Moving from R to Python: The Libraries You Need to Know

One quick nitpick:

> glm, knn, randomForest, e701 (yes, this is actually a meaningful package's name)

e701 is not a meaningful package's name. e1071 is! e1071 is a package Misc Functions of the Department of Statistics, Probability Theory Group (Formerly: E1071), Technische Universität Wien.

I've never heard of pandasql before. Looks very intriguing. I'm always looking up "how to do this sql thing in pandas" or similar cheatsheets when working with pandas.

big fan of the yhat blog.

> reshape/reshape2,pylr/dplyr -> pandas

R users have a choice here, so this line could be

data.table -> pandas

and in fact data.table has a significant performance advantage compared to the reshape/reshape2,plyr/dplyr option