This change has been reverted in pandas 1.2.1, and the behaviour to not align DataFrames
an() and Rolling.sum() use Kahan summation to calculate the mean to avoid numerical problems ( GH10319, GH11645, GH13254, GH32761, GH36031)ĭarchsorted(), archsorted(), archsorted(), and archsorted() with datetime-like dtypes will now try to cast string arguments (list-like and scalar) to the matching datetime-like type ( GH36346)Īdded methods IntegerArray.prod(), IntegerArray.min(), and IntegerArray.max() ( GH33790)Ĭalling a NumPy ufunc on a DataFrame with extension types now preserves the extension types when possible ( GH23743)Ĭalling a binary-input NumPy ufunc on multiple DataFrame objects now aligns, matching the behavior of binary operations and ufuncs on Series ( GH23743). Styler now allows direct CSS class name addition to individual data cells ( GH36159) t_table_styles() now allows the direct styling of rows and columns and can be chained ( GH35607) Io.sql.get_schema() now supports a schema keyword argument that will add a schema into the create table statement ( GH28486)ĭataFrame.explode() and Series.explode() now support exploding of sets ( GH35614)ĭataFrame.hist() now supports time series (datetime) data ( GH32590) Index with object dtype supports division and multiplication ( GH34160) Other enhancements ¶Īdded day_of_week (compatibility alias dayofweek) property to Timestamp, DatetimeIndex, Period, PeriodIndex ( GH9605)Īdded day_of_year (compatibility alias dayofyear) property to Timestamp, DatetimeIndex, Period, PeriodIndex ( GH9605)Īdded set_flags() for setting table-wide flags on a Series or DataFrame ( GH28394)ĭataFrame.applymap() now supports na_action ( GH23803) Numba must be installed as an optional dependency DataFrame () In : df Out: A B 0 a 0 1 b 1 2 a 2 3 b 3 In : df. If pandas does not automatically detect whether the file handle is opened in binary or text mode, With encoding ( GH13068 and GH23854) and compression ( GH22555). To_csv() supports file handles in binary mode ( GH19827 and GH35058) Support for binary file handles in to_csv ¶ GH35655 added fsspec support (including storage_options)įor reading excel files. Of the individual storage backends (detailed from the fsspec docs forīuiltin implementations and linked to external ones). Parameters can be passed to which backends can be found in the documentation
This allows, forĮxample, for passing credentials to S3 and GCS storage. To pass a dictionary of parameters to the storage backend. Many read/write functions have acquired the storage_options optional argument, Thisĭiffers from DataFrame.attrs, which stores information that applies to
This stores global attributes that apply to the pandas object. The allows_duplicate_labels flag is stored in the new DataFrame.flagsĪttribute. It is expected that every method taking or returning one or moreĭataFrame or Series objects will propagate allows_duplicate_labels. Propagate the allows_duplicate_labels value. Prevent accidental introduction of duplicate labels, which can affect downstream operations.īy default, duplicates continue to be allowed. Series and DataFrame can now be created with allows_duplicate_labels=False flag toĬontrol whether the index or columns can contain duplicate labels ( GH28394). Enhancements ¶ Optionally disallow duplicate labels ¶ xlsx files using the openpyxl engine instead. It can be globally set and the warning suppressed.
While this option is now deprecated and will also raise a FutureWarning, This is no longer supported, switch to using openpyxl instead.Īttempting to use the xlwt engine will raise a FutureWarning Please do not report issues when using ``xlrd`` to read ``.xlsx`` files.
Thus, it is strongly encouraged to install openpyxl to read Excel 2007+ See the read_excel() documentation for more details. Many of these cases will now default to using the openpyxl engine. Would result in using the xlrd engine in many cases, including new Previously, the default argument engine=None to read_excel()