Finding unique dates in vector

Hi all.

I have a vector of dates where each date is repeated many times. How would i create a vector with every date but no repetitions efficiently?

Thx.

[Jan]

What exactly means "vector of dates"? Please post the type of the input exactly. I assume that UNIQUE will help you directly.

Answer 1

[Andrei Bobrov]

Date = ['02-Aug-2010 19:28:01'
'13-Nov-2010 09:10:20'
'10-Nov-2010 14:44:14'
'02-Aug-2010 19:28:01'
'08-Oct-2010 08:58:46'
'30-Jan-2010 08:24:16'
'10-Nov-2010 14:44:14'
'13-Nov-2010 09:10:20'
'08-Oct-2010 08:58:46']

[a n] = unique(datenum(Date));
D1 = sortrows([a n],2);
Dout = datestr(D1(:,1));

Hi Jan! agree with you. In our case, it is possible and so:

[a n] = unique(Date,'rows')
[~,idx] = sort(n);
Dout = a(idx,:);

[Jan]

What about "unique(cellstr(Date))"?

[Mads]

Thanks a lot.

[Andrei Bobrov]

Hi Jan! agree with you.

[Matt Tearle]

...assuming that the dates are all in the exact same format. Andrei's solution (going via datenum) avoids the problem of two different formats for the same date. (Unlikely, but it could happen.)

Answer 2

[Daniel Svedbrand]

unique_dates = unique(dateshift(Date,'start','day'));

if Date are datetimes, otherwise

unique_dates = unique(dateshift(datetime(Date),'start','day'));