tessierplot

Installation

Go into the project folder and type:

python setup.py install

If you're feeling frisky

python setup.py develop

installs in develop mode and the project folder is where the module lives. Any editing done will immediately carry over.

Usage

tessier view

Make your life easier by making thumbnails of all measurement files and plot them in a nice grid.

import imp
from tessierplot import view
imp.reload(view)

view.tessierView(rootdir='/where/my/naturepublicationmeasurements/are',filterstring='',override=False, showfilenames=True)

As can be seen tessierView takes 4 potential arguments:

• The rootdir is where view begins recursively looking for files matching the filterstring.
• The filterstring is expected to be a regular expression.
• override determines if datafile preview files are replotted, even if they already exist. The default value is False.
• showfilenames determines if the file name of every plotted data file is displayes. The default value is False

plotR

This is the main plotting object, it takes a measurement file as argument, after which a specific command can be given to plot either 2d, 3d , or some other type of plot.

from tessierplot import plot

p = plot.plotR('mymeasurementfilelocation.dat.gz')
p.quickplot() #automagically figures out if it's a 2d or 3d plot and plots accordingly
p.plot2d() #plot in 2d
p.plot3d() #plot in..hey 3d.
p.starplot() #starplot measurement files only, only plot datapoints for each separate axis

uniques_col_str

By supplying a uniques_col_str parameter to a plot command the way the data is segmented can be altered.

E.g. the data consists of 4 data columns. x y z r

with x,y,z coordinates and r a value that needs plotting. For a 3d plot, values y,z are most likely the coordinates with r being the corresponding value needing plotting. Sometimes x is varied slowly taking only a limited amount (n) of unique values. Thus, it would be logical to plot only n 3d plots with the value of x being indicated per plot.

You can manually supply which columns contain these 'unique' coordinates. In this particular example one could supply p.plot3d(uniques_col_str=['x']) or, if y is also pretty sparse p.plot2d(uniques_col_str=['x','y']) .

styles

plotR support several default ''styles'' that can be applied to the data to make your life a bit easier.

e.g.

p.plot3d(style=['mov_avg(n=1,m=2)','didv'])

applies a moving average filtering, and a subsequent derivative. Other filters at this time are:

STYLE_SPECS = {
	'deinterlace': {'param_order': []},
	'deinterlace0': {'param_order': []},
	'deinterlace1': {'param_order': []},
	'didv': {'param_order': []},
	'log': {'param_order': []},
	'normal': {'param_order': []},
	'flipaxes': {'param_order': []},
	'flipyaxis': {'param_order': []},
	'flipxaxis': {'param_order': []},
# 	'threshold_offset': {'threshold':0.2,'start':0.0,'stop':1.0,'param_order':[]},
	'mov_avg': {'m': 1, 'n': 5, 'win': None, 'param_order': ['m', 'n', 'win']},
	'abs': {'param_order': []},
	'savgol': {'samples': 11, 'order': 3, 'param_order': ['samples', 'order']},
	'sgdidv': {'samples': 11, 'order': 3, 'param_order': ['samples', 'order']},
	'fancylog': {'cmin': None, 'cmax': None, 'param_order': ['cmin', 'cmax']},
	'minsubtract': {'param_order': []},
	'crosscorr': {'peakmin':None,'peakmax':None,'toFirstColumn':True,'param_order': ['peakmin','peakmax','toFirstColumn']},
	'massage': {'param_order': []},
	'deint_cross': {'param_order': []}
}

The massage style

you can also custom make a style without modifying the tessierplot module.

Supplying the plot command with a massage_func=special_style. Will cause special_style to be called and you can process the data inline.

def special_style(wrapper):
	#do some fancy manipulation of your data
	wrapper['X'] #xdata in 2d, empty for 3d
	wrapper['XX'] #ydata/zdata, depending on 2d or 3d

	return wrapper
p.quickplot(style=['mov_avg', 'massage', 'didv'], massage_func=special_style)

the colorbar

The colorbar supports modifying the colormap nonlinearly by clicking in it. This will divide the colormap in n+1 segments, where n is the number of marks. Each mark can be dragged.

e.g. for 2 marks, the modifying points will be at 1/3 and 2/3 of the colormap. Dragging a mark will effectively drag these points to an arbitrary new point, scaling the colormap in the process.

fiddle, linedraw, and linecut

By clicking one of the three buttons in the 3d plot window it's possible to modify and extract more info from a plot.

Fiddle changes the colormap range by clicking and dragging in the figure once fiddle has been enabled. Look to the colorbar for the effect. This feature is useful to bring out features that are maybe otherwise hidden.

With linedraw enabled, you can draw a line, and observe its length and slope.

Linecut gives a linecut of the data. For a horizontal linecut, hold the alt-key.