News for hackers. Stuff that matters.
Re:A Roadmap for Rich Scientific Data Structures in Python by rcthompson on Thursday July 21, 2011 @04:26PM (#2791152)
Data frames (and the associated functionaly for reading and writing csv files) are one of the main features that make me use R over python. Whenever I need to manipulate, merge, slice, and dice table-like data, I turn to R. I would love to have something feature-equivalent in Python. Although maybe rpy2 and rnumpy are sufficient for now.
Re:A Roadmap for Rich Scientific Data Structures in Python by wesm on Thursday July 21, 2011 @04:54PM (#2791252)
It is my project but I think pandas.DataFrame is already at 90+% feature-equivalency (especially if you're using the current git version...new release forthcoming). You don't have the integration with a million CRAN libraries. pandas.DataFrame actually does a lot more for you in many places than data.frame does-- for example data alignment is deeply intrinsic whereas it's very much a DIY affair in R.
Re:A Roadmap for Rich Scientific Data Structures in Python by rcthompson on Wednesday August 03, 2011 @07:45PM (#2843757)
Yes, I do intend to try out pandas at some point.