https://github.com/amazon-science/goal-gps-ordered-activity-labels
Dataset of GPS trajectories with driving/walking activity labels of delivery agents
https://github.com/amazon-science/goal-gps-ordered-activity-labels
activity-recognition dataset geospatial-data
Last synced: 3 months ago
JSON representation
Dataset of GPS trajectories with driving/walking activity labels of delivery agents
- Host: GitHub
- URL: https://github.com/amazon-science/goal-gps-ordered-activity-labels
- Owner: amazon-science
- License: cc-by-4.0
- Created: 2022-10-10T18:41:48.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2022-10-10T19:22:12.000Z (over 3 years ago)
- Last Synced: 2025-09-09T05:11:53.984Z (9 months ago)
- Topics: activity-recognition, dataset, geospatial-data
- Homepage:
- Size: 1.5 MB
- Stars: 7
- Watchers: 2
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
Awesome Lists containing this project
README
## GPS-ordered Activity Labels
Dataset of GPS trajectories with driving/walking activity labels of delivery agents
If you use this dataset, please cite
```
@inproceedings{facnet,
author = {Emre Eftelioglu and Gil Wolff and Sai Krishna Tejaswi Nimmagadda and Vishal Kumar and Amber Roy Chowdhury},
title = {Deep Classification of Frequently-Changing Activities from GPS Trajectories},
booktitle = {ACM SIGSPATIAL Workshop: International Workshop on Computational Transportation Science (IWCTS)},
year = {2022},
}
```
## Accessing the data
The data is found under `trajectories.zip`. The dataset is organized into 804 separate csv files, each containing a 6-minute chunk of a GPS trace. The `x,y` coordinates are in meters relative to an arbitrary center point, i.e. there is no absolute location information in the trajectories. The `timestamp`s as well are not absolute but relative to an arbitrary date. The sampling frequency is approximately one GPS point per 5 seconds. The activity labels are a binary label, `OnFoot/Driving`. The `OnFoot` activity includes time spent in the vehicle not driving.
## Security
See [CONTRIBUTING](CONTRIBUTING.md#security-issue-notifications) for more information.
## License
This library is licensed under the CC-BY-4.0 License.