x[["a"]][[1]] is equivalent to zip_n(x)[[1]][["a"]], in the same way z[a,b] is equivalent to t(z)[b, a].
That suggests there might be interesting generalisations of zip_n() that correspond to aperm() - i.e. maybe you want to switch the order of the 2nd and 3rd depths. What is the data frame equivalent of this operation? select()?
I think we can probably remove zip2() and zip3() as they don't seem that useful (since they're not symmetric)
Haskell's transpose: https://hackage.haskell.org/package/base-4.8.1.0/docs/Data-List.html#v:transpose
@jennybc
x[["a"]][[1]]is equivalent tozip_n(x)[[1]][["a"]], in the same wayz[a,b]is equivalent tot(z)[b, a].That suggests there might be interesting generalisations of
zip_n()that correspond toaperm()- i.e. maybe you want to switch the order of the 2nd and 3rd depths. What is the data frame equivalent of this operation?select()?I think we can probably remove
zip2()andzip3()as they don't seem that useful (since they're not symmetric)Haskell's transpose: https://hackage.haskell.org/package/base-4.8.1.0/docs/Data-List.html#v:transpose
@jennybc