fritzvd

    def cokriging_in_r(self, x, y, z):
        '''
        Cokriging (and ordinary kriging) is quite fast in R.
        This would anyway be more pragmatic than rewriting/porting it to Python.
        For the moment this will be the 'best' way as R makes it very easy to
        use kriging without fitting a variogram model, but using a standard 
        variogram.
        '''
        import rpy2
        import rpy2.robjects as robj

kevincolten

GitHub Pages is great for building a personal or project website. It'll default to http://username.github.io, or you can even use your own custom domain name from services like Namecheap, which I will write about in another post soon.

So you set up your GitHub Page for yourself or project, but what if you want to show off some of your other projects you are working on. You can go the poor man's route, and simply just copy everything from another repository into a folder in your username.github.io repository, commit and push the changes to GitHub, and you'll be able to navigate to http://username.github.io/project.

But this comes with some seriously inconvienent and non-conventional drawbacks, such as duplication of code across repositories and manually copy/paste-ing everytime changes are made to update your hosted codebase. Good luck maintaining that.

The other

brunoventura

// vim: ft=javascript:
/*jslint sloppy: true, vars: true, white: true, nomen: true, browser: true  */
/*global SpreadsheetApp, UiApp, UrlFetchApp, Utilities */
/*
 * A script to automate requesting data from an external url that outputs CSV data.
 *
 * Adapted from the Google Analytics Report Automation (magic) script.
 * @author [email protected] (Nick Mihailovski)
 * @author [email protected] (Ian Lewis)
*/

abridgland

jbants

# ----------------------------------------------------------------------
# Remote Sensing Image Classification Workflow for Landsat data with soft
# voting on a SVM and Gradient Boosting classifier. Outlier in the 
# training data are flagged through an Isolation Forest algorithm. 
# Feature Selection is done by a Recursive Feature Elimination method.
# The results are classification and classification probability raster 
# images in TIF format.
#
# Written by Dimo Dimov, MapTailor, 2017
# ----------------------------------------------------------------------