ModelisationGeochimie

Material for the course Modélisation en Géochimie

November 2023 > !!! Course is ongoing, not all the material is online at the moment !!!

Dependencies

You will need a working installation of Python, please see that with Google. I recommend Anaconda Python, that will provide a fully-featured Python distro with Jupyter Notebooks.

In term of libraries, we will need:

• matplotlib
• scipy
• numpy
• pandas
• emcee
• pymc3
• corner
• uncertainties

Install that on your computer with this command in the terminal (Linux/MacOS, search Google for Windows as I don't know this OS):

'pip install -r requirements.txt'

Running on Binder

Alternatively, the notebooks can be run without installing anything on your computer on the website MyBinder at this address:

https://mybinder.org/v2/gh/charlesll/ModelisationGeochimie/master

Running on Google Colab

You can download the zip of the repository and run it on Google Colab.

You either start a Google Colab notebook and upload your data, or host a copy of the folder on Google Drive and access it via Colab. I recommend doing the latter. In any case, there is a good documentation from google to import data on Colab here: https://colab.research.google.com/notebooks/io.ipynb

Please remember that each time you will start again your Colab Notebook, you will need to install the necessary libraries by typing

'!pip install -r requirements.txt'

in the first cell.

Ressources

https://github.com/valentineap/ComputationalGeoscienceCourse

https://scipy-lectures.org/intro/

https://github.com/jrjohansson/scientific-python-lectures

https://towardsdatascience.com/from-scratch-bayesian-inference-markov-chain-monte-carlo-and-metropolis-hastings-in-python-ef21a29e25a

Contributors

Charles Le Losq, Institut de physique du globe de Paris