No description, website, or topics provided.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
parallel-gaussian-mean
readme

readme

README file for parallelized Gaussian Mean Change Point Detection

Contact (unless otherwise noted): nsong@alumni.princeton.edu
(C) Nancy Song, Princeton University

Overview: this program executes a parallel version of change point detection
of changes in the Gaussian mean. Note that for fluorescence spectroscopy,
input are the intensities at each time point, not photon counts.

Requirements: GNU C Library 2.12.2+, OpenMPI 1.4.2+

Instructions:
(1)	Build Parallel-GM program using the makefile
(2)	Run Parallel-GM using the following arguments: filename, alpha, beta
	(a) alpha is the Type-I error
	(b) beta is the confidence region. Only accepts values from {0.69, 0.90, 0.95, 0.99}
(3)	Output is a file with the name of the file appended with .cp that has the time index of the 
	change point detected and the left and right confidence intervals

Example:	The following runs Gaussian mean change point detection on the file file1 with two processes
		and a Type-I error of 0.05 and a confidence interval of 0.95

		Type into command terminal:
		mpirun -np 2 Parallel-GM filename 0.05 0.95

		Output file would be file1.cp
		Sample from file1.cp:	
		99 97 104

		This indicates that a change point was found at the 99th time step. 
		Confidence region is from index 97 to 104