dplyr_pandas.ipynb

Wonderful notebook! Helps me know R better. There might be something that will help others know pandas better.

Anyway, as in In[8], flights.iloc[:9] is fine. But wouldn't flights.loc[:10] be more equivalent?

And In[12]:

# select(flights, year:day) 

# No real equivalent here. Although I think this is OK.
# Typically I'll have the columns I want stored in a list
# somewhere, which can be passed right into __getitem__ ([]).

I'm only guessing what the R code does, but how about flights.loc[:, 'year':'day'] ? And if there's a list of columns to get, named columns for example, I think flights[columns] will do, other than __getitem__ ([]).

And In[13]:

# select(flights, -(year:day)) 
...

If I'm guessing right, flights.loc[:, ~flight.columns.isin(a.loc[:, 'year':'day'])] might do. But it's a little bit tedious and less elegant than R.

Again, a very helpful notebook.

For those who are struggling to translate your R code into python, you might want to take a look at this package:

https://github.com/pwwang/datar

which reimages pandas APIs and aligns them with tidyverse's.