# Simple Physics CAM6 Codebase for Training Machine Learning Algorithms

## Creator

Garrett Limon (glimon@umich.edu)

## Funding

NSF GRFP

## Research and Description

This is a collection of Python and NCL scripts intended to process data, train
machine learning algorithms to emulate physical parameterization schemes,
as well as to analyze the results. See Associated Work for detailed information
and findings.

## Format

Directory is organized as follows.
simplePhysicsML/data_process/ 
      - contains data processing scripts and functions.
      	We use input data from the Community Atmosphere
	Model (CAM) for these tasks; see Associated Work(2)
	for data initially used for this work.
simplePhysicsML/19/
	- contains directories for each case studied (HS,TJ,TJBM), each case
	containing the fields (PTTEND, PTEQ, PRECL, etc) that were emulated
	with machine learning, and each field directory containing a sub-
	directory for each machine learning method used (random forests and
	neural networks). This final directory in this sub-structure is
	associated with the training of our machine leraning emulators using
	Python scripts.
simplePhysicsML/analysis
	- contains python and NCL scripts that analyze and plot relevant fields
	and metrics for us to understand the skillfullness of our emulators.

## Data Citation

Limon, G. C. (2023) Simple Physics CAM6 Codebase for Training Machine Learning Algorithms [Data set], University of Michigan - Deep Blue Data. 

## Associated Work

(1) Limon, G. C., Jablonowski, C. (2022) Probing the Skill of Random Forest Emulators for Physical Parameterizations via a Hierarchy of Simple CAM6 Configurations [Preprint], ESSOAr. https://10.1002/essoar.10512353.1

(2) Limon, G. C. (2022) Simple Physics CAM6 Dataset for Training Machine Learning Algorithms [Data set], University of Michigan - Deep Blue Data. https://doi.org/10.7302/r6v3-s064