This work is part of a submission to Unearthed’s ExploreSA Gawler Challenge. The source code is posted at this link: Gawler Public Repository The ExploreSA Gawler Challenge is a contest to predict areas of potential mineralization in the Gawler region of South Australia. In this post I will lay some more of the groundwork for a computational approach to this mineral favorability mapping challenge. The SARIG data package contains a whole series of CSVs, sarig_md_*_exp.

This work is part of a submission to Unearthed’s ExploreSA Gawler Challenge. The source code is posted at this link: Gawler Public Repository The ExploreSA Gawler Challenge is a contest to predict areas of potential mineralization in the Gawler region of South Australia. In this post I will lay some of the groundwork for a computational approach to this mineral favorability mapping challenge. We’ll begin by loading some of the geochemistry data from the SARIG data package into Dask and extracting a smaller set of soil geochemistry records.

kRiging (Kriging in R) As a programmer, it is often frustrating to solve equations by hand, knowing that a clever implementation in software could solve an infinite variety of similar problems. This cannot be said for the paper-and-pen approach. As a <<Mastère spécialisé>> student in the Cycle de Formation Spécialisée en Géostatistique at Mines Paristech, I have spent the better part of these past few weeks building and solving kriging systems.

Hugo Tips: Part 2 I only have 3 tips this time, but they are important: Always always always remember to run env HUGO_ENV="production" hugo (or the equivalent on your system) rather than just plain hugo. It is possible that running hugo will write the files in development mode, which puts NOCRAWL, NOINDEX tags on every page. These tags will prevent search engines from crawling your site. For Windows, it should be set HUGO_ENV=production.

This is a continuation of my first post on the SEG Facies Classification competition. In the previous post I left off before implementing a baseline learner to see what results we might expect. Baseline Learner After Witten, Frank, and Hall’s Data Mining, I use depth-1 decision stumps and shallow decision trees as baseline learners. Decision trees are easily interpretable once plotted. They also give us a sense of feature importance, since the greedy algorithm will split the more informative features earlier in the tree.