Please don't push them to 'master' and... in fact, any time you do work on this, work on a branch. Even if it works and you merge it right in after. 'master' should /always/ be functioning code. Branch, code, merge. It keeps things self consistant.
So, the way you name a branch is this: initials.descriptive and then you can work on it. Don't fork the project, it makes it harder for me to merge your improvements back into the mainline.
To make a branch to work on a feature, I (Ryan) will type: git branch rd.laserdetect
Automate filtering, zeroing, and interactive fitting. Call with a list of filenames, all steps shoudl proceed automatically. System limitations may preclude analysis of more than approximately 50 files at once.
Data filtering for the CSULA quantum-dot spectroscopy experiment
Basic noise-reject algorythm:
Some experimentation shows minimal noise rejection until threshhold is below 0.5 and/or delta grows over 4. Suggest values of 0.10 and 10.
New options file format. New, more powerful command line syntax. Actually writes the default config file! Doesn't do globbing, so windows users will need to specify full filnemaes, not just *.dat
Options:
-
Read more filenames from STDIN--optname
shorthand for --optname=1--optname=value
set the option named 'optname' to 'value'--
all things after this are treated as filenames.`Order isn't important unless you have some wierd filenames that look like options. Options on the command line over-ride those in the defaults file.
Options list:
Valid column names are: "none", "ev", "lambda", "corrected", "intensity", and "temp". Valid Laser Names are "red" and "green" No quotes please.
--laser=red is short for --laserlow=7000 --laserhigh=7250
--laser=green is short for --laserlow=10300 --laserhigh=11600
--laser=foo will override high/low settings.
NOTE: This needs a rewrite.
Fit many data files at once. When fit is acceptable, data is saved to a file with a keypress.
MultiIPF instructions: Generally called by MCP. On command line, call multiIPF.sh with filenames. Within Matlab, multiIPF requires specially prepared input. MultiIPF uses 3 parameters: The first is the name of a file that is a list of all the files you want to analyze (which must be in the same directory as multiIPF) The files must include extensions and must be line delimited. The second two parameters are the columns in the data files that you wish to analyze. These are columns 1 and 2. All the files that are to be analyzed should be sorted, in eV, and no header columns. There is a catch statement that will correct the header columns, but this has not been tested.
multiIPF('FileList.dat', 1, 2)
The program will run ipf on each file. The normal window will appear, asking you to set the number of peaks, baseline, etc. Proceed as usual, but after fitting, press = to save your fit results to NewStuff.dat. The name of the file will be saved before your peaks. Note that it will be appended, so you should clear NewStuff.dat beforehand. mcp.sh automatically does this. Also, if there is an error with the file that should be analyzed, multiIPF will catch it and will report that in NewStuff.dat: 'There was an error with this file' and move on.
Caution with =: Also note that pressing = the very first time without fitting anything will cause an error: = reads out the FitResults to NewStuff.dat, so if there is no fit, an error will be thrown. A catch statement has been implemented but not tested. If = is pressed accidentally twice after a successful fit, the same list of peaks will be appended again.
After finishing all the files, the program will exit, and the peak results will be in NewStuff.dat.
Provides options to Filter. example.filterrc2 should be copied into your home directory under the name .filterrc2 Change username to your own name for logging purposes. See logs in QD/data
Subtracts a baseline from data based on a fit to a background run. Use baseline.sh if possible. To use directly Open MatLAB and type: zero('filename') filename should be in the same directory as the zero.m unless pathed This program will output filename_corrected.dat
Creates zeroed versions of files passed to it.
Spreadsheet that creates data similar to our experiments. The centroids, height, and width factors of the peaks are adjustable on the Controls tab. The NoiseEx tab contains some functions for different kinds of noise. They can be summed up in the left column and the graph and first tab automatically update. To export data switch to the Out tab, Save As csv, and check the box to adjust options. The text delimiter should be " and the separator should be space or tab.