Weather monitoring station

README.md

@@ -1,3 +1,5 @@
+![Weather](https://user-images.githubusercontent.com/12642226/126882902-fed11a73-99d1-41a8-8533-7b5cd16f4c8d.png)
+
 # To work better it is recommended:
 - The main code in the project folder
 - The data in a subfolder called "Data"

src/PlotFigure.m

@@ -0,0 +1,13 @@
+clear;clc;
+Datos=load('agriculture.mat');Datos=Datos.agriculture;
+Min=min(Datos);
+Max=max(Datos);
+%nomalization
+nDatos=(Datos./Max);%-Min;
+%% plot
+figure
+plot(nDatos(:,2:5));
+hold on
+plot(nDatos(:,6),'--');
+legend("Relative Humidity","Environment Temperature","Soil moisture"...
+    ,"Light intensity","Rain Occurrence",'NumColumns',3);

src/fBar_MaeMseR2.m

@@ -0,0 +1,27 @@
+% Bar graph root mean absolute error (MAE), mean square error (MSE) and R squared
+% input: yest, y youtput
+% return: mae,mse, r2
+% More examples: https://github.com/vasanza/Matlab_Code
+% Read more: https://vasanza.blogspot.com/
+function [mae,mse,r2] = fBar_MaeMseR2(yest,youtput)
+    mae = sum(abs(yest-youtput)/length(yest));
+    %mse = mean((output - yest).^2);
+    mse = immse(yest, youtput);
+    r2 = fR2(youtput,yest);
+    % Difference between the mean square error and the true value 
+    %dmser=mean(sqrt((yest-youtput).^2)./youtput);
+
+    c = categorical({'MAE','MSE','1-R2'});
+    values = [mae mse 1-r2];
+    figure;
+    b=bar(c,values);
+
+    %xlabel('xlabel')
+    ylabel('Error')
+    title('MAE, MSE and 1-R2')
+
+    xtips1 = b(1).XEndPoints - 0.2;
+    ytips1 = b(1).YEndPoints + 0.0003;
+    labels1 = string(b(1).YData);
+    text(xtips1,ytips1,labels1,'VerticalAlignment','middle')
+end

src/fBar_MseR2.m

@@ -0,0 +1,26 @@
+% Bar graph root mean square error (MSE) and R squared
+% input: yest, y youtput
+% return: mse, r2
+% More examples: https://github.com/vasanza/Matlab_Code
+% Read more: https://vasanza.blogspot.com/
+function [mse,r2] = fBar_MseR2(yest,youtput)
+    %rmse = sqrt(mean((output - yest).^2));
+    %rmse = sqrt(immse(yest, youtput));
+    %mse = mean((output - yest).^2);
+    mse = immse(yest, youtput);
+    r2 = fR2(youtput,yest);
+
+    c = categorical({'MSE','1-R2'});
+    values = [mse 1-r2];
+    figure;
+    b=bar(c,values);
+
+    %xlabel('xlabel')
+    ylabel('Error')
+    title('MSE and 1-R2')
+
+    xtips1 = b(1).XEndPoints - 0.2;
+    ytips1 = b(1).YEndPoints + 0.0003;
+    labels1 = string(b(1).YData);
+    text(xtips1,ytips1,labels1,'VerticalAlignment','middle')
+end

src/fBar_RmseMseMae.m

@@ -0,0 +1,28 @@
+% Bar graph root mean square error (RMSE), mean square error (MSE) and mean absolute error (MAE)
+% input: yest, y youtput
+% return: rmse,mse, mae
+% More examples: https://github.com/vasanza/Matlab_Code
+% Read more: https://vasanza.blogspot.com/
+function [rmse,mse,mae] = fBar_RmseMseMae(yest,youtput)
+    %rmse = sqrt(mean((output - yest).^2));
+    rmse = sqrt(immse(yest, youtput));
+    %mse = mean((output - yest).^2);
+    mse = immse(yest, youtput);
+    mae = sum(abs(yest-youtput)/length(yest));
+    % Difference between the mean square error and the true value 
+    %dmser=mean(sqrt((yest-youtput).^2)./youtput);
+
+    c = categorical({'MAE','MSE','RMSE'});
+    values = [rmse mse mae];
+    figure;
+    b=bar(c,values);
+
+    %xlabel('xlabel')
+    ylabel('Error')
+    title('RMSE, MSE and MAE')
+
+    xtips1 = b(1).XEndPoints - 0.2;
+    ytips1 = b(1).YEndPoints + 0.0003;
+    labels1 = string(b(1).YData);
+    text(xtips1,ytips1,labels1,'VerticalAlignment','middle')
+end

src/myNeuralNetworkFunction_BR.m

@@ -0,0 +1,73 @@
+function [y1] = myNeuralNetworkFunction_BR(x1)
+%MYNEURALNETWORKFUNCTION neural network simulation function.
+%
+% Auto-generated by MATLAB, 12-Jul-2021 12:16:11.
+%
+% [y1] = myNeuralNetworkFunction(x1) takes these arguments:
+%   x = Qx5 matrix, input #1
+% and returns:
+%   y = Qx1 matrix, output #1
+% where Q is the number of samples.
+
+%#ok<*RPMT0>
+
+% ===== NEURAL NETWORK CONSTANTS =====
+
+% Input 1
+x1_step1.xoffset = [0.608695652173913;0.82370820668693;0.433566433566434;0.0124984476084822;0];
+x1_step1.gain = [5.11111111111111;11.3448275862069;3.53086419753086;2.02531327181859;4];
+x1_step1.ymin = -1;
+
+% Layer 1
+b1 = [0.00035174026259418046669;0.00051092816047968955297;-0.00048391175546749315125;-0.00050981894364561498;0.00053107338220117399161;0.00042076275179294889269;-0.08001026386788413769;-0.00052123117927782200952;0.15745448440093531839;-0.067634534449094715902];
+IW1_1 = [0.00029940427633355415132 -0.00091351092567174430742 0.00035050945272335786752 0.00054201144080147822986 -0.0006553261154915732481;0.00043483496913422018532 -0.0013268466275537572387 0.00050908684357890937426 0.00078723075997475118991 -0.00095181600158640057574;-0.00041185539151189476913 0.0012567042505490203137 -0.00048217772185979798508 -0.00074561911857309566901 0.00090150388804861110621;-0.00043389146051596373839 0.0013239667398766085359 -0.00050798200136364002506 -0.00078552225583856721478 0.00094975028080367834317;0.00045196848775252984609 -0.0013791473561833049108 0.00052915087847809467135 0.00081825737359259107656 -0.00098932999443373213777;0.00035813397574167632129 -0.0010927404007482076527 0.00041927332239120770592 0.00064834552211519617081 -0.00078389209306386792731;0.60348615247481129931 -0.31381583708022858792 0.57257217686362416043 0.58517705395397245915 -0.00041552308730919947435;-0.00044359788315542270955 0.0013535954267334779529 -0.00051934848696569009107 -0.000803099149488242358 0.00097100231079647308235;-0.17588209086211484267 -0.11370070401550631811 -0.98882602763530946799 0.27919569916621544969 0.068327356243596343299;-0.8882180602117247803 0.1560561953219604947 -0.4013236356394274118 -0.18485228807893513969 0.051201117667770981723];
+
+% Layer 2
+b2 = 0.21194446183604778722;
+LW2_1 = [0.0013771378901323138438 0.0020005088943848372586 -0.0018946378611525214845 -0.0019967650998719966952 0.0020801898158251267215 0.0016473761574264276716 0.56432693796989685797 -0.0020410398899658637253 -1.1382824405110656407 0.52130642137983940199];
+
+% Output 1
+y1_step1.ymin = -1;
+y1_step1.gain = 3.53086419753086;
+y1_step1.xoffset = 0.433566433566434;
+
+% ===== SIMULATION ========
+
+% Dimensions
+Q = size(x1,1); % samples
+
+% Input 1
+x1 = x1';
+xp1 = mapminmax_apply(x1,x1_step1);
+
+% Layer 1
+a1 = tansig_apply(repmat(b1,1,Q) + IW1_1*xp1);
+
+% Layer 2
+a2 = repmat(b2,1,Q) + LW2_1*a1;
+
+% Output 1
+y1 = mapminmax_reverse(a2,y1_step1);
+y1 = y1';
+end
+
+% ===== MODULE FUNCTIONS ========
+
+% Map Minimum and Maximum Input Processing Function
+function y = mapminmax_apply(x,settings)
+y = bsxfun(@minus,x,settings.xoffset);
+y = bsxfun(@times,y,settings.gain);
+y = bsxfun(@plus,y,settings.ymin);
+end
+
+% Sigmoid Symmetric Transfer Function
+function a = tansig_apply(n,~)
+a = 2 ./ (1 + exp(-2*n)) - 1;
+end
+
+% Map Minimum and Maximum Output Reverse-Processing Function
+function x = mapminmax_reverse(y,settings)
+x = bsxfun(@minus,y,settings.ymin);
+x = bsxfun(@rdivide,x,settings.gain);
+x = bsxfun(@plus,x,settings.xoffset);
+end

src/myNeuralNetworkFunction_LM.m

@@ -0,0 +1,73 @@
+function [y1] = myNeuralNetworkFunction_LM(x1)
+%MYNEURALNETWORKFUNCTION neural network simulation function.
+%
+% Auto-generated by MATLAB, 12-Jul-2021 12:17:06.
+%
+% [y1] = myNeuralNetworkFunction(x1) takes these arguments:
+%   x = Qx5 matrix, input #1
+% and returns:
+%   y = Qx1 matrix, output #1
+% where Q is the number of samples.
+
+%#ok<*RPMT0>
+
+% ===== NEURAL NETWORK CONSTANTS =====
+
+% Input 1
+x1_step1.xoffset = [0.608695652173913;0.82370820668693;0.433566433566434;0.0124984476084822;0];
+x1_step1.gain = [5.11111111111111;11.3448275862069;3.53086419753086;2.02531327181859;4];
+x1_step1.ymin = -1;
+
+% Layer 1
+b1 = [2.1949853578616629335;-1.7469328441935882967;0.99240755064494678983;1.0218472696166676084;-0.20364962239710274194;0.075006696145808071652;0.86849978954689988075;-1.7528113775439608801;-2.1635286289685735639;1.9533738151333810418];
+IW1_1 = [-1.0200111942236536056 0.18371207872449493714 -1.4761242942088723673 -0.68794352417100779196 1.0905174575592302411;0.15849883407858353368 -0.76811703825605215368 1.6988832804854816505 -0.68283686162590440105 0.67432444400053681566;-0.64349214417223599138 0.40030808532394601684 1.1057920048993226114 0.21412508236109187298 1.1990683467370717441;-0.48795612410084338029 -1.2250250302922145451 0.67407093082124247552 0.81674746062738967645 -1.5287905283841298765;0.36787248393377131039 0.64062360566131126838 1.4243599683899237363 -0.54873720927763891542 -1.4872545148778710811;-0.9268852109842726783 -1.0196075464390572662 -0.34597136023448776809 -1.4471455571856632893 -0.5530734974790058045;0.48424032314825560253 -0.1811170079280729206 0.97468818170933824163 -1.3859231772694791118 1.142501422637789199;-0.64255807181742796708 0.67010085976123234808 -2.6434921749012407766 -0.68891688899908154475 1.4610370159121313094;0.15189272482242013873 0.50924382354552510943 1.5003722988369436742 -1.9880704228772694275 0.8904186285136883594;0.28925144735049329592 0.5742974445922626181 -1.0987458626795434391 -2.0475217260707032629 -0.7219189729790551624];
+
+% Layer 2
+b2 = 0.49227585916764610152;
+LW2_1 = [-0.5167614963278497342 -0.11906540081385030838 0.61823532419030025054 0.19156553538645498813 0.23647371873576578105 -0.012726615235693293157 0.092360116876703701738 -0.334773211566524731 0.42646907290742258612 -0.34115551012335215697];
+
+% Output 1
+y1_step1.ymin = -1;
+y1_step1.gain = 3.53086419753086;
+y1_step1.xoffset = 0.433566433566434;
+
+% ===== SIMULATION ========
+
+% Dimensions
+Q = size(x1,1); % samples
+
+% Input 1
+x1 = x1';
+xp1 = mapminmax_apply(x1,x1_step1);
+
+% Layer 1
+a1 = tansig_apply(repmat(b1,1,Q) + IW1_1*xp1);
+
+% Layer 2
+a2 = repmat(b2,1,Q) + LW2_1*a1;
+
+% Output 1
+y1 = mapminmax_reverse(a2,y1_step1);
+y1 = y1';
+end
+
+% ===== MODULE FUNCTIONS ========
+
+% Map Minimum and Maximum Input Processing Function
+function y = mapminmax_apply(x,settings)
+y = bsxfun(@minus,x,settings.xoffset);
+y = bsxfun(@times,y,settings.gain);
+y = bsxfun(@plus,y,settings.ymin);
+end
+
+% Sigmoid Symmetric Transfer Function
+function a = tansig_apply(n,~)
+a = 2 ./ (1 + exp(-2*n)) - 1;
+end
+
+% Map Minimum and Maximum Output Reverse-Processing Function
+function x = mapminmax_reverse(y,settings)
+x = bsxfun(@minus,y,settings.ymin);
+x = bsxfun(@rdivide,x,settings.gain);
+x = bsxfun(@plus,x,settings.xoffset);
+end