1.Matlab实现MTF-CNN-Mutilhead-Attention基于马尔可夫转移场-卷积神经网络融合多头注意力多特征数据分类预测(完整源码和数据)
2.自带数据,多输入,单输出,多分类。图很多,混淆矩阵图、预测效果图等等。MTF将一维信号转换为二维特征图,而CNN可以对这些特征图进行自适应的特征提取和分类,融合多头注意力机制有效把握提取特征的贡献程度。
3.直接替换数据即可使用,保证程序可正常运行。运行环境MATLAB2022及以上。
4.代码特点:参数化编程、参数可方便更改、代码编程思路清晰、注释明细。
%--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
P_train = res(temp(1: 60), 2: 16)';
T_train = res(temp(1: 60), 1)';
M = size(P_train, 2);
%--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
P_test = res(temp(61: end), 2: 16)';
T_test = res(temp(61: end), 1)';
N = size(P_test, 2);
%--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
%% 数据归一化
[P_train, ps_input] = mapminmax(P_train, 0, 1);
P_test = mapminmax('apply', P_test, ps_input);
%--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
t_train = categorical(T_train)';
t_test = categorical(T_test )';
————————————————
版权声明:本文为CSDN博主「机器学习之心」的原创文章,遵循CC 4.0 BY-SA版权协议,转载请附上原文出处链接及本声明。
原文链接:https://blog.csdn.net/kjm13182345320/article/details/127177589
版权声明:本文为CSDN博主「机器学习之心」的原创文章,遵循CC 4.0 BY
[1] https://blog.csdn.net/kjm13182345320/article/details/129036772?spm=1001.2014.3001.5502
[2] https://blog.csdn.net/kjm13182345320/article/details/128690229