This program implements the analysis pipeline for an ICL project. The project examines how human memory can affect saccadic reaction times. The results are reported in a separate research report Effects of memory to variation of saccadic reaction times of infants by Akseli Palén and Jukka Leppänen.
The steps of the pipeline are defined under directory tasks/ and written in Python. The source data, intermediate results, and the final results are stored under directory data/. The pipeline is defined and visually presented in the Python script pipeline.py.
Conda is not necessary but gives convenient environment handling. Dependencies include gazelib, numpy, and scipy:
$ conda create --name variability python=3
$ source activate variability
(variability)$ pip install gazelib
Run analysis pipeline by:
$ source activate variability
(variability)$ python pipeline.py
The results of each step are cached under data/ so their unnecessary computation can be skipped. The following tip reveals the caching logic: to force single computation of a step, modify the script of the step or remove or modify one of the input files.
If a file close to the pipeline root is modified, it might require a couple of runs before the change reaches the leaf steps. The pipeline logic needed to prevent this remains to be implemented.
Akseli Palén, akseli.palen@gmail.com
The pipeline is released under MIT license.
The results and the source data under data/ are property of Infant Cognition Laboratory at University of Tampere. Contact Jukka Leppänen (jukka.leppanen@staff.uta.fi) for further permissions.