3 ms·
A lot of data scientists and analysts favour PostgreSQL for ETL and heavy duty data manipulation. PL/pgSQL lets you perform fairly complex tasks on datasets th
by mr_toad 8y ago
A lot of data scientists and analysts favour PostgreSQL for ETL and heavy duty data manipulation. PL/pgSQL lets you perform fairly complex tasks on datasets that may be too large to handle in memory (e.g too big for R and pandas).
And unlike a lot of ETL and reporting tools it doesn’t involve a lot of pointing and clicking. You can put PL/pgSQL in version control along with your R and Python code. That’s very handy for repeatability.