Python implementation of Event Level Cross-Category Metric (ELC) proposed in article "Gaze-in-wild: A dataset for studying eye and head coordination in everyday activities".
The Event-based error metric is used for reporting and debugging label sequences from human labelers and classification algorithms. It provides a way to compare and evaluate event sequences, particularly useful in the context of eye-tracking and gaze analysis.
To use this package, clone the repository and install the required dependencies:
git clone https://github.com/your-username/event-level-metric.git
cd event-level-metric
pip install numpy
The main functionality is provided by the elc function in the elc.py file. Here's a basic example of how to use it:
import numpy as np
from elc import elc
test_r = np.hstack([[0]*30, [2]*15, [0]*45, [1]*35])
# test_t = np.copy(test_r)
test_t = np.hstack([[0] * 27, [2] * 16, [0] * 47, [2] * 35])
# set the window size for the metric
fs_win = 7 # choose winsize=3.5 for 25ms window
fp_win = 7 # chhose winsize=14 for 100 ms window
winsize = [fs_win, fp_win]
l2dis_all, olr_all, conf_mat, percent_detach = elc(test_r, test_t, winsize)
print("l2dis_all: ", l2dis_all)
print("olr_all: ", olr_all)
print("confusion matrix: ", conf_mat)
print("detached percentage: ", percent_detach)
Please use the bibtext below for your citation:
@article{kothari2020gaze,
title={Gaze-in-wild: A dataset for studying eye and head coordination in everyday activities},
author={Kothari, Rakshit and Yang, Zhizhuo and Kanan, Christopher and Bailey, Reynold and Pelz, Jeff B and Diaz, Gabriel J},
journal={Scientific reports},
volume={10},
number={1},
pages={2539},
year={2020},
publisher={Nature Publishing Group UK London}
}