replace values in char array if conditions are met

11 visualizzazioni (ultimi 30 giorni)
I have a datestr of 15 minute data from a certain location (minDNB):
ex) minDNB = '00','15','30','45','00','15',30','45' etc.
my goal is to replace each value or create a new matrix with fractions of an hour:
goal: newminDNB = 0.00, 0.25, 0.50, 0.75, 1.00, 1.25, 1.50, 1.75, 2.00, etc.
How should I go about this? I would like to loop through minDNB because some time stamps are missing and I need to retain gaps in the data. Thanks.
time00 = '00';
time25 = '15';
time50 = '30';
time75 = '45';
A = size(minDNB)
for i = 1:numel(minDNB)
if minDNB(i) == time00
A(i) = 0.00;
else if minDNB == time25
A(i) = 0.25 ;
else if minDNB(i) == time50
A(i) = 0.50 ;
else if minDNB(i) == time75
A(i) = 0.75 ;
end
end
end
end
end

Risposta accettata

Dave B
Dave B il 3 Nov 2021
Modificato: Dave B il 3 Nov 2021
I think your input is a cell array of character vectors:
minDNB={'00','15','30','45','00','15', '30','45'}
minDNB = 1×8 cell array
{'00'} {'15'} {'30'} {'45'} {'00'} {'15'} {'30'} {'45'}
and you want the output to be doubles, could you just do:
str2double(minDNB)/60
ans = 1×8
0 0.2500 0.5000 0.7500 0 0.2500 0.5000 0.7500
Or a fun alternative:
hours(minutes(str2double(minDNB)))
ans = 1×8
0 0.2500 0.5000 0.7500 0 0.2500 0.5000 0.7500
  4 Commenti
Nicholas Modar
Nicholas Modar il 3 Nov 2021
See attached. S is the structure with all station data. totalDNB is the full time stamp. minDNB and hourDNB are minute and hour seperated from the total date string.
The original data is in 15 minute intervals. My output needs to be in fractions of an hour. ( minute 15 -> minute 0.25)
My main point of confusion is how do I loop through the file the get 1) each fifteen minute interval converted to fractions of an hour and 2) how can I also keep the hour.
My desired output is a variable with size(minDNM). and somehow include the hours with minutes. (0.00 00.25 00.50 0.75 01.25 01.50 01.75 02.00...23.75 = 1 days worth of data. Let me know if I can explain further.
Dave B
Dave B il 3 Nov 2021
Modificato: Dave B il 3 Nov 2021
This is pretty similar to @Stephen's answer below but with a modification because your data are one big char and not a cell array of chars. I used double(string( for this but you could allso use str2double(cellstr(. And also I changed the < to a <= note this caveat:
if you're missing 1 or 2 consecutive measurements it's fine, if you want to increment the hour when you're missing 3 measurements (i.e. '00' '15' '30' '30' '45') then < should be <=, and if you're missing 4 measurements it's of course indistinguishable from just being the next hour!
mins=double(string(minDNB));
hrs = mins/60;
hrs=cumsum([0;diff(hrs)<0])+hrs;
But note: if you started with timeDNB, there's a way more natural path:
dt=datetime(timeDNB); % convert to datetime
hrs2=hours(dt-dt(1)); % or alternatively: [0;cumsum(hours(diff(dt)))]
This catches that you're missing measurements on March 11th (I think?)

Accedi per commentare.

Più risposte (1)

Stephen23
Stephen23 il 3 Nov 2021
C = {'00','15','30','45','00','15','30','45'};
V = str2double(C)/60;
V = V+cumsum([0,diff(V)<0])
V = 1×8
0 0.2500 0.5000 0.7500 1.0000 1.2500 1.5000 1.7500
  3 Commenti
Nicholas Modar
Nicholas Modar il 4 Nov 2021
Thanks everyone! Ended kind of combining everyones suggestions.
minDNB_fmt = datetime(dateDNB,'Format','mm');
hourDNB_fmt = datetime(dateDNB,'Format','HH');
minDNB = str2double(cellstr(minDNB_fmt));
hourDNB = str2double(cellstr(hourDNB_fmt));
duration_DNB = duration(hourDNB,minDNB,zeros(numel(minDNB),1));
[hh,mm] = hms(duration_DNB);
minutes_DNB = mm /60;
hourmin_DNB = horzcat(hh+minutes_DNB);
Dave B
Dave B il 4 Nov 2021
if you're just going through cellstrs to get from datetime to double, you can skip that step:
d=datetime(2010,1,1,10,30,00);
d.Hour
ans = 10
d.Minute
ans = 30

Accedi per commentare.

Categorie

Scopri di più su Dates and Time in Help Center e File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by