Changeset 1528

Show
Ignore:
Timestamp:
10/08/08 12:29:25 (2 months ago)
Author:
mattknox_ca
Message:

Some doc cleanup

Files:

Legend:

Unmodified
Added
Removed
Modified
Copied
Moved

  • trunk/timeseries/scikits/timeseries/doc/source/core/TimeSeries.rst

    r1521 r1528  
    367367.. autofunction:: fill_missing_dates 
    368368 
     369.. autofunction:: first_unmasked_val 
     370.. autofunction:: last_unmasked_val 
     371 
    369372.. autofunction:: empty_like 
    370373 

  • trunk/timeseries/scikits/timeseries/tseries.py

    r1518 r1528  
    9494 
    9595def last_unmasked_val(marray): 
    96     """Retrieve the last unmasked value in a 1d MaskedArray. 
     96    """ 
     97    Retrieve the last unmasked value in a 1d MaskedArray. 
    9798 
    9899    Parameters 
    …… 
    244245    Raises 
    245246    ------ 
    246         A TimeSeriesCompatibilityError exception is raised if something goes 
    247         wrong. 
     247    A TimeSeriesCompatibilityError exception is raised if something goes 
     248    wrong. 
    248249 
    249250    """ 
    …… 
    910911    Returns 
    911912    ------- 
    912     TimeSeries: 
    913 <<<<<<< .mine 
    914         A new TimeSeries with the `.dates` attribute at the specified frequency 
    915         (the :meth:`.asfreq` method of the :attr:`.dates` 
    916 ======= 
    917         A new TimeSeries with the :attr:`.dates` :class:`DateArray` at the 
    918         specified frequency (the :meth`.asfreq` method of the :attr:`.dates` 
    919 >>>>>>> .r1253 
    920         property will be called). 
    921         The data in the resulting series will be a VIEW of the original series. 
     913    A new TimeSeries with the :attr:`.dates` :class:`DateArray` at the 
     914    specified frequency (the :meth`.asfreq` method of the :attr:`.dates` 
     915    property will be called). 
     916    The data in the resulting series will be a VIEW of the original series. 
    922917 
    923918    Notes 
    924919    ----- 
    925 <<<<<<< .mine 
    926920    The parameters are the exact same as for 
    927     :meth:`~scikit.timeseries.tdates.DateArray.asfreq`. 
    928     Please see the docstring for that method for details on the parameters and 
    929 ======= 
    930     The parameters are the exact same as for :meth:`DateArray.asfreq`, please 
    931     see the `__doc__` string for that method for details on the parameters and 
    932 >>>>>>> .r1253 
    933     how the actual conversion is performed. 
     921    :meth:`~scikit.timeseries.DateArray.asfreq`. Please see the docstring for 
     922    that method for details on the parameters and how the actual conversion is 
     923    performed. 
    934924 
    935925    """ 
    …… 
    16071597    Converts a series from one frequency to another, by manipulating both the 
    16081598    `data` and `dates` attributes. 
    1609     The input series should not have any missing nor duplicated dates. 
    1610  
    1611     When converting from one frequency to a lower one, use the ``func`` parameter 
    1612     to control how data sharing the same new date should be handled. For example, 
    1613     when converting a daily series to a monthly series, use ``numpy.ma.mean`` 
    1614     to get a series of monthly averages. 
    1615     If ``func`` is not given, the output series group the points of the initial 
    1616     series that share the same new date. For example, if the initial series has 
    1617     a daily frequency and is 1D, the output series is 2D. 
    1618  
    1619     When converting to a higher frequency, use the ``position`` parameter to 
    1620     determine where the points should fall in the new period. 
    1621     For example, when converting a monthly series to daily, set ``precision`` 
    1622     to ``'START'`` to force the points to fall on the first of each month. 
     1599 
     1600    If the input series has any missing dates, it will first be filled in with 
     1601    masked values prior to doing the conversion. 
    16231602 
    16241603    Parameters 
    …… 
    16301609        Frequency to convert the TimeSeries to. Accepts any valid frequency 
    16311610        specification (string or integer) 
    1632     func : {None,function}, optional 
    1633         Function controlling how data sharing the same new dates should be 
    1634         manipulated. 
    1635         This function should handle masked values appropriately. 
    1636         If ``func`` is None (default), the output series groups the points of the 
    1637         initial series that share the same new date. 
    1638         The parameter is used only when converting to a lower frequency. 
     1611    func : function, optional 
     1612        When converting a series to a lower frequency, you can use the 
     1613        ``func`` parameter to perform a calculation on each period of values 
     1614        to aggregate results. 
     1615        For example, when converting a daily series to a monthly series, use 
     1616        :func:`numpy.ma.mean` to get a series of monthly averages. If you wish 
     1617        to get the first or last value from a period, use the functions 
     1618        :func:`scikits.timeseries.first_unmasked_val` and 
     1619        :func:`scikits.timeseries.last_unmasked_val`. 
     1620        If ``func`` is not given, the output series group the points of the 
     1621        initial series that share the same new date. For example, if the 
     1622        initial series has a daily frequency and is 1D, the output series is 
     1623        2D. 
    16391624    position : {'END', 'START'}, optional 
    1640         Determines whether the points should fall at the beginning (``'START'``) 
    1641         or at the end (``'END'``) of the new period. 
     1625        When converting a series to a higher frequency, use this parameter to 
     1626        determine where the points should fall in the new period. 
     1627        For example, when converting a monthly series to daily, specifying 
     1628        position='START' will cause the values to fall on the first day of 
     1629        each month (with all other values being masked). 
    16421630    *args : {extra arguments for func parameter}, optional 
    16431631        Additional mandatory parameters of the ``func`` function.