GIS, Fossils and Loads of Data – Friday Faves

This week, we have loads of geoscience-y favourites. Between great Jupyter widgets, augmented reality and a huge new dataset to play with, in the Friday Faves.

Geoscience and Python

Martin Renou gave a fantastic talk about Jupyter Voila at EuroScipy. He tweeted out their newest developments in iPyLeaflet, which brings GIS capabilities to Jupyter.

Once you got that done, you might even build a Dashboard using Voila after. Check out the full Blog post here, it’s awe-inspiring. So many possibilities, I thought should be much harder to do!

Geoscience and Augmented Reality

In this amazing Tweet, the University of Wyoming built a killer application for Paleontology, combining AR and 3D models for education.

What’s more?

You can print them yourself! And get the Apps here:

Machine Learning in Geoscience

GeoProvider (http://geoprovider.no/) has sponsored the preparation of a new dataset of 1240 Norwegian wells was made available CC-BY 4.0.

The dataset includes core porosity, permeability and lithology, available on Google Drive. The data are licensed CC-BY 4.0. The dataset contains 270000 individual core plug measurements, 6500m of depth referenced core images and digitized core log descriptions. A I first heard about a new dataset via Peter Borman on Software Underground.

Brendon Hall has further popularized the dataset:

I can’t wait to read the roll up from the event, to see what people came up with this year.

Machine Learning for Code

Everyone knows, searching for code is at the core of getting software done.

But it turns out, searching code is a really hard problem. So GitHub teamed up with Microsoft Research Cambridge to provide 2 million code-comment pairs (out of 6 million total) to train a new machine learning solution to make coding a bit less of a hassle.

The Blog Post is here: https://github.blog/
The Github Repo is here: https://github.com/github/CodeSearchNet
The Leaderboard is provided by Weights & Biases with the submission instructions here.