fitting data form a matrix using loops
显示 更早的评论
hello,
I made the following chart with averages, error bars and fittings starting from my data set. It took me some time as I am quite new to Matlab, but I know I did it wrong. I should be using loops but I have tried several times with no success.
Can someone please help me to understand it better?
Thank you very much!!!
d=readmatrix("shear_Stress_more.csv");
ctrl = mean(d(:,2:5),2);
pef = mean(d(:,6:9),2);
USa = mean(d(:,10:11),2);
USb = mean(d(:,12:13),2);
enz = mean(d(:,14:17),2);
medie=[ctrl pef USa USb enz];
st_ctrl = std(d(:,2:5),0,2);
st_pef = std(d(:,6:9),0,2);
st_USa = std(d(:,10:11),0,2);
st_USb = std(d(:,12:13),0,2);
st_enz = std(d(:,14:17),0,2);
stdev = [st_ctrl st_pef st_USa st_USb st_enz];
ft = fittype('a+(b-a)*exp(-c*x)', ...
'dependent',{'y'}, ...
'independent',{'x'}, ...
'coefficients',{'a','b','c',});
hold on
[f,gof] = fit(d(:,1),medie(:,1), ft,'StartPoint',[3, 2, 0.01]);
errorbar(d(:,1),medie(:,1),stdev(:,1));
plot(f);
[f2,gof2] = fit(d(:,1),medie(:,2), ft,'StartPoint',[3, 2, 0.01]);
errorbar(d(:,1),medie(:,2),stdev(:,2));
plot(f2);
[f3,gof3] = fit(d(:,1),medie(:,3), ft,'StartPoint',[3, 2, 0.01]);
errorbar(d(:,1),medie(:,3),stdev(:,3));
plot(f3);
[f4,gof4] = fit(d(:,1),medie(:,4), ft,'StartPoint',[3, 2, 0.01]);
errorbar(d(:,1),medie(:,4),stdev(:,4));
plot(f4);
[f5,gof5] = fit(d(:,1),medie(:,5), ft,'StartPoint',[3, 2, 0.01]);
errorbar(d(:,1),medie(:,5),stdev(:,5));
plot(f5);
xlabel({'Time t [s]'});
ylabel({'Shear stress σ [Pa]'});
title({''});
legend('Control','PEF','USa','USb','Enzyme');
hold off
采纳的回答
Walter Roberson
2022-9-19
You do not need all those variables at the same time, not for the purpose of the plot. You could look
for K = 1 : 5
start = (K-1)*4 + 2;
stop = start + 3;
thisdata = data(:,start:stop);
m = mean(thisdata,2);
s = std(thisdata);
[f, gof] = fit(d(:,1), m, ft, 'StartPoint', [3, 2, 0.01]);
errorbar(d(:,1), m, s);
h(K) = plot(f);
hold on
end
hold off
xlabel({'Time t [s]'});
ylabel({'Shear stress σ [Pa]'});
title({''});
legend(h, {'Control','PEF','USa','USb','Enzyme'});
更多回答（1 个）
dpb
2022-9-19
No real deal...your first part is somewhat more of a pain since you're averaging over a variable number of columns -- if it's a one-time exercise, I'd probably basically leave it along and only deal with it further if have to change these column IDs manuall
d=readmatrix("shear_Stress_more.csv");
% build summary data arrays
davg=[mean(d(:,2:5),2) mean(d(:,6:9),2) mean(d(:,10:11),2) mean(d(:,12:13),2) mean(d(:,14:17),2)];
dstd=[std(d(:,2:5),2) std(d(:,6:9),2) std(d(:,10:11),2) std(d(:,12:13),2) std(d(:,14:17),2)];
ft = fittype('a+(b-a)*exp(-c*x)', ...
'dependent',{'y'}, ...
'independent',{'x'}, ...
'coefficients',{'a','b','c',});
hold on
% build fit across columns
for i=1:size(davg,2);
[f{i},gof{i}] = fit(d(:,1),davg(:,i),ft,'StartPoint',[3, 2, 0.01]);
errorbar(d(:,1),davg(:,i),dstd(:,i));
plot(f{i})
end
xlabel({'Time t [s]'});
ylabel({'Shear stress σ [Pa]'});
title({''});
legend('Control','PEF','USa','USb','Enzyme');
hold off
类别
在 帮助中心 和 File Exchange 中查找有关 Image Processing Toolbox 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
