how i do loop over column?

hi all,

I have a matrix with the dim of 978 * 744 which I want to loop over a number of a column with time step equal to 168 (hr/week) to have the final matrices of each week separately with the same number of rows 978

could anyone help me with, please?

thank you in advance.

[KSSV]

Read about reshape

[Lilya]

KSSV, thank you for your response. but I cant get the result

Answer 1

[KL]

If you want to do time-based calculation with your data, for example, weekly average or something I'd highly recommend using timetables (version 16b and above). Even for versions a bit older than that, use tables with datevec and then varfun.

Take a look at the examples on this page: https://www.mathworks.com/help/matlab/timetables.html

[Lilya]

KL, thank you for your answer. Yes, I did. the point is: this function calculating the weekly mean that I don't need it I just need the all values in each weed and each has a separet matrix.

[Guillaume]

@KL, I would use datetime which is a lot more powerful than datevec.

@Lina, with timetables you can calculate, weekly, monthly, hourtly, quaterly, etc. anything, not just mean. One thing for sure, you do not want to split weeks in separate matrices. Whatever it is you're trying to do, it will be harder once you've split your original data into several matrices.

If you give more details about what you're doing, we can show you how to do it simply.

[Lilya]

thanks for the clarification Guillaume :).

ok. as you know the number of weeks in each month are different regarding the number of days. so I decided to 'trim' the data like just to have 4 weeks in each month therefore the original matrix (978*744 for example) should give 4 weeks each has (978*168) with neglecting the remain couple days.

I hope this clear now

[KL]

@Guillaume: hmm, I kind of thought I might need datevec components to group but yeah, I agree, using proper datetime format is a better approach.

...I decided to 'trim' the data ...

@Lina: Don't do that. You don't have to lose any data. Simply import your data using readtable

 data = readtable(filename);

convert it to timetable,

 TT = table2timetable(data,'rowtimes',here goes your datetime column)

then use retime,

 TT_weekly = retime(TT,'weekly','mean')

[Lilya]

Fantastic * _ * ... thanks so much!