0

I have a large CSV dataset (>2GB) but I would only like to read in a small subset where the first column = 'a' for example into R. How do I do this quickly, hopefully without having to read the whole dataset in? I don't mind subsetting with a large text editor/command line first if that is what's necessary. As long it's quick.

Vykta Wakandigara
  • 339
  • 2
  • 10
  • You can with `read.csv.sql` from the `sqldf` package, see here: https://stackoverflow.com/questions/35847044/specific-rows-in-fread, but I don't think you get performance bonuses over just using `fread` from the `data.table` package and then subsetting, see here: https://stackoverflow.com/questions/23502974/using-fread-to-select-rows-and-columns-the-way-read-csv-sql-does. – caldwellst Oct 06 '19 at 10:06
  • I think the fread option appears to be better and faster. Thanks! – Vykta Wakandigara Oct 06 '19 at 11:02

0 Answers0