Model Analysis ToolKit

Software · Fri May 15 00:00:00 EDT 2015

OSTI ID:1232133

Harp, Dylan R.

MATK provides basic functionality to facilitate model analysis within the Python computational environment. Model analysis setup within MATK includes: - define parameters - define observations - define model (python function) - define samplesets (sets of parameter combinations) Currently supported functionality includes: - forward model runs - Latin-Hypercube sampling of parameters - multi-dimensional parameter studies - parallel execution of parameter samples - model calibration using internal Levenberg-Marquardt algorithm - model calibration using lmfit package - model calibration using levmar package - Markov Chain Monte Carlo using pymc package MATK facilitates model analysis using: - scipy - calibration (scipy.optimize) - rpy2 - Python interface to R

• Short Name / Acronym: MATK; 003429MLTPL00
• Site Accession Number: C14023; LA-CC-13-132
• Version: 00
• Programming Language(s): Medium: X; OS: Multi-platform; Compatibility: Multiplatform
• Research Organization: Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
• Sponsoring Organization: USDOE
• DOE Contract Number: AC52-06NA25396
• OSTI ID: 1232133
• Country of Origin: United States