ARSET Fundamentals of Machine Learning (ML) for Earth Science

Materials for ARSET Fundamentals of Machine Learning for Earth Science. This repository contains materials for Session 1, 2, and 3.

Assignments

The assignments listed for each session are practice assignments with questions that will be included in the final assignment after Session 3 conclusion. The final assignment will be through a Google Form where you will be answering a set of questions from each one of the Sessions.

Session 1 Materials:

Lecture Topic	Interactive Link
ML Algorithms Introduction
Assignment Session 1

Session 2 Materials:

Lecture Topic	Interactive Link
MODIS EDA
MODIS Train & Eval
Assignment Session 2

Session 3 Materials:

Lecture Topic	Interactive Link
MODIS Model Tuning
MODIS Explainability
MODIS AutoML
Assignment Session 3

Additional Resources

The NASA ASTG provides additional introductory materials related to Python and data science in general. You can access some of this interactive material directly from their repository NASA ASTG py_materials or under the links below.

Installing the Anaconda Python Distribution

It is not required to have a Python distribution installed on your local machine. However, we believe that it is important to have one in order to write and run your own Python applications. We recommend that you install the Anaconda Python distribution by following the instructions at: Anconda installation Guide

Installing Git

To install Git on your local machine, follow the installation instructions: Getting Started - Installing Git

To fully follow all the topics below, you need to have a gmail account in order to access Google Colaboratory. Each course will be taught through the Google cloud based Jupyter notebook.

Starting Point

Lecture Topic	Interactive Link
Introduction to Jupyter Notebook
Introduction to Git

Introduction to Python

If you have never been exposed to Python, you need to take this Introduction to Python course. In case you did some Python programming in the past and you want to assess your Python knowledge, take the following test (in less that 15 minutes and without using any help):

Python Assessment Test

If you score at least 80% then only take the I/O on Text Files topic. Otherwise, take the entire course.

Lecture Topic	Interactive Link
Running Python
Data Types
Conditional Statements
Loops
Advanced Data Types
Functions
Modules
I/O on Text Files

Lecture Topic	Interactive Link
Introduction to Turtle
A place to run the code	https://repl.it/

Data Science Tools

Lecture Topic	Interactive Link
Introduction to Numpy
Basic Visualization with Matplotlib
Introduction to Pandas

Additional References

• awesome-earth-artificial-intelligence
• awesome-open-geoscience
• segment-anything-geospatial