How do I create a For loop from a string array that will fill and load h5 files?

10 次查看（过去 30 天）
GDT
GDT 2023-4-12
评论： GDT 2023-4-15
Hello,
I'm working with .h5 files and trying to read them in and export accelerometer data from them for each trial. Every participants has differing amount of trials that they were able to conduct, so I use ls to identify the trials within a subject directory. The problem arises because each .h5 file is named after the day, time and task that it was collected on (example "20230313-085403_Walk.h5"). The entire name of the file is needed to load the h5 files but for some reason I can't wrap my head around this part with h5read. This can range from 5 trials to 30 trials. Any help would be greatly appreciated!
Code:
Trials = ls('*Walk.h5'); % could change ls to DIRFF, but we should be in that folder.
n_Trials=size(Trials,1);
%%% idenitfying all Walk.h5 files in subject directory, creating a vector the size of how many trials there are
for i = 1:length(n_Trials)
%% Loops through as many times there are trials.
Head = h5read({i}, '/Sensors/10075/Accelerometer');
Chest = h5read({i}, '/Sensors/10235/Accelerometer');
Lumbar = h5read('{i}', '/Sensors/10008/Accelerometer');
Right_Foot = h5read('{i}', '/Sensors/10152/Accelerometer');
Left_Foot = h5read('{i}', '/Sensors/10156/Accelerometer');
Right_Wrist = h5read('{i}', '/Sensors/10134/Accelerometer');
%%% Below is unrelated to issue, but to help show where its going.
x = Head(1,385:end);
y = Head(2,385:end);
z = Head(3,385:end);
res1 = sqrt((x.^2)+(y.^2)+(z.^2));
x2 = Chest(1,385:end);
y2 = Chest(2,385:end);
z2 = Chest(3,385:end);
res2 = sqrt((x2.^2)+(y2.^2)+(z2.^2));
x3 = Lumbar(1,385:end);
y3 = Lumbar(2,385:end);
z3 = Lumbar(3,385:end);
res3 = sqrt((x3.^2)+(y3.^2)+(z3.^2));
res_total = padcat(res1, res2, res3);
end
Any help would be greatly appreciated! My apologies if this has been addressed somewhere, but I can't quite find it with the search terms.
  2 个评论
Stephen23
Stephen23 2023-4-12
Do not use LS. Use DIR, as the documentation shows:
https://www.mathworks.com/help/matlab/import_export/process-a-sequence-of-files.html
"but we should be in that folder."
Rather than restricting your user like that, a much better approach is to use absolute/relative filenames:
https://www.mathworks.com/help/matlab/ref/fullfile.html
GDT
GDT 2023-4-12
Thank you, I'll try DIR instead. I do use fullfile to navigate to the subject directory for each subject/timepoint.

请先登录，再进行评论。

请先登录，再回答此问题。

采纳的回答

dpb
dpb 2023-4-13
I didn't see @Stephen23 had already addressed the main iissue, but had already started, so...
d=dir('*Walk.h5');
for i = 1:numel(d)
Head = h5read(fullfile(d(i).folder,d(i).name), '/Sensors/10075/Accelerometer');
...
Looks like you could read the metadata from the file first to determine the sizes and then save some manipulations later...as a general rule, don't bury "magic numbers" like the 385 value inside code; if must use absolute addressing (really should try to avoid this), then at least put the number in a variable where it can be fixed in one place instead of scattered all throughout the code.
  1 个评论
GDT
GDT 2023-4-15
Thank you, using DIR and the lower portion of code from the link below worked perfectly. Also a good point on the number. I was trying to use it to remove the first three seconds of a trial, but you're right it makes way more sense to at least make it a variable.
https://www.mathworks.com/help/matlab/import_export/process-a-sequence-of-files.html

请先登录，再进行评论。

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 File Operations 的更多信息

标签

产品


版本

R2019a

Community Treasure Hunt

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

Start Hunting!

Translated by