You can clone with
HTTPS or Subversion.
If possible, call lapply directly from llply
Replace for loop with custom C code
Change to more functional style in llply in preparation for foreach
Slightly shorter llply benchmark and vis of progress
Add benchmark for straight llply
Add helper method for rapidly counting large data frames
A immutable data frame data structure.
The immutable data frame returns references to subsets, not copies. This makes certain operations much faster and simplifies the splitting code.
Treat drop consistently
Optimised code for splitting indices
Only store file name, not full path
Initial benchmarks for splitting
Improve speed of id by eliminating use of factor.
Still slow for strings
Two variable timings
Basic benchmarking infrastructure
Construct both character and factor variables