Skip to content

jlnh/SeizurePrediction

Repository files navigation

American Epilepsy Society Seizure Prediction Challenge

Description

The code was written for the American Epilepsy Society Seizure Prediction Challenge. The software is written in Python. The standard numpy, scipy, scikit-learn and matplotlib packages are used extensively.

Dependency

Hardware and Runtime

The computations were done in a desktop with intel i7 quad core CPU and 12GB RAM. The total computational time is about 2 hours when the classifier is set to use all the CPU threads.

How to Generate the Solution

  • Modify SETTINGS.json file and put the data in the data dir. Sample SETTINGS.json is given here
{
  "competition-data-dir": "seizure-data",
  "data-cache-dir": "data-cache",
  "submission-dir": "submissions",
  "figure-dir": "figure"
}
  • Run predict.py
  • Check the submission file in submissions directory and the analytical graphic PDF file in figure directory

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages