Where/mask methods for Series

[jreback]

add where and mask methods for Series, analagous to DataFrame methods added for 0.9.1
passes all tests

where is equivalent to: s[cond].reindex_like(s).fillna(other)

In [7]: s = pd.Series(np.random.rand(5))

In [8]: s
Out[8]: 
0    0.638664
1    0.574688
2    0.460510
3    0.641840
4    0.044129

In [10]: s[0:2] = -s[0:2]

In [11]: s
Out[11]: 
0   -0.638664
1   -0.574688
2    0.460510
3    0.641840
4    0.044129

boolean selection

In [12]: s[s>0]
Out[12]: 
2    0.460510
3    0.641840
4    0.044129

In [13]: s.where(s>0)
Out[13]: 
0         NaN
1         NaN
2    0.460510
3    0.641840
4    0.044129

In [14]: s.where(s>0,-s)
Out[14]: 
0    0.638664
1    0.574688
2    0.460510
3    0.641840
4    0.044129

In [15]: s.mask(s<=0)
Out[15]: 
0         NaN
1         NaN
2    0.460510
3    0.641840
4    0.044129

support setting as well (though not used anywhere explicity)

In [16]: s2 = s.copy()

In [17]: s2.where(s2>0,inplace=True)
Out[17]: 
0         NaN
1         NaN
2    0.460510
3    0.641840
4    0.044129

In [18]: s2
Out[18]: 
0         NaN
1         NaN
2    0.460510
3    0.641840
4    0.044129

[changhiskhan]

I cherry-picked this.
I changed the implementation of where a little bit as _set_value would have failed for non-scalar other.
I also added a few more test cases.

Thanks for the PR!

[jreback]

added docs for these in commit: 2d57979

[changhiskhan]

cherry-picked. Thank you!

[durden]

Are the where and mask methods supposed to be included in 0.9.1? I'm not seeing them, but maybe I've got something setup incorrectly?

>>> pandas.__version__
'0.9.1'
>>> df['col2'].where
Traceback (most recent call last):
  File "<ipython-input-19-885524901cf2>", line 1, in <module>
    df['col2'].where
AttributeError: 'Series' object has no attribute 'where'

>>> dir(df['col2'])
['T', '_AXIS_ALIASES', '_AXIS_NAMES', '_AXIS_NUMBERS', '__abs__', '__add__', '__and__', '__array__', '__array_finalize__', '__array_interface__', '__array_prepare__', '__array_priority__', '__array_struct__', '__array_wrap__', '__class__', '__contains__', '__copy__', '__deepcopy__', '__delattr__', '__delitem__', '__delslice__', '__dict__', '__div__', '__divmod__', '__doc__', '__eq__', '__float__', '__floordiv__', '__format__', '__ge__', '__getattribute__', '__getitem__', '__getslice__', '__gt__', '__hash__', '__hex__', '__iadd__', '__iand__', '__idiv__', '__ifloordiv__', '__ilshift__', '__imod__', '__imul__', '__index__', '__init__', '__int__', '__invert__', '__ior__', '__ipow__', '__irshift__', '__isub__', '__iter__', '__itruediv__', '__ixor__', '__le__', '__len__', '__long__', '__lshift__', '__lt__', '__mod__', '__module__', '__mul__', '__ne__', '__neg__', '__new__', '__nonzero__', '__oct__', '__or__', '__pos__', '__pow__', '__radd__', '__rand__', '__rdiv__', '__rdivmod__', '__reduce__', '__reduce_ex__', '__repr__', '__rfloordiv__', '__rlshift__', '__rmod__', '__rmul__', '__ror__', '__rpow__', '__rrshift__', '__rshift__', '__rsub__', '__rtruediv__', '__rxor__', '__setattr__', '__setitem__', '__setslice__', '__setstate__', '__sizeof__', '__str__', '__sub__', '__subclasshook__', '__truediv__', '__weakref__', '__xor__', '_agg_by_level', '_binop', '_can_hold_na', '_check_bool_indexer', '_constructor', '_get_axis', '_get_axis_name', '_get_axis_number', '_get_repr', '_get_val_at', '_get_values', '_get_values_tuple', '_get_with', '_index', '_ix', '_reindex_indexer', '_repr_footer', '_set_labels', '_set_values', '_set_with', '_tidy_repr', 'abs', 'add', 'align', 'all', 'any', 'append', 'apply', 'argmax', 'argmin', 'argsort', 'asfreq', 'asof', 'astype', 'at_time', 'autocorr', 'base', 'between', 'between_time', 'byteswap', 'choose', 'clip', 'clip_lower', 'clip_upper', 'combine', 'combine_first', 'compress', 'conj', 'conjugate', 'copy', 'corr', 'count', 'cov', 'ctypes', 'cummax', 'cummin', 'cumprod', 'cumsum', 'data', 'describe', 'diagonal', 'diff', 'div', 'dot', 'drop', 'dropna', 'dtype', 'dump', 'dumps', 'fill', 'fillna', 'first', 'first_valid_index', 'flags', 'flat', 'flatten', 'from_array', 'from_csv', 'get', 'get_value', 'getfield', 'groupby', 'head', 'hist', 'idxmax', 'idxmin', 'iget', 'iget_value', 'imag', 'index', 'interpolate', 'irow', 'isin', 'isnull', 'item', 'itemset', 'itemsize', 'iteritems', 'iterkv', 'ix', 'keys', 'kurt', 'last', 'last_valid_index', 'load', 'mad', 'map', 'max', 'mean', 'median', 'min', 'mul', 'name', 'nbytes', 'ndim', 'newbyteorder', 'nonzero', 'notnull', 'nunique', 'order', 'pct_change', 'plot', 'prod', 'ptp', 'put', 'quantile', 'rank', 'ravel', 'real', 'reindex', 'reindex_like', 'rename', 'reorder_levels', 'repeat', 'replace', 'resample', 'reset_index', 'reshape', 'resize', 'round', 'save', 'searchsorted', 'select', 'set_value', 'setasflat', 'setfield', 'setflags', 'shape', 'shift', 'size', 'skew', 'sort', 'sort_index', 'sortlevel', 'squeeze', 'std', 'str', 'strides', 'sub', 'sum', 'swapaxes', 'swaplevel', 'tail', 'take', 'to_csv', 'to_dict', 'to_sparse', 'to_string', 'tofile', 'tolist', 'tostring', 'trace', 'transpose', 'truncate', 'tshift', 'tz_convert', 'tz_localize', 'unique', 'unstack', 'update', 'valid', 'value_counts', 'values', 'var', 'view', 'weekday']

[jreback]

in 0.9.1 they only exist in a DataFrame
0.10.0 will support Series as well

docs will explain the slight difference

eg

s[cond] will return a subset of rows
but df[cond] returns a same sized frame
if s is a series and df is a frame

On Nov 28, 2012, at 3:22 PM, Luke Lee notifications@github.com wrote:

Are the where and mask methods supposed to be included in 0.9.1? I'm not seeing them, but maybe I've got something setup incorrectly?

pandas.version
'0.9.1'
df['col2'].where
Traceback (most recent call last):
File "", line 1, in
df['col2'].where
AttributeError: 'Series' object has no attribute 'where'
dir(df['col2'])
['T', '_AXIS_ALIASES', '_AXIS_NAMES', '_AXIS_NUMBERS', 'abs', 'add', 'and', 'array', 'array_finalize', 'array_interface', 'array_prepare', 'array_priority', 'array_struct', 'array_wrap', 'class', 'contains'
, 'copy', 'deepcopy', 'delattr', 'delitem', 'delslice', 'dict', 'div', 'divmod', 'doc', 'eq', 'float', 'floordiv', 'format', 'ge', 'getattribute', 'getitem', 'getslice', 'gt', 'hash', 'hex', 'iadd', 'iand', 'idiv', 'ifloordiv', 'ilshift', 'imod', 'imul', 'index', 'init', 'int', 'invert', 'ior', 'ipow', <
span class="s">'irshift', 'isub', 'iter', 'itruediv', 'ixor', 'le', 'len', 'long', 'lshift', 'lt', 'mod', 'module', 'mul', 'ne', 'neg', 'new', 'nonzero', <
span class="s">'oct', 'or', 'pos', 'pow', 'radd', 'rand', 'rdiv', 'rdivmod', 'reduce', 'reduce_ex', 'repr', 'rfloordiv', 'rlshift', 'rmod', 'rmul', 'ror', 'rpow',
'rrshift', 'rshift', 'rsub', 'rtruediv', 'rxor', 'setattr', 'setitem', 'setslice', 'setstate', 'sizeof', 'str', 'sub', 'subclasshook', 'truediv', 'weakref', 'xor', '
_agg_by_level', '_binop', '_can_hold_na', '_check_bool_indexer', '_constructor', '_get_axis', '_get_axis_name', '_get_axis_number', '_get_repr', '_get_val_at', '_get_values', '_get_values_tuple', '_get_with', '_index', '_ix', '_reindex_indexer', 'repr
footer', '_set_labels', '_set_values', '_set_with', '_tidy_repr', 'abs', 'add', 'align', 'all', 'any', 'append', 'apply', 'argmax', 'argmin', 'argsort', 'asfreq', 'asof', 'astype', 'at_time', 'autocorr', 'base', 'between', 'between_time', 'byteswap', 'choose', 'clip', 'clip_lower', 'clip_upper', 'combine', 'combine_first', 'compress', 'conj', 'conjugate', 'copy', 'corr', 'count', 'cov', 'ctypes', 'cummax', 'cummin', 'cumprod', 'cumsum', 'data', 'describe', 'diagonal', 'diff', 'div', 'dot', 'drop', 'dropna', 'dtype', 'dump', 'dumps', 'fill', 'fillna', 'first', 'first_valid_index', 'flags', 'flat', 'flatten', 'from_array', 'from_csv', 'get', 'get_value', 'getfield', 'groupby', 'head', 'hist', 'idxmax', 'idxmin', 'iget', 'iget_value', 'imag', 'index', 'interpolate', 'irow', 'isin', 'isnull', 'item', 'itemset', 'itemsize', 'iteritems', 'iterkv', 'ix', 'keys', 'kurt', 'last', 'last_valid_index', 'load', 'mad', 'map', 'max', 'mean', 'median', 'min', 'mul', 'name', 'nbytes', 'ndim', 'newbyteorder', 'nonzero', 'notnull', 'nunique', 'order', 'pct_change', 'plot', 'prod', 'ptp', 'put', 'quantile', 'rank', 'ravel', 'real', 'reindex', 'reindex_like', 'rename', 'reorder_levels', 'repeat', 'replace', 'resample', 'reset_index', 'reshape', 'resize', 'round', 'save', 'searchsorted', 'select', 'set_value', 'setasflat', 'setfield', 'setflags', 'shape', 'shift', 'size', 'skew', 'sort', 'sort_index', 'sortlevel', 'squeeze', 'std', 'str', <
span class="s">'strides', 'sub', 'sum', 'swapaxes', 'swaplevel', 'tail', 'take', 'to_csv', 'to_dict', 'to_sparse', 'to_string', 'tofile', 'tolist', 'tostring', 'trace', 'transpose', 'truncate', 'tshift', 'tz_convert', 'tz_localize', 'unique', 'unstack', 'update', 'valid', 'value_counts', 'values', 'var', 'view', 'weekday']

—
Reply to this email directly or view it on GitHub.

[durden]

@jreback Ah, that clears up the confusion. Thanks!