Kriging three-dimensional interpolation using MATLAB

Kriging three-dimensional interpolation matlab

theta = [10 10]; lob = [1e-1 1e-1]; upb = [20 20];

[dmodel, perf] = dacefit([lat,lon], tem, @regpoly0, @corrgauss, theta, lob, upb);

LonLat = gridsamp([min(latlim) min(lonlim);max(latlim) max(lonlim)], 60);

TemNew = predictor(LonLat, dmodel);

LatNew = reshape(LonLat(:,1),[60,60]);

LonNew = reshape(LonLat(:,2),[60,60]);

TemNew = reshape(TemNew, size(LonNew));

geoshow(LatNew,LonNew,TemNew,'DisplayType','surface');

hold on

plotm(lat,lon,'k.');

colorbar;

What does nargin mean in matlab

In matlab, epochs is the number of times the neuron weights and thresholds are adjusted based on the output error return during calculation.

Authentication method:

(1) Using network linearlayer

1,cell input form

Input P={[1;2] [2;1] [2;3] [3;1]};

Target value T={4 5 7 7}

Use adapt;

input the command:

P={[1;2] [2;1] [2;3] [3;1]};

T={4 5 7 7};

net=linearlayer(0,0.1);

net=configure(net,P,T);

net.IW{1,1}=[0,0];

net.b{1}=0;

[net,a,e]=adapt(net,P,T);

The weight is updated 4 times, the final value is:

net.IW{1,1}= 1.5600 1.5200

net.b{1}=0.9200

Simulation results: [0] [2] [6.0000] [5.8000]

2, matrix input form

Input P=[1 2 2 3;2 1 3 1];

Output T=[4 5 7 7]

Use adapt;

input the command:

P=[1 2 2 3;2 1 3 1];

T=[4 5 7 7];

net=linearlayer(0,0.01);

net=configure(net,P,T);

net.IW{1,1}=[0,0];

net.b{1}=0;

[net,a,e]=adapt(net,P,T);

The weight is updated once, and the final value is:

net.IW{1,1}=0.4900 0.4100

net.b{1}= 0.2300

3, matrix input form

Input P=[1 2 2 3;2 1 3 1];

Output T=[4 5 7 7]

Use train; (set epochs=1)

Prerequisite: Add explicit calling commands to the learning function and training function;

P=[1 2 2 3;2 1 3 1];

T=[4 5 7 7];

net=linearlayer(0,0.01);

net=configure(net,P,T);

net.IW{1,1}=[0,0];

net.b{1}=0;

net=trian(net,P,T);

The weight is updated once, and the final value is:

net.IW{1,1}=0.4900 0.4100

net.b{1}= 0.2300

Conclusion: For static networks, the cell input of linearlayer and adapt is online learning, while the matrix input is offline learning, which is equivalent to one round of train.

As for the dynamic network: do it when you have time.

The above is the detailed content of Kriging three-dimensional interpolation using MATLAB. For more information, please follow other related articles on the PHP Chinese website!