how i do weekly average?
10 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
Lilya
am 27 Sep. 2017
Bearbeitet: Tanziha Mahjabin
am 6 Mär. 2020
Hi all,
I have a matrix of 978*744 which is = no. observation*hours in a month. and the attached mat file is the DateTime vector. I want to make a weekly average.
Thank you in advance.
2 Kommentare
Jan
am 27 Sep. 2017
The MAT file contains the datetime vector of what? I do not have access to a Matlab engine currently, therefore I cannot inspect the MAT file. Otherwise perhaps I could guess, what it contains. But maybe guessing is less efficient than if you explain it.
Akzeptierte Antwort
Andrei Bobrov
am 28 Sep. 2017
Bearbeitet: Andrei Bobrov
am 28 Sep. 2017
Let data - your array [978 x 744].
For MATLAB >= R2016b (here week -> [Sun : Sat]):
Time = (datetime([2017 7 1 0 0 0]):hours(1):datetime([2017 7 31 23 0 0]))';
% or: Time = datetime(timeCopy);
TT = array2timetable(data','R',Time);
TT_out = retime(TT,'weekly','mean');
For MATLAB < R2016b ( here week -> [Mon : Sun]):
data = data';
Time = (datetime([2017 7 1 0 0 0]):hours(1):datetime([2017 7 31 23 0 0]))';
wd = rem(weekday(Time) + 5, 7) + 1;
ii = [true;diff(wd == 1) == 1];
g = cumsum(ii);
[r,c] = ndgrid(g,1:size(TT,2));
TT_out = [table(Time(ii),'V',{'Time'}),...
array2table(accumarray([r(:),c(:)],data(:),[],@mean))];
2 Kommentare
Tanziha Mahjabin
am 3 Mär. 2020
Bearbeitet: Tanziha Mahjabin
am 6 Mär. 2020
Hi,
what about the array with 3 dimensions? I have daily average of SST data (340X285X36) which are (lon,lat,time). I want to have weekly average.
Is the [r,c] line correct? Then how can i mean it? And if possible how can i modify the 'for loop'?So far i have this:
clear all; close all; clc;
path(path,'D:\Radar_job\SST')
ncfile='IMOS_aggregation_20200212T151211Z.nc'; %short time
time2=ncread(ncfile,'time');
time2=ncread(ncfile,'time')./86400+datenum(1981,01,01);
SST_lon=ncread(ncfile,'lon');
SST_lat=ncread(ncfile,'lat');
SST=ncread(ncfile,'sea_surface_temperature');
SST=SST-273.15;
id1=find(SST_lat>-39.9&SST_lat<-36.2);
id2=find(SST_lon>137.1&SST_lon<141.9);
lat1=SST_lat(id1);
lon1=SST_lon(id2);
tvec=datevec(time2);
t=datetime(tvec)
[C,uq,iuq]=unique((tvec(:,1:3)),'rows','stable'); %manually look which rows are full weeks
time3=datenum(C);
wd = rem(weekday(time3) + 5, 7) + 1;
ii = [true;diff(wd == 1) == 1];
g = cumsum(ii);
[r,c] = ndgrid(g,1:size(SST,3));
for i=1:length(time3)
h=figure;
set(gcf,'Position',[200 100 1500 800])
set(gcf,'PaperPositionMode','auto')
m_proj('mercator','lon<gitude>',[137.1 141.9],'lat<itude>',[-39.9 -36.2]);
m_pcolor(lon1,lat1,(SST(id2,id1,i))');
shading interp;
m_proj('mercator','lon<gitude>',[137.1 141.9],'lat<itude>',[-39.9 -36.2]);
m_gshhs_h('patch',[0.7 0.7 0.7],'LineWidth',1.5);
m_grid('ytick',[-90:0.5:90],'xtick',[-180:0.5:180],'tickdir','in','FontSize',14);
colormap jet
colorbar
caxis manual
caxis([15 19])
end
Weitere Antworten (0)
Siehe auch
Kategorien
Mehr zu Dates and Time finden Sie in Help Center und File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!