This is pyarbus - eyetracking analysis software written in python.
At the moment, it is very rough and experimental. The intention is to have this be a software package used for analysis of eyetracking data in general, but at the moment, it is focused on data generated by SR Research's Eyelink eye trackers.
This is a for-public-consumption rewrite of previously unreleased internal code, and it may still contain specifics unique to the way my experiments (which would be considered a bug and should be removed). Many docstrings have not yet been updated to reflect the changes introduced by the rewrite.
To start using pyarbus right away, just run:
python setup.py develop
This will make it so that you don't have to reinstall every time you pull in new changes from the git repository. If you just wanted to install the standard way, run:
python setup.py install
- python >= 2.6 or >= 3.2
- numpy
- matplotlib
- nitime >= 0.4
In [23]: import pyarbus
In [24]: from pyarbus.tests import short_test_file
In [25]: el = pyarbus.read_eyelink(short_test_file)
In [26]: el.r # right eye data
Out[26]:
Eye:
TimeArray([ 13899085., 13899086., 13899087., ..., 13908082., 13908083.,
13908084.], time_unit='ms')
{'pupA': masked_array(data = [3875.0 3874.0 3867.0 ..., 1588.0 1590.0 1594.0],
mask = [False False False ..., False False False],
fill_value = 1e+20)
, 'xres': masked_array(data = [24.2 24.2 24.2 ..., 25.6 25.6 25.6],
mask = [False False False ..., False False False],
fill_value = 1e+20)
, 'y': masked_array(data = [327.1 327.2 327.2 ..., 475.7 475.8 477.0],
mask = [False False False ..., False False False],
fill_value = 1e+20)
, 'x': masked_array(data = [365.7 365.6 365.0 ..., 610.4 610.5 610.5],
mask = [False False False ..., False False False],
fill_value = 1e+20)
, 'xv': masked_array(data = [-- -- -- ..., -0.4 -0.5 -0.5],
mask = [ True True True ..., False False False],
fill_value = 1e+20)
, 'yres': masked_array(data = [25.5 25.5 25.5 ..., 27.7 27.7 27.7],
mask = [False False False ..., False False False],
fill_value = 1e+20)
, 'yv': masked_array(data = [-- -- -- ..., 0.7 0.7 0.9],
mask = [ True True True ..., False False False],
fill_value = 1e+20)
}
In [27]: print el.l # this is monocular data, so left eye is None
None
In [28]: el.r.x
Out[28]:
masked_array(data = [365.7 365.6 365.0 ..., 610.4 610.5 610.5],
mask = [False False False ..., False False False],
fill_value = 1e+20)
In [29]: el.r.sacepochs
Out[29]:
Epochs([(13899265.0 ms, 13899287.0 ms), (13899394.0 ms, 13899408.0 ms), (13902078.0 ms, 13902104.0 ms), (13903000.0 ms, 13903021.0 ms), (13903339.0 ms, 13903371.0 ms), (13904212.0 ms, 13904264.0 ms), (13904442.0 ms, 13904483.0 ms), (13904708.0 ms, 13904723.0 ms), (13904892.0 ms, 13904957.0 ms), (13905162.0 ms, 13905185.0 ms), (13905684.0 ms, 13905697.0 ms), (13905948.0 ms, 13905985.0 ms), (13906228.0 ms, 13906258.0 ms), (13906404.0 ms, 13906447.0 ms), (13906747.0 ms, 13906765.0 ms), (13907322.0 ms, 13907401.0 ms), (13907766.0 ms, 13907794.0 ms)], as (start,stop) tuples)
In [30]: el.r.pupA
Out[30]:
masked_array(data = [3875.0 3874.0 3867.0 ..., 1588.0 1590.0 1594.0],
mask = [False False False ..., False False False],
fill_value = 1e+20)
In [31]: during_saccade = el.r[el.r.sacepochs[4]]
In [32]: during_saccade.x
Out[32]:
masked_array(data = [348.2 347.8 347.6 346.5 346.3 344.6 343.7 341.3 339.1 335.5 333.0 328.1
323.7 319.7 315.6 314.7 313.1 311.7 311.7 311.7 313.3 313.3 315.2 315.2
315.3 316.7 316.7 315.6 315.2 313.0 312.1 311.5],
mask = [False False False False False False False False False False False False
False False False False False False False False False False False False
False False False False False False False False],
fill_value = 1e+20)
In [33]: during_saccade.y
Out[33]:
masked_array(data = [363.0 363.0 362.8 362.3 362.3 363.7 365.3 365.5 367.7 368.1 371.8 374.2
376.0 378.2 382.1 384.7 386.0 389.5 390.2 390.7 390.7 389.7 387.7 387.7
385.7 383.0 380.2 379.1 376.8 376.8 376.8 376.1],
mask = [False False False False False False False False False False False False
False False False False False False False False False False False False
False False False False False False False False],
fill_value = 1e+20)
In [34]: during_saccade.vel
Out[34]:
masked_array(data = [-- 0.0129398193923 0.0299923468798 0.0284499922247 0.0475870289105
0.0790078183227 0.076276294298 0.105381781955 0.129392460549
0.147789431206 0.19157977555 0.206206547975 0.187775556203 0.202036125135
0.160674261959 0.0901188754409 0.110041557531 0.0850538300641
0.0229007633588 0.0338913119245 0.0377063581096 0.0690588310646
0.0542967098007 0.0382220185382 0.0947345641721 0.108765761834
0.0778624397933 0.071856471239 0.0689702250133 0.0632653061224
0.0334206015821 --],
mask = [ True False False False False False False False False False False False
False False False False False False False False False False False False
False False False False False False False True],
fill_value = 1e+20)
In [35]: import pyarbus.viz as viz
In [36]: viz.plot_xyp(el.r)
In [37]: viz.plot_xyp?
Type: function
Base Class: <type 'function'>
String Form:<function plot_xyp at 0x3f901b8>
Namespace: Interactive
File: /home/pi/code/pyarbus/pyarbus/viz.py
Definition: viz.plot_xyp(eye, axes=None, subtract_t0=True)
Docstring:
Plots, on three separate subplots, the pupil area, x, and y position
reported by the eyetracker as a function of time.
If you have any questions, my email address is pi berkeley edu with the
appropriate characters replacing the spaces.
- best,
- Paul Ivanov
2011-10-18 - initial github posting (pi)