Comments on: Faster csvwrite/dlmwrite https://undocumentedmatlab.com/articles/faster-csvwrite-dlmwrite?utm_source=rss&utm_medium=rss&utm_campaign=faster-csvwrite-dlmwrite Professional Matlab consulting, development and training Tue, 10 Oct 2017 23:51:29 +0000 hourly 1 https://wordpress.org/?v=6.7.3 By: Yair Altman https://undocumentedmatlab.com/articles/faster-csvwrite-dlmwrite#comment-414780 Tue, 10 Oct 2017 23:51:29 +0000 http://undocumentedmatlab.com/?p=7080#comment-414780 In reply to Saket.

@Saket – thanks for sharing, useful indeed

]]> By: Saket https://undocumentedmatlab.com/articles/faster-csvwrite-dlmwrite#comment-414481 Thu, 05 Oct 2017 07:52:58 +0000 http://undocumentedmatlab.com/?p=7080#comment-414481 @Andrew, @Yair – Try out the Mex Version of CSV Reader from Stanislaw Adaszewski: http://adared.ch/efficient-csv-reader-for-matlab

You might need to modify the C code as per your need and create a .mex file from the c-code. I have been using it for last 2 Years. This is quick – very quick..!

]]> By: Michal Kvasnicka https://undocumentedmatlab.com/articles/faster-csvwrite-dlmwrite#comment-414421 Wed, 04 Oct 2017 09:07:52 +0000 http://undocumentedmatlab.com/?p=7080#comment-414421 @Yair – thanks for very interesting post!

Any examples how to read/write HUGE text files (CSV) by fread+textscan in chunks?

]]> By: Yair Altman https://undocumentedmatlab.com/articles/faster-csvwrite-dlmwrite#comment-414392 Tue, 03 Oct 2017 18:09:18 +0000 http://undocumentedmatlab.com/?p=7080#comment-414392 In reply to AndrewV.

@Andrew – my advise is to read the entire CSV file into memory (a long string) using fread and then parse the string, for example using textscan. This works well with files that are not huge; with huge files you may want to read them in chunks.

]]> By: AndrewV https://undocumentedmatlab.com/articles/faster-csvwrite-dlmwrite#comment-414388 Tue, 03 Oct 2017 17:04:44 +0000 http://undocumentedmatlab.com/?p=7080#comment-414388 Any resources out there for faster csvread (or any other text parsing) functions? I believe there were a few on file exchange but I had no luck with them, either from minimal speed improvements or functions that just do not work.

]]>