bigdata - Reading big data and logistic regression in R -


Status: 1 GB CSV file, 100000 rows, 4000 independent digit variables, 1 dependent variable. On windows citrix server, with 16 GB memory

Problem: I took 2 hours! To:

  read.table ("full_data.csv", header = t, sp ",")   

and glm process crashes, programs Not responding, and I have to close it in Task Manager.

I often use the package sqldf to load larger .csv in memory Is a good indicator.

Comments