Guide · updated 2026-11-04
pandas nullable dtypes and explicit parsing
Int64, boolean and string dtypes, and why default inference changes your data.
Why int becomes float
A classic int64 column becomes float64 when one value is missing, so 7 becomes 7.0. Nullable Int64 keeps integers exact with <NA>.
Explicit parsing
Use pd.to_numeric(errors="coerce") or explicit formats for dates, then quarantine values that became missing during parsing but weren't missing before.
Guard your merges
Use validate= on merges and remember that pandas matches missing keys to each other.