What’s new in 1.1.5 (December 07, 2020)¶
These are the changes in pandas 1.1.5. See Release notes for a full changelog including other versions of pandas.
Fixed regressions¶
Fixed regression in addition of a timedelta-like scalar to a
DatetimeIndex
raising incorrectly (GH37295)Fixed regression in
Series.groupby()
raising when theIndex
of theSeries
had a tuple as its name (GH37755)Fixed regression in
DataFrame.loc()
andSeries.loc()
for__setitem__
when one-dimensional tuple was given to select fromMultiIndex
(GH37711)Fixed regression in inplace operations on
Series
withExtensionDtype
with NumPy dtyped operand (GH37910)Fixed regression in metadata propagation for
groupby
iterator (GH37343)Fixed regression in
MultiIndex
constructed from aDatetimeIndex
not retaining frequency (GH35563)Fixed regression in
Index
constructor raising aAttributeError
when passed aSparseArray
with datetime64 values (GH35843)Fixed regression in
DataFrame.unstack()
with columns with integer dtype (GH37115)Fixed regression in indexing on a
Series
withCategoricalDtype
after unpickling (GH37631)Fixed regression in
DataFrame.groupby()
aggregation with out-of-bounds datetime objects in an object-dtype column (GH36003)Fixed regression in
df.groupby(..).rolling(..)
with the resultingMultiIndex
when grouping by a label that is in the index (GH37641)Fixed regression in
DataFrame.fillna()
not fillingNaN
after other operations such asDataFrame.pivot()
(GH36495).Fixed performance regression in
df.groupby(..).rolling(..)
(GH38038)Fixed regression in
MultiIndex.intersection()
returning duplicates when at least one of the indexes had duplicates (GH36915)Fixed regression in
GroupBy.first()
andGroupBy.last()
whereNone
was considered a non-NA value (GH38286)
Bug fixes¶
Bug in pytables methods in python 3.9 (GH38041)