Deriving specific rows from a large text files

4 次查看（过去 30 天）
lostatsea
lostatsea 2018-10-28
评论： dpb 2018-11-4
Hello! I am very new at MATLAB and am still learning. I have a 66 million row dataset in the form of a text file. I need to pull certain rows from the list to create a new list. Specifically, the data presents as:
Time,Pressure,Sea pressure,Depth
2018-08-25 16:10:26.000,10.2011833,0.0686833,0.0681232
The data isn't in separate columns, it's comma separated (except date and time, which are separated by a space). The new list needs to contain anything with a time stamp that ends in milliseconds of:
.000
.125
.375
.500
.625
.750
.875
I am not sure what the best way would be to address this. Any help would be greatly appreciated!
  14 个评论
显示 12更早的评论
lostatsea
lostatsea 2018-10-28
Thanks Jonas, I have tried adjusting the code to this:
>> TT1 = readtable('sample1.txt');
TT2 = etime(TT1,'regular', 'TimeStep', seconds(1/8));
Error using etime
Too many input arguments.
dpb
dpb 2018-10-28
The function Walter used is retime not etime.

请先登录，再进行评论。

请先登录，再回答此问题。

采纳的回答

Akira Agata
Akira Agata 2018-10-29
I would recommend updating your MATLAB to the latest version, since R2015a does not support many useful functions for your task, such as retime. If you have to do your task in old MATLAB for some reasons, the following would be one possible solution.
T = readtable('sample1.txt',...
'ReadVariableNames', false,...
'HeaderLines', 1);
T.Properties.VariableNames = {'Time','Pressure','SeaPressure','Depth'};
c = regexp(T.Time,'(000|125|250|375|500|625|750|875)$','match');
idx = cellfun(@isempty,c);
T = T(~idx,:);
  1 个评论
lostatsea
lostatsea 2018-11-1
Thank you Akira! This worked perfectly. I am taking your suggestion and will be updating our MATLAB. Thank you again for your help!

请先登录，再进行评论。

更多回答（1 个）

dpb
dpb 2018-10-28
Read as table or timetable; deal with the asterisks as needs must depending on whether they're real or not...
Two alternatives, retime to a specific vector means making a new time vector; the alternative would be to find the locations matching the desired times..
t=readtable('yourfile');
ix=ismember((t.Time-fix(t.Time))*1000,[0:7]*125)); % locate millisec multiples of 125
t=t(ix,:); % save those into the desired table
  15 个评论
显示 13更早的评论
Walter Roberson
Walter Roberson 2018-11-3
t = readtable('sample1.txt'); %up to R2018b still cannot autodetect times with fractions of a second, even if detectImportOptions is used
tt = datetime(t.Time, 'inputformat', 'yyyy-MM-dd HH:mm:ss.SSS');
s = mod(second(tt),1) * 1000;
mask = ismember(s, 0:125:875);
selected = t(mask,:);
However, with that particular data, this can be shorted to
t = readtable('sample1.txt');
selected = t(1:2:end, :);

请先登录，再进行评论。

类别

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

标签

Community Treasure Hunt

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

Start Hunting!

Translated by