nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

ArcGIS Pro 3.1.0: What is the python code for the Geographically Weighted Regression (GWR)?

JamalNUMAN

In the screenshot below, I couldn’t figure out how to obtain the python code for the Geographically Weighted Regression (GWR) by which the dependent variable (y) can be expected based on the independent variables (x1, x2, x3, x4) considering the location of each observations where u column represents the x – coordinate and v column represents the y – coordinate

Encuentra más publicaciones etiquetadas con

Accepted answers

All comments

JamalNUMAN

https://developers.arcgis.com/geoanalytics/tools/gwr/

https://github.com/geeksnome/machine-learning-made-easy/blob/master/locally_weighted_regression.py

JamalNUMAN

I have little knowledge in coding. All what I wanted is a python code for the Geographically Weighted Regression (GWR) that can work in Jupyter or Spyder

danretired

spreg isn't an available package

You can do the same if you examine the code I referenced earlier

JamalNUMAN

I used the ChatGPT in order to get the python code for the GWR as per the request below

---

Write a python code to read the table saved here:

"C:\D\H3.xlsx"

Write a python code to predict the value of the dependent variable based on the geographically Weighted Regression (GWR) displaying how the weights of neighbouring observations are used to calculate the prediction of the focal observation

The first 4 columns are the independent variables

The fifth column is the dependent variable

The u column represents the x-coordinate

The v column represents the y coordinate

Use the Gaussian as the kernel function

Assume the bandwidth to be 400 meters

-------

The ChatGPT provided the code below:

--------

import pandas as pd

import numpy as np

from spreg import GWR

from spreg.utils import read_spatialdata

# Load the data from the Excel file

file_path = r'C:\D\H3.xlsx'

data = pd.read_excel(file_path)

# Define the independent variables (X1, X2, X3, X4)

X = data[['X1', 'X2', 'X3', 'X4']].values

# Define the dependent variable (y)

y = data['y'].values

# Define the coordinates (u and v)

u = data['u'].values

v = data['v'].values

# Define the bandwidth for the Gaussian kernel

bandwidth = 400 # meters

# Create a spatial weights matrix based on Gaussian kernel

distances = np.sqrt((u[:, np.newaxis] - u)**2 + (v[:, np.newaxis] - v)**2)

kernel_weights = np.exp(-distances**2 / (2 * bandwidth**2))

# Create a GWR model

model = GWR(y, X, kernel_weights)

# Fit the GWR model

results = model.fit()

# Predict the values of the dependent variable

predicted_values = results.predict(X)

# You can now use predicted_values for your predictions

print(predicted_values)

-------

However, running this code on the JupyterLab ends up with an error as below

danretired

For JupyterLab?

you would have to provide all the input variables into this expression

ARCPY.GeographicallyWeightedRegression_analysis(
                in_features, dependent_field, explanatory_field, out_feature,
                kernel_type, band_width, distance, nn, weight_field, crw,
                anaCellSize, in_pred, pred_field, out_pred)

I don't think that is what you want

JamalNUMAN

Not sure how can I can take the advantage of this python code to stick it in JupyterLab in order to apply the analysis

danretired

C:\... install folder ...\Resources\ArcToolBox\toolboxes\Spatial Statistics Tools.tbx\toolbox.module.py

line 3619 : contains the class GeographicallyWeightedRegression description

line 3864: conts the class GWR

you can follow the various imports fromtthat module

Únete a la Conversación

Los miembros registrados pueden publicar, seguir actualizaciones y más. ¿Nuevo aquí? Regístrate gratis.

Enlaces Rápidos

Community Help
Find useful guides, FAQs, and documents to help you navigate and make the most of Esri Community.