What’s new in 1.1.3 (October 5, 2020)¶
These are the changes in pandas 1.1.3. See Release notes for a full changelog including other versions of pandas.
Enhancements¶
Added support for new Python version¶
pandas 1.1.3 now supports Python 3.9 (GH36296).
Development Changes¶
The minimum version of Cython is now the most recent bug-fix version (0.29.21) (GH36296).
Fixed regressions¶
Fixed regression in
DataFrame.agg()
,DataFrame.apply()
,Series.agg()
, andSeries.apply()
where internal suffix is exposed to the users when no relabelling is applied (GH36189)Fixed regression in
IntegerArray
unary plus and minus operations raising aTypeError
(GH36063)Fixed regression when adding a
timedelta_range()
to aTimestamp
raised aValueError
(GH35897)Fixed regression in
Series.__getitem__()
incorrectly raising when the input was a tuple (GH35534)Fixed regression in
Series.__getitem__()
incorrectly raising when the input was a frozenset (GH35747)Fixed regression in modulo of
Index
,Series
andDataFrame
usingnumexpr
using C not Python semantics (GH36047, GH36526)Fixed regression in
read_excel()
withengine="odf"
causedUnboundLocalError
in some cases where cells had nested child nodes (GH36122, GH35802)Fixed regression in
DataFrame.replace()
inconsistent replace when using a float in the replace method (GH35376)Fixed regression in
Series.loc()
on aSeries
with aMultiIndex
containingTimestamp
raisingInvalidIndexError
(GH35858)Fixed regression in
DataFrame
andSeries
comparisons between numeric arrays and strings (GH35700, GH36377)Fixed regression in
DataFrame.apply()
withraw=True
and user-function returning string (GH35940)Fixed regression when setting empty
DataFrame
column to aSeries
in preserving name of index in frame (GH36527)Fixed regression in
Period
incorrect value for ordinal over the maximum timestamp (GH36430)Fixed regression in
read_table()
raisedValueError
whendelim_whitespace
was set toTrue
(GH35958)Fixed regression in
Series.dt.normalize()
when normalizing pre-epoch dates the result was shifted one day (GH36294)
Bug fixes¶
Bug in
read_spss()
where passing apathlib.Path
aspath
would raise aTypeError
(GH33666)Bug in
Series.str.startswith()
andSeries.str.endswith()
withcategory
dtype not propagatingna
parameter (GH36241)Bug in
Series
constructor where integer overflow would occur for sufficiently large scalar inputs when an index was provided (GH36291)Bug in
DataFrame.sort_values()
raising anAttributeError
when sorting on a key that casts column to categorical dtype (GH36383)Bug in
DataFrame.stack()
raising aValueError
when stackingMultiIndex
columns based on position when the levels had duplicate names (GH36353)Bug in
Series.astype()
showing too much precision when casting fromnp.float32
to string dtype (GH36451)Bug in
Series.isin()
andDataFrame.isin()
when usingNaN
and a row length above 1,000,000 (GH22205)Bug in
cut()
raising aValueError
when passed aSeries
of labels withordered=False
(GH36603)
Other¶
Reverted enhancement added in pandas-1.1.0 where
timedelta_range()
infers a frequency when passedstart
,stop
, andperiods
(GH32377)