neuroph12 / peacca

PEACCA

This repository contains the implementation code used in the experiments of the following paper:

Francisco Raposo, David Martins de Matos, Ricardo Ribeiro, Suhua Tang, Yi Yu, "Towards Deep Modeling of Music Semantics using EEG Regularizers", CoRR, vol. arXiv:1712.05197, 2017 https://arxiv.org/pdf/1712.05197.pdf

Please cite the previously mentioned article if you use PEACCA in your research and/or software.

download and install Anaconda

wget https://repo.continuum.io/archive/Anaconda3-5.1.0-Linux-x86_64.sh; bash Anaconda3-5.0.1-Linux-x86_64.sh

create anaconda environment

conda create -n peacca python=3.6.4

activate environment

source activate peacca

install tensorflow

pip install --ignore-installed --upgrade https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.5.0-cp36-cp36m-linux_x86_64.whl

install keras, pywt, h5py

pip install keras pywavelets h5py

filter and cut openbci raw eeg files

for i in 000 001 002 003 004 005 006 007 008 009 011 012 014 015 996 997 998 999; do python filter_and_cut_eeg.py datasets/audio_eeg/eeg/openbci/$i.openbci $i 16 datasets/audio_eeg/audio/info/$i.info datasets/audio_eeg/eeg/sources; done

preprocess eeg source files

for i in datasets/audio_eeg/eeg/sources/*; do o=$( basename $i ); python preprocess_eeg.py --war 5 --wsd 0.5 $i datasets/audio_eeg/eeg/preprocessed/$o; done

create models directory

mkdir models

train audio-eeg model and generate embeddings

python compute_audio_embeddings.py

evaluate audio-lyrics model (choi embeddings)

python evaluate_audio_lyrics_retrieval.py datasets/audio_lyrics/audio/features/choi

evaluate audio-lyrics model (spotify features)

python evaluate_audio_lyrics_retrieval.py datasets/audio_lyrics/audio/features/spotify

evaluate audio-lyrics model (audio-eeg model embeddings)

for run in {0..4}; do for fold in {0..4}; do python evaluate_audio_lyrics_retrieval.py datasets/audio_lyrics/audio/features/embeddings/run-${run}_fold-${fold}; done; done