This repository contains source code for simulation experiments on models, specifically developed for muscle models, but it not limited to those class models. It also contains scripts to perform these experiments on high performance computer with the slurm scheduler.
In order to run many of the optimization methods and sensitivity anaylsis the following libabry needs to be in the search path of MatLab. https://www.safetoolbox.info/
In order to run the genetic algrothim the following Matlab package needs to be installed. https://www.mathworks.com/products/optimization.html
In order to run many of programs devoloped for a HPC you will need this tool box installed.
This directory contains scripts to run a variantly of differnet methods to find the optimal parameter sets and the optimal training set (data set which is used during optimization). The optimization methods includes a large scale random sampling expirment, a genetic algorthim, and the build in Matlab optimizer fmnincon.
Script examples to run the built in MatLab optimizer.
Scripts to run a genetic algrothim for optimization on a slurm super computer.
Scripts to a run a large scale random sampling expirment on a slurm HPC
Scripts to run an optimization / testing expirment to see which dataset is ideal for training the reset of the data sets.
Contains scripts to run different types of senstivity anaylsis including one at a time and a varaince based approach. Please not many of these scripts require the Sensitivity Anaylsis For Everyone (SAFE) tool kit. Find link in the dependencies.
Contains functions for basic statistices for evaluating model results. These functions are used during optimization and single model evaluations.
Contains a varity of different helper functions which accomplish common tasks throughout the kit.
Contains different functions for different visualization techinques to evaluate a model run. This includes a post-flight anaylsis which looks at indiviaul components of the model.