Multiple input and multiple output | MATLAB implements CNN-BiGRU convolution bidirectional gated cyclic unit multiple input and multiple output

Multiple input and multiple output | MATLAB implements CNN-BiGRU convolution bidirectional gated cyclic unit multiple input and multiple output

Prediction effect

basic introduction

MATLAB implements CNN-BiGRU convolutional bidirectional gated cyclic unit multi-input and multi-output, and the running environment is Matlab2020 and above. Using the feature fusion method, the shallow features and deep features are extracted and connected through the convolution network, the features are fused through convolution, and the obtained vector information is input into the BiGRU unit.
Matlab implements CNN-BiGRU convolution bidirectional gated cyclic unit multi-input multi-output prediction
1. data is a data set, 10 input features, 3 output variables.
2.MainCNN_BiGRUNM.m is the main program file, and the running environment is Matlab2020b and above.
3. The command window outputs MAE, MBE and R2, and the data and program content can be obtained in the download area.

programming

• How to download the complete program and data: Private message the blogger to reply to MATLAB to implement CNN-BiGRU convolutional bidirectional gated cyclic unit multiple input and multiple output .

%-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
%%
% 层设置，参数设置
inputSize = size(xnorm{
    
    1},1);   %数据输入x的特征维度
outputSize = size(ynorm,2);     %数据输出y的维度  
numhidden_units1=50;            %网络单元
numhidden_units2= 20;
numhidden_units3=100;
%%
% gru
————————————————
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/127418340
%%  参数设置
options = trainingOptions('adam', ...       % Adam 梯度下降算法
    'MiniBatchSize', 100, ...               % 批大小
    'MaxEpochs', 1000, ...                  % 最大迭代次数
    'InitialLearnRate', 1e-2, ...           % 初始学习率
    'LearnRateSchedule', 'piecewise', ...   % 学习率下降
    'LearnRateDropFactor', 0.1, ...         % 学习率下降因子
    'LearnRateDropPeriod', 700, ...         % 经过700次训练后 学习率为 0.01 * 0.1
    'Shuffle', 'every-epoch', ...           % 每次训练打乱数据集
    'ValidationPatience', Inf, ...          % 关闭验证
    'Plots', 'training-progress', ...       % 画出曲线
    'Verbose', false);
%--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
%%  训练模型
net = trainNetwork(p_train, t_train, layers, options);
%--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
%%  仿真预测
t_sim1 = predict(net, p_train); 
t_sim2 = predict(net, p_test ); 

Past highlights

MATLAB implements RBF radial basis neural network multi-input multi-output prediction
MATLAB implements BP neural network multi-input multi-output prediction
MATLAB implements DNN neural network multi-input multi-output prediction

References

[1] https://blog.csdn.net/kjm13182345320/article/details/116377961
[2] https://blog.csdn.net/kjm13182345320/article/details/127931217
[3] https://blog.csdn.net/kjm13182345320/article/details/127894261