Skip to content

Pipeline of processing and modelisation for the detection of seizure on EEG data (dataset TUH/TUSZ)

License

Notifications You must be signed in to change notification settings

cetic/TUH_EEG_Seizure_Detection

Repository files navigation

TUH_EEG_Seizure_Detection

GitHub Super-Linter

This repository will meanly contain the Python code used Vincent STRAGIER for his intership and master dissertation at the CETIC. More information will be added to the project Wiki in order to explain how to use this pipeline.

UMONS cluster related instructions

1 - Get access to the cluster:

ssh username@******.ac.be -p 22000
  • username: ******************
  • password: ******************
  • password hint: ******************

2 - Get rid of the password

3 - Get VSCode working (+ use socket).

4 - Install the Python version you want without privilege

  • Set it up as a module with Lmod

    • mkdir ~/modulefiles/python

    • nano 3.8.6rc1.lua

    • add the following lines:

      help([==[This module sets the PATH variable for python/3.8.6rc1]==])
      
      whatis([==[Description:Python is a programming language that lets you work quickly and integrate systems more effectively. ]==])
      whatis([==[Homepage: https://python.org ]==])
      
      local pythonroot = "$HOME/python-3.8/Python-3.8.6rc1"
      -- Workaround to use the library not compiled with Python (copied from Python 3.6)
      local lib_dynload = "$HOME/python-3.8/lib/python3.8/lib-dynload"
      -- Allows to use ipython and pip
      local python_utils = "$HOME/python-3.8/bin"
      
      conflict("python")
      --setenv("PYTHONPATH", pythonroot)
      prepend_path("PATH", pythonroot)
      prepend_path("PATH", python_utils)
      prepend_path("PYTHONPATH", pythonroot)
      
      whatis("Name         : Python 3.8.6rc1")
      whatis("Version      : 3.8.6rc1")
      whatis("Category     : Interpreter")
      whatis("Description  : Python environment ")
      whatis("URL          : https://python.org/ ")
      
      family("python")
  • Load the module on startup

    • $nano ~/.bashrc

    • add the following lines to the end of the file (check if python exists and if the modules folder exists):

      if [ -f ~/python-3.8/Python-3.8.6rc1/python ] && [ -d ~/modulefiles/python ]; then
        module use ~/modulefiles >/dev/null
        module load python/3.8.6rc1 >/dev/null
      fi

      N.B. : Remove >/dev/null if you want to show the output. Replace it with &> out.log if you want to log the output in the file out.log. You can also add the line alias py='python' (Windows like command).

    • $source ~/.bashrc (restart the script)

5 - Download the dataset (TUH_EEG_SZ - v1.5.2)

$python ./tools/tuh_sz_download.py "https://www.isip.piconepress.com/projects/tuh_eeg/downloads/tuh_eeg_seizure/v1.5.2/" nedc_tuh_eeg nedc_tuh_eeg --path /home_nfs/stragierv/TUH_SZ_v1.5.2/

6 - Remove VSCode watch on the TUH folder

About

Pipeline of processing and modelisation for the detection of seizure on EEG data (dataset TUH/TUSZ)

Resources

License

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •