Hacker News Reader

HNHacker News

Top New Best Ask Show Jobs

Comment by sandGorgon | Hacker News Reader

Full thread

sandGorgon·

we have recently started migrating to python from R because of notebooks and pyspark.

I dont see us moving back anytime soon, because production code in python is orders of magnitude better than R.

I just wish there was a decent dplyr for python though :(

View on HN

There is a HUGE difference between running production code and exploratory analysis. Python with Jupyter (however good in production they are) are just not fit for that task. After all, Python is a general-purpose programming language and R is largely a DSL designed specifically for figuring things out as quick as possible. And R is really good at it.

Not entirely true - I can understand you like R, but Pandas+Numpy is fantastic.

"DSL for figuring out things aa quick as possible " is a religious statement. I suggest you spend some time on the Python side to understand how good it is.

However I can quantify some of what you said - that tge ecosystem of analytics libraries is bigger in CRAN. I agree with that... however at this point, the only library I really miss is dplyr.

Already doing both R and Python in equal proportions, thank you for the suggestion anyways.

I haven't used dplyr much, is is not similar to pandas?

I think you should read the discussion here - https://news.ycombinator.com/item?id=11335491

It explains it better than i can

dplyr abstracts away so much bullshit... they're not even playing the same sport. Pandas brought R-style data.frames to Python; dplyr makes databases, data.tables, etc. all behave the same. Out-of-memory, in-memory, all the abstractions just work the same.

Oh and it's fast as shit (the tight loops are in C++).

Reply on news.ycombinator.com