Machine Learning in Climate Change - 02/25/2022

Welcome to Machine Learning in Climate Change! The goals of this lesson are introduce you to the basics of ScikitLearn, Keras, and Tensorflow on High Processing Computers within the Python. We're approaching this lesson in two parts: - Part 1: Simple Time Series Prediction Using Long-Short-Term-Memory Techniques, and - Part 2: Using Time Series Prediction on Geospatial Data

Part 1

Part 1 will focus on using a singular time-series of sea surface temperatures (SST) from NOAA buoy data, found here (insert link here). After this, you'll be provided quick scripts to prepare the data for processing, and begin creating your machine learning script. The Python skills and packages can be found here: - Scikit Learn and Keras (for the machine learning) - Matplotlib (to plot your timeseries)

Part 2

Part 2 will expand on the initial time series into a whole set of geospatial information. Instead of using buoy data, we'll be training our model using """"""" model output data. We'll be able to evetually compare the singular-point SST data to the model output, as well as predict entire geographic reigons. The python skills we'll build upon are: - netcdf4 (to process the geospatial images) - Scikit Learn and Keras - Matplotlib

Learning Outcomes:

Select, apply, and evaluate ML models. Describe data, models, and modeling assumptions

CI Tools:

Open OnDemand, GPU computing Scikit-Learn Tensorflow

Prerequsites

• Background in Python and coding in general. Familiarity with Pandas, NETCDF4, and matplotlib is preferred.
• An account on the UH HPC, MANA
• Tensorflow installed on your personal computer if you do not have access to the cluster.

Contributing

We welcome all contributions to improve the lesson! Maintainers will do their best to help you if you have any questions, concerns, or experience any difficulties along the way.

We'd like to ask you to familiarize yourself with our Contribution Guide and have a look at the more detailed guidelines on proper formatting, ways to render the lesson locally, and even how to write new episodes.

Please see the current list of [issues][FIXME] for ideas for contributing to this repository. For making your contribution, we use the GitHub flow, which is nicely explained in the chapter Contributing to a Project in Pro Git by Scott Chacon. Look for the tag . This indicates that the maintainers will welcome a pull request fixing this issue.

Maintainer(s)

Current maintainers of this lesson are

• Peter Sadowski, PhD, UH Manoa Dept of Computer Sciences
• Katie Ackerman, UH Manoa Dept of Atmospheric Sciences
• Mingyue Tang, UH Manoa Dept of Atmospheric Sciences

Authors

A list of contributors to the lesson can be found in AUTHORS

Citation

To cite this lesson, please consult with CITATION