Research with Dr. Hoover and Dr. Caudle at SDSMT by Timothy Ford.
Find me here: timothywford.com
- README.md - this
- requirements.txt - pip venv requirements, needed to run jupyter notebooks (maybe help to come on site if I feel like it or someone wants it)
- abstract.odt - My abstract for the presentation
Holds all the original Tensor Operations we were given from Matlab. Someone else's work, attribution to them is buried in those files at the header I believe.
- 22matrix.m - test 2x2 matrix in matlab
- /matlab - all of the matlab functions
- epidemic models.pdf - paper given to go off of (someone elses' work)
- Social Network Generation.ipynb/pdf - my attempt at some network generation
Work on the matrix version of an ARMA model.
- generateData.py - generate data for testing a MAR (don't remember if this was useful or not)
- model.py - a fairly built out object oriented MAR, but it's not complete and doesn't work
- /coloringCommunities - an tsvd project where the two distinct communities were colored
- /firstRun - first attempt at doing a tsvd with two distinct communities
- /matlab - matlab files used for testing
- .ipynb files - Various iterations of getting results
- L_svd.py and L_trans.py - code stolen from Riley I believe (github: https://github.com/VintageDesign)
- tensorOps.py - the tensor operations converted from matlab to python (everything in here should work!)
Initial exploration/learning in time forecasting using an ARMA model. Essentially building one from scratch, using maximum likelihood.