https://github.com/matthias-wright/sarcos
Four different regression models applied to the SARCOS dataset.
https://github.com/matthias-wright/sarcos
gaussian-processes linear-regression machine-learning nearest-neighbors random-forest
Last synced: 6 months ago
JSON representation
Four different regression models applied to the SARCOS dataset.
- Host: GitHub
- URL: https://github.com/matthias-wright/sarcos
- Owner: matthias-wright
- Created: 2019-06-14T16:59:25.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2019-06-21T17:51:02.000Z (over 6 years ago)
- Last Synced: 2025-02-07T18:53:49.074Z (8 months ago)
- Topics: gaussian-processes, linear-regression, machine-learning, nearest-neighbors, random-forest
- Language: Python
- Homepage:
- Size: 3.78 MB
- Stars: 2
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# SARCOS regression
## Description
I applied four different regression models; Nearest neighbor, Linear regression, Regression forests, and Gaussian Processes. The data originates from the [SARCOS dataset](http://gaussianprocess.org/gpml/data/), which consists of 21 features (arising from seven joint positions, seven joint velocities, and seven joint accelerations). The goal is to estimate the first of the seven corresponding joint torques.
## Instructions
In order to execute a model, run main.py file.
## Results
| | Training error [RMSE] | Test error [RMSE] |
| :--- | :---: | :---: |
| Nearest Neighbors | - | 3.3386 |
| Linear Regression | 5.5527 | 5.4600 |
| Regression forests | 4.1356 | 4.6956 |
| Gaussian Processes | - | 5.4386 |
## Dependencies
* [Python (Anaconda 3.6.5)](https://anaconda.org/)
* [NumPy (1.15.4)](http://www.numpy.org/)
* [Matplotlib (3.0.2)](https://matplotlib.org/)
## License
This project is licensed under the [MIT Licence](https://choosealicense.com/licenses/mit/)