Libeep

libeep is python package to load eego data files with python.

Take a look at the most recent release at https://github.com/translationalneurosurgery/libeep/releases and you might be lucky and skip the need to compile the binaries.

Contribution

My tiny contribution stands on the shoulders of the giant Robert Smies, from whom i forked this from https://gitlab.com/smeeze/libeep/. I added instructions and automated the compilation of an installable python package, revised the API for a cleaner object-oriented approach, minimized the risk of locked files and segmentation faults, and expanded the documentation.

Installation

Linux / Ubuntu

You might need a recent cmake (tested with cmake 3.17.1) and the build-essential, so maybe run first

wget -O - https://apt.kitware.com/keys/kitware-archive-latest.asc 2>/dev/null | sudo apt-key add -
sudo apt-add-repository 'deb https://apt.kitware.com/ubuntu/ bionic main'
sudo apt-get update
sudo apt-get install cmake
sudo apt-get install build-essential

See also Kitwares instructions for installation of cmake. You might also need wheels for Python, so run e.g. pip install wheels. Afterwards this should do the trick:

make clean python
pip install -e .

You will find a file called pyeep.so in build/python/v3. This is a python extension to load eego files.

Linux / Arch

You need cmake, e.g. from the snap store (https://snapcraft.io/install/cmake/arch) and base-devel. Afterwards this should do the trick:

make clean python
pip install -e .

Windows

In Windows you need Visual Studio with C++ support and CMake installed. You might also need wheels for Python, so run e.g. pip install wheels. Then, from the project root, run in the Developer Command Prompt

mkdir build
cmake -S . -B build
cd build
cmake --build . --config Release

You will find a set of files in build/python/v3/Release. This are the python extension files required to load eego files.

Create and install the python package by running from the project root

mkdir libeep
copy build\python\v3\Release\* libeep
copy python\__init__.py libeep
pip install -e .

Usage

Object-oriented-interface for python

After running pip install -e ., you will then be able to import eep and use its object oriented interface.

import libeep
import matplotlib.pyplot as plt

fname = "example.cnt"
cnt = libeep.CntReader(fname)
print(f"Within the file {fname} are")
print(f"{cnt.get_channel_count()} channels and ")
print(f"{cnt.get_sample_count()} samples and")
print(f"{cnt.get_trigger_count()} events")

# load 1s before and after the second event
fs = cnt.get_sample_frequency()
marker, tstamp, *info = cnt.get_trigger(1)
data = cnt.get_samples(tstamp-fs, tstamp+fs)
# plot the data
plt.plot(data)

Low-level interface

Because it is a python extension, pyeep.so can also be simply imported with import pyeep. If the python package is installed, you can use from libeep import pyeep. This does only offer a low-level interface. You can find an example use case here or in the following:

import pyeep
fname = "example.cnt"
fh = pyeep.read(fname)  # get the file handle

sampling_rate = pyeep.get_sample_frequency(fh)

chan_count = pyeep.get_channel_count(fh)
channel_labels = [pyeep.get_channel_label(fh, chan) for chan in range(chan_count)]

sample_count = pyeep.get_sample_count(fh) # get how many samples are there
data = pyeep.get_samples(fh, 0, sample_count)  # load them all


trigger_count = pyeep.get_trigger_count(fh)
markers = [pyeep.get_trigger(fh, trigger) for trigger in range(trigger_count)]