PyGRF

PyGRF: An improved Python Geographical Random Forest (GRF) model.

Installation:

PyGRF can be installed from PyPI:

$ pip install PyGRF

Example:

Below shows an example on how to fit a PyGRF model and use it to make predictions.

from PyGRF import PyGRF
from sklearn.model_selection import train_test_split

# Split your data into training and test sets
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)

#Create a PyGRF model by specifying hyperparameters
pygrf_example = PyGRF.PyGRFBuilder(n_estimators=60, max_features=1, band_width=39, train_weighted=True, predict_weighted=True, bootstrap=False,
                          resampled=True, random_state=42)

#Fit the created PyGRF model based on training data and their spatial coordinates
#xy_coord is the two-dimensional coordinates of training samples						  
pygrf_example.fit(X_train, y_train, xy_coord)

#Make predictions for testing data using the fitted PyGRF model and you specified local model weight 
predict_combined, predict_global, predict_local = pygrf_example.predict(X_test, coords_test, local_weight=0.46)

Parameters:

If you want to learn more about the major parameters in this package, please refer to the Description of Parameters.

Reference:

Kai Sun, Ryan Zhenqi Zhou, Jiyeon Kim, and Yingjie Hu. 2024. PyGRF: An improved Python Geographical Random Forest model and case studies in public health and natural disasters. Transactions in GIS.