MSCS Thesis

Thesis code for credit toward masters degree in computer science from Carnegie Mellon University.

Code

• toy.py: implements / demonstrates distribution -> distribution regression for data drawn from distribution functions of a single variable
• dtdr.py: implements scalable distribution -> distribution regression
• manage_files.py: tools for scaling, binning, and otherwise manipulating large data files
• math_helpers.py: mathematical helper functions for use in distribution-to-distribution regression
• pll_helpers.py: tools for parallelizing coefficient computation
• my_parser.cpp: custom parser for extracting particles from simulation binaries

Requirements

This project requires the following scientific computing libraries:

• numpy
• scipy
• sklearn
• pylab (if making plots)

Acknowledgements

My work was supervised by:

• Shirley Ho (Assistant Professor, Physics) - shirleyh@andrew.cmu.edu
• Jeff Schneider (Associate Research Professor, Robotics Institute / School of Computer Science) - jeff.schneider@cs.cmu.edu

Simulations were provided by:

• Hy Trac (Assistant Professor, Physics) - hytrac@cmu.edu