Metadata-Version: 2.1
Name: GPyS
Version: 0.0.3
Summary: Gaussian Process Subspace Prediction
Author-email: Taiwo Adebiyi <taadebi2@cougarnet.uh.edu>
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: GNU General Public License v3 or later (GPLv3+)
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
License-File: LICENSE

This is a prototypical implementation of GPS in the Python programming language. 
For the original research article documenting the method, see the Citation section.

## Citation

- Ruda Zhang, Simon Mak, and David Dunson. Gaussian Process Subspace Prediction for Model Reduction. SIAM Journal on Scientific Computing, 2022. https://epubs.siam.org/doi/10.1137/21M1432739

## Installation

Install the package via pip using the following command:

- pip install GPyS==0.0.3

## Example Use 

After installing the package you can load it via: 

#### For GPS Preprocessor: 
  - from GPyS_preprocessor import Preprocessor
  - Note that only Preprocessor.setup(X) takes in argument X and this must be called first before any other functions
  - The remaining functions merely returns preprocessing quantities of interests

#### For GPS Hyperparameter training:
- from GPyS_LOOCV_error import LOOCV
- Utilize hSSDist(length) method for the objective function computation at a given (default) length scale
- Refer to the working script to see an example computation of optimal lengthscale for GPS. 

#### For GPS Prediction: 
  - from GPyS_prediction import Prediction
  - All the functions can be independently called here. 
  - Also, user can directly call Prediction.GPS_Prediction() to immediately obtain prediction results
