Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

[solved] Analysis of AOI or fixation shifts to the next image

edited June 2016 in PyGaze

Hi,
I am trying to use my new Eye Tribe. I was running script from the following link:
https://github.com/esdalmaijer/PyTribe

the script runs well but when i analyse the data, the fixation points and heat map points shows differently. later I figured out that the fixation point of one image is carried on to the next image.
I think the image display timing and the data obtained from the eye-tribe does not synchronise well.
could anyone tell me what to do????

Comments

  • edited 5:04AM

    Hi Mathew,

    This issue stems from the way log messages were timestamped. In previous versions, they were anchored to the latest sample. However, on some PCs, the sampling started to lag behind. Therefore, log messages that were sent later, appeared as though they were sent earlier. Unfortunately, this issue was very hard to debug, as I couldn't replicate it on any of the computers available to me.

    I did fix the log times, though, with this commit. Download PyGaze from GitHub and give it a test run.

    NOTE: This might still not get rid of the laggy sampling! Make sure to compare your last sample in a trial against a 'trial_end' log message to check whether your system suffers from this lag.

    Cheers,

    Edwin

  • edited 5:04AM

    Hi Edwin

    Thank you so much now I am getting the exact fixation points but during the analysis I am getting the following error message
    Warning (from warnings module):
    File "H:\mathew eyetribe test ready\New\New trial\pygazeanalyser\gazeplotter.py", line 227
    heatmap[y:y+gwh,x:x+gwh] += gaus * fix['dur'][i]
    VisibleDeprecationWarning: using a non-integer number instead of an integer will result in an error in the future

    kindly let me know what to do

    thank you
    Mathew

  • edited 5:04AM

    Hi Mathew,

    That's not an error, that's a warning. It tells us that some of the functionality that we are using is not going to be supported in future editions of NumPy.

    Thanks for the heads-up, I'll have a look at what's causing it, and will fix it. For now, though, there's nothing to worry about. (Assuming the script does what it's supposed to do?)

    Cheers,

    Edwin

  • edited 5:04AM

    Thank you so much

    I think now I am ready to start my work.

    your help, commitment and contribution is really remarkable

    greetings of peace

    mathew

Sign In or Register to comment.