python - How do Rpy2, pyrserve and PypeR compare? -
i access r within python program. aware of rpy2, pyrserve , pyper.
what advantages or disadvantages of these 3 options?
i know 1 of 3 better others, in order given in question:
rpy2:
- c-level interface between python , r (r running embedded process)
- r objects exposed python without need copy data over
- conversely, python's numpy arrays can exposed r without making copy
- low-level interface (close r c-api) , high-level interface (for convenience)
- in-place modification vectors , arrays possible
- r callback functions can implemented in python
- possible have anonymous r objects python label
- python pickling possible
- full customization of r's behavior console (so possible implement full r gui)
- mswindows limited support
pyrserve:
- native python code (will/should/may work cpython, jython, ironpython)
- use r's rserve
- advantages , inconveniences linked remote computation , rserve
pyper:
- native python code (will/should/may work cpython, jython, ironpython)
- use of pipes have python communicate r (with advantages , inconveniences linked it)
edit: windows support rpy2
Comments
Post a Comment