Feature Extraction:
Firstly, extract features or load final_workspace.mat which contains a backup
To extract features:
1) Download DEAP dataset and store it in code folder
2) Run psd_hamming_overlap.m to extract psd features. These are stored in a variable 'feature_vector'
3) Run 'msce_features_matrix.m' to extract MSCE features. These are stored in a variable ' msce_features_matrix'
(Note: msce_features_matrix.m requires MATLAB 2013a or plus and many hours on 8GB machine)
Alternatively, load 'final_workspace.mat' which has the 'feature_vector' and 'msce_features_matrix' the variables
Running SVM with PCA and RBF Kernel and MSCE Features:
Install 'libsvm' and set path
Run pca_svm.m
The gamma, c values and number of PCA dimensions can be altered inside pca_svm.m
It performs 10 trials by default and reports mean and standard deviation of accuracies on console
Running Neural Network Classifier with PCA:
Run pca_neural.m
This uses neural_classifier.m, verify.m and threshold.m
The number of number of hidden nodes can be altered inside neural_classifier.m
The number of PCA dimensions and number of trials can be changed in pca_neural.m
The mean and standard deviation of accuracy is reported on console
Running Logistic Regression:
Run pca_logistic.m
This uses reg.m and check_reg.m
The stopping threshold for logistic regression can be changed from inside reg.m
The number of PCA dimensions and number of trials can be changed in pca_logistic.m
The mean and standard deviation of accuracy is reported on console
Running SVM with PCA and MSCE Features on per person basis:
Run pca_svm_personwise.m
Number of persons, PCA dimensions, gamma and c values can be changed inside pca_svm_personwise.m
Reports a vector of mean of accuracies for each person
Reports a vector of std. of accuracies for each person
Running Neural Network with PCA and MSCE Features on per person basis:
Run pca_neural_personwise.m
The number of number of hidden nodes can be altered inside neural_classifier.m
Number of persons, PCA dimensions can be changed inside pca_svm_personwise.m
Reports a vector of mean of accuracies for each person
Reports a vector of std. of accuracies for each person
Running Logistic Regression with PCA and MSCE Features on per person basis:
Run pca_logistic_personwise.m
This uses reg.m and check_reg.m
This uses neural_classifier.m, verify.m and threshold.m
The stopping threshold for logistic regression can be changed from inside reg.m
Number of persons, PCA dimensions can be changed inside pca_svm_personwise.m
Reports a vector of mean of accuracies for each person
Reports a vector of std. of accuracies for each person
deap_learning.rar
版权申诉
5星 · 超过95%的资源 148 浏览量
2021-04-12
17:26:08
上传
评论
收藏 11.01MB RAR 举报
大大U
- 粉丝: 722
- 资源: 136
最新资源
- 基于Python的新能源承载力计算及界面设计源码 - HAINING-DG
- 基于Java的本科探索学习项目设计源码 - 本科探索
- 基于Javascript和Python的微商城项目设计源码 - MicroMall
- 基于Java的网上订餐系统设计源码 - online ordering system
- 基于Javascript的超级美眉网络资源管理应用模块设计源码
- 基于Typescript和PHP的编程知识储备库设计源码 - study-php
- Screenshot_2024-05-28-11-40-58-177_com.tencent.mm.jpg
- 基于Dart的Flutter小提琴调音器APP设计源码 - violinhelper
- 基于JavaScript和CSS的随寻订购网页设计源码 - web-order
- 基于MATLAB的声纹识别系统设计源码 - VoiceprintRecognition
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈