R - 使用自定义函数进行高效数据清理

Question

I am working on cleaning a dataset composed of 1M names. The cleaning is done by a function that includes around 40 greps such as name=gsub("Johnmichael", "John Michael",name,ignore.case=TRUE) and name=gsub("Mihcael", "Michael",name,ignore.case=TRUE)

I am currently using the cleaning function straight-up like this:

contacts$first_name=clean_name(contacts$first_name)

My issue is that my code is very slow since it applies the function to the whole vector one at a time. I am trying to find a way to use the function in parallel for each string, I have tried sapply but I do not seem to find any improvements. Any advice?

Answer 1

2

Install the OpenBLAS in R in Windows x64

Open the url http://sourceforge.net/projects/openblas/files/

Open the the latest version folder

download OpenBLAS-v0.2.13-Win64-int32.zip and mingw64_dll.zip

Unpack the "OpenBLAS-v0.2.13-Win64-int32.zip" find "libopenblas.dll" and rename this file to "Rblas.dll",copy the file to the path like this "\R\R-3.1.2\bin\x64"(Remember to backup) Unpack the "mingw64_dll.zip" and copy all the DLL to the same path "\R\R-3.1.2\bin\x64"