Skip to content

This repository contains the tensorflow implementation for the paper: "Emotion Recognition from Multi-Channel EEG through Parallel Convolutional Recurrent Neural Network"

Notifications You must be signed in to change notification settings

ynulonger/ijcnn

Repository files navigation

Code for IJCNN 2018 submission

This repository contains the tensorflow implementation for the paper: "Emotion Recognition from Multi-Channel EEG through Parallel Convolutional Recurrent Neural Network"

About the paper

Instructions

  • Before running the code, please download the DEAP dataset, unzip it and place it into the right directory. The dataset can be found here. Each .mat data file contains the EEG signals and consponding labels of a subject. There are 2 arrays in the file: data and labels. The shape of data is (40, 40, 8064). The shape of label is (40,4). Each .pkl file contains a numpy.ndarray variable. It stores the pre_processed data with the shape of (segments, window_size, width, height), in this paper, it is (2400,128,9,9).
  • Please run the deap_pre_process.py to Load the origin .mat data file and transform it into .pkl file.
  • Using cv.py to train and test the model (10-fold cross-validation), result of each fold will be saved in a .xls file.
  • count_accuracy.py is used to caculate the final accuracy of the model.

Requirements

  • Pyhton 3
  • scipy
  • numpy
  • pandas
  • pickle
  • sk-learn
  • pickle
  • tensorflow (1.4 or higher version)
  • xlrd
  • xlwt

About

This repository contains the tensorflow implementation for the paper: "Emotion Recognition from Multi-Channel EEG through Parallel Convolutional Recurrent Neural Network"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published