Acknowledgments
Built with [OSMnx](https://github.com/gboeing/osmnx), [NetworkX](https://github.com/networkx/networkx), and [GeoPandas](https://github.com/geopandas/geopandas). Data © [OpenStreetMap](https://www.openstreetmap.org/) contributors.
For complete license information and attributions, see [NOTICE.md](NOTICE.md).
---
**Ready to analyze any location? Start with `