Using datastore and tall arrays for files that are too big for memory
Show older comments
I have been looking at using datastore on files that have proven too much for matlab's memory. I was using fopen and textscan, fgetl, sscanf and dlmread. A recurring problem with each different command is memory. My data is saved in two different formats, an extremely large file containing 200 individual matrices and as the individual 200 matrices. When looking at creating tall arrays to fix the memory issues, I came across a problem. Which datastore command should I use? The individual files could be added to datastore using fileDatastore or the bulk file could use tabularTextDatastore but no matter which method i use, i get the same error:
Undefined function 'datastore' for input arguments of type 'char'.
I have tried multiple different methods of specifying the file\folder. I have used uigetfile and uigetdir to specify the file and I have also explicitly stated the file, either by setting the current folder and naming the file or with the 'C:\Users\c13459232\Documents....' method. They all return the same error. Is there something I am missing here?
3 Comments
Steven Lord
on 3 Jul 2017
What release of MATLAB are you using? The datastore function was introduced in release R2014b.
Aaron Smith
on 4 Jul 2017
Edited: Aaron Smith
on 4 Jul 2017
Jeremy Hughes
on 7 Jul 2017
Yes, tall arrays are available in R2016b.
Answers (0)
Categories
Find more on Large Files and Big Data in Help Center and File Exchange
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!