What’s new in 1.4.1 (February 12, 2022) — pandas 2.2.3 documentation (original) (raw)

These are the changes in pandas 1.4.1. See Release notes for a full changelog including other versions of pandas.

Fixed regressions#

Regression in Series.mask() with inplace=True and PeriodDtype and an incompatible other coercing to a common dtype instead of raising (GH 45546)
Regression in assert_frame_equal() not respecting check_flags=False (GH 45554)
Regression in DataFrame.loc() raising ValueError when indexing (getting values) on a MultiIndex with one level (GH 45779)
Regression in Series.fillna() with downcast=False incorrectly downcasting object dtype (GH 45603)
Regression in api.types.is_bool_dtype() raising an AttributeError when evaluating a categorical Series (GH 45615)
Regression in DataFrame.iat() setting values leading to not propagating correctly in subsequent lookups (GH 45684)
Regression when setting values with DataFrame.loc() losing Index name if DataFrame was empty before (GH 45621)
Regression in join() with overlapping IntervalIndex raising an InvalidIndexError (GH 45661)
Regression when setting values with Series.loc() raising with all False indexer and Series on the right hand side (GH 45778)
Regression in read_sql() with a DBAPI2 connection that is not an instance of sqlite3.Connection incorrectly requiring SQLAlchemy be installed (GH 45660)
Regression in DateOffset when constructing with an integer argument with no keywords (e.g. pd.DateOffset(n)) would behave like datetime.timedelta(days=0) (GH 45643, GH 45890)

Fixed segfault in DataFrame.to_json() when dumping tz-aware datetimes in Python 3.10 (GH 42130)
Stopped emitting unnecessary FutureWarning in DataFrame.sort_values() with sparse columns (GH 45618)
Fixed window aggregations in DataFrame.rolling() and Series.rolling() to skip over unused elements (GH 45647)
Fixed builtin highlighters in Styler to be responsive to NA with nullable dtypes (GH 45804)
Bug in apply() with axis=1 raising an erroneous ValueError (GH 45912)

Reverted performance speedup of DataFrame.corr() for method=pearson to fix precision regression (GH 45640, GH 42761)

A total of 19 people contributed patches to this release. People with a “+” by their names contributed a patch for the first time.