Introduction
zeebeez3 is a refactor of the zeebeez project. The code for zeebeez produced all the analysis and figures for the Schachter 2016 thesis.
In this version, everything is python3 compatible, and utilizes the soundsig package. All dependencies to neosound have been removed.
The base file format for zeebeez3 are nwb files, created with pynwb. There are a series of transformations applied to the nwb files that produce hdf5 files with highly preprocessed data. There is one nwb file per recording site.
Event though the original codebase was developed for experiments in anaesthetized Zebra finch with repeated trials of randomly interleaved stimuli, the representation of the data in the Experiment object is relatively neutral to the type of recording. The data is represented in it's continuous format.
Local Installation
The first step is to clone the repository:
git clone https://github.com/theunissenlab/zeebeez3
This project requires that you have anaconda installed. The next step is to create a virtual environment with the dependencies that zeebeez3 needs:
cd zeebeez3
conda env create -f environment.yml -n zeebeez3
Now activate that environment:
source activate zeebeez3
and install zeebeez3
python setup.py install
Note that at the moment, you will have to have the latest version of soundsig installed. To do this, pull the latest version or clone soundsig, and then from the soundsig directory:
source activate zeebeez3
python setup.py install
Running Notebooks
To run the notebooks, first start jupyter notebooks from the zeebeez3 notebooks directory:
cd notebooks
jupyter notebook
A web browser tab will pop up. Select the notebook that you're interested in.