Source code for astropy.stats.sigma_clipping

# Licensed under a 3-clause BSD style license - see LICENSE.rst

from __future__ import (absolute_import, division, print_function,
                        unicode_literals)
import numpy as np
import warnings
from ..utils.exceptions import AstropyUserWarning
from ..extern.six.moves import range


__all__ = ['SigmaClip', 'sigma_clip', 'sigma_clipped_stats']


[docs]class SigmaClip(object): """ Class to perform sigma clipping. The data will be iterated over, each time rejecting points that are discrepant by more than a specified number of standard deviations from a center value. If the data contains invalid values (NaNs or infs), they are automatically masked before performing the sigma clipping. For a functional interface to sigma clipping, see :func:`sigma_clip`. .. note:: `scipy.stats.sigmaclip <http://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.sigmaclip.html>`_ provides a subset of the functionality in this class. Parameters ---------- sigma : float, optional The number of standard deviations to use for both the lower and upper clipping limit. These limits are overridden by ``sigma_lower`` and ``sigma_upper``, if input. Defaults to 3. sigma_lower : float or `None`, optional The number of standard deviations to use as the lower bound for the clipping limit. If `None` then the value of ``sigma`` is used. Defaults to `None`. sigma_upper : float or `None`, optional The number of standard deviations to use as the upper bound for the clipping limit. If `None` then the value of ``sigma`` is used. Defaults to `None`. iters : int or `None`, optional The number of iterations to perform sigma clipping, or `None` to clip until convergence is achieved (i.e., continue until the last iteration clips nothing). Defaults to 5. cenfunc : callable, optional The function used to compute the center for the clipping. Must be a callable that takes in a masked array and outputs the central value. Defaults to the median (`numpy.ma.median`). stdfunc : callable, optional The function used to compute the standard deviation about the center. Must be a callable that takes in a masked array and outputs a width estimator. Masked (rejected) pixels are those where:: deviation < (-sigma_lower * stdfunc(deviation)) deviation > (sigma_upper * stdfunc(deviation)) where:: deviation = data - cenfunc(data [,axis=int]) Defaults to the standard deviation (`numpy.std`). See Also -------- sigma_clip Examples -------- This example generates random variates from a Gaussian distribution and returns a masked array in which all points that are more than 2 sample standard deviations from the median are masked:: >>> from astropy.stats import SigmaClip >>> from numpy.random import randn >>> randvar = randn(10000) >>> sigclip = SigmaClip(sigma=2, iters=5) >>> filtered_data = sigclip(randvar) This example sigma clips on a similar distribution, but uses 3 sigma relative to the sample *mean*, clips until convergence, and does not copy the data:: >>> from astropy.stats import SigmaClip >>> from numpy.random import randn >>> from numpy import mean >>> randvar = randn(10000) >>> sigclip = SigmaClip(sigma=3, iters=None, cenfunc=mean) >>> filtered_data = sigclip(randvar, copy=False) This example sigma clips along one axis on a similar distribution (with bad points inserted):: >>> from astropy.stats import SigmaClip >>> from numpy.random import normal >>> from numpy import arange, diag, ones >>> data = arange(5) + normal(0., 0.05, (5, 5)) + diag(ones(5)) >>> sigclip = SigmaClip(sigma=2.3) >>> filtered_data = sigclip(data, axis=0) Note that along the other axis, no points would be masked, as the variance is higher. """ def __init__(self, sigma=3., sigma_lower=None, sigma_upper=None, iters=5, cenfunc=np.ma.median, stdfunc=np.std): self.sigma = sigma self.sigma_lower = sigma_lower self.sigma_upper = sigma_upper self.iters = iters self.cenfunc = cenfunc self.stdfunc = stdfunc def __repr__(self): return ('SigmaClip(sigma={0}, sigma_lower={1}, sigma_upper={2}, ' 'iters={3}, cenfunc={4}, stdfunc={5})' .format(self.sigma, self.sigma_lower, self.sigma_upper, self.iters, self.cenfunc, self.stdfunc)) def __str__(self): lines = ['<' + self.__class__.__name__ + '>'] attrs = ['sigma', 'sigma_lower', 'sigma_upper', 'iters', 'cenfunc', 'stdfunc'] for attr in attrs: lines.append(' {0}: {1}'.format(attr, getattr(self, attr))) return '\n'.join(lines) def _perform_clip(self, _filtered_data, axis=None): """ Perform sigma clip by comparing the data to the minimum and maximum values (median + sig * standard deviation). Use sigma_lower and sigma_upper to get the correct limits. Data values less or greater than the minimum / maximum values will have True set in the mask array. """ if _filtered_data.size == 0: return _filtered_data max_value = self.cenfunc(_filtered_data, axis=axis) std = self.stdfunc(_filtered_data, axis=axis) min_value = max_value - std * self.sigma_lower max_value += std * self.sigma_upper if axis is not None: if axis != 0: min_value = np.expand_dims(min_value, axis=axis) max_value = np.expand_dims(max_value, axis=axis) if max_value is np.ma.masked: max_value = np.ma.MaskedArray(np.nan, mask=True) min_value = np.ma.MaskedArray(np.nan, mask=True) _filtered_data.mask |= _filtered_data > max_value _filtered_data.mask |= _filtered_data < min_value return _filtered_data
[docs] def __call__(self, data, axis=None, copy=True): """ Perform sigma clipping on the provided data. Parameters ---------- data : array-like The data to be sigma clipped. axis : int or `None`, optional If not `None`, clip along the given axis. For this case, ``axis`` will be passed on to ``cenfunc`` and ``stdfunc``, which are expected to return an array with the axis dimension removed (like the numpy functions). If `None`, clip over all axes. Defaults to `None`. copy : bool, optional If `True`, the ``data`` array will be copied. If `False`, the returned masked array data will contain the same array as ``data``. Defaults to `True`. Returns ------- filtered_data : `numpy.ma.MaskedArray` A masked array with the same shape as ``data`` input, where the points rejected by the algorithm have been masked. """ if self.sigma_lower is None: self.sigma_lower = self.sigma if self.sigma_upper is None: self.sigma_upper = self.sigma if np.any(~np.isfinite(data)): data = np.ma.masked_invalid(data) warnings.warn('Input data contains invalid values (NaNs or ' 'infs), which were automatically masked.', AstropyUserWarning) filtered_data = np.ma.array(data, copy=copy) if self.iters is None: lastrej = filtered_data.count() + 1 while filtered_data.count() != lastrej: lastrej = filtered_data.count() self._perform_clip(filtered_data, axis=axis) else: for i in range(self.iters): self._perform_clip(filtered_data, axis=axis) # prevent filtered_data.mask = False (scalar) if no values are clipped if filtered_data.mask.shape == (): # make .mask shape match .data shape filtered_data.mask = False return filtered_data
[docs]def sigma_clip(data, sigma=3, sigma_lower=None, sigma_upper=None, iters=5, cenfunc=np.ma.median, stdfunc=np.std, axis=None, copy=True): """ Perform sigma-clipping on the provided data. The data will be iterated over, each time rejecting points that are discrepant by more than a specified number of standard deviations from a center value. If the data contains invalid values (NaNs or infs), they are automatically masked before performing the sigma clipping. For an object-oriented interface to sigma clipping, see :func:`SigmaClip`. .. note:: `scipy.stats.sigmaclip <http://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.sigmaclip.html>`_ provides a subset of the functionality in this function. Parameters ---------- data : array-like The data to be sigma clipped. sigma : float, optional The number of standard deviations to use for both the lower and upper clipping limit. These limits are overridden by ``sigma_lower`` and ``sigma_upper``, if input. Defaults to 3. sigma_lower : float or `None`, optional The number of standard deviations to use as the lower bound for the clipping limit. If `None` then the value of ``sigma`` is used. Defaults to `None`. sigma_upper : float or `None`, optional The number of standard deviations to use as the upper bound for the clipping limit. If `None` then the value of ``sigma`` is used. Defaults to `None`. iters : int or `None`, optional The number of iterations to perform sigma clipping, or `None` to clip until convergence is achieved (i.e., continue until the last iteration clips nothing). Defaults to 5. cenfunc : callable, optional The function used to compute the center for the clipping. Must be a callable that takes in a masked array and outputs the central value. Defaults to the median (`numpy.ma.median`). stdfunc : callable, optional The function used to compute the standard deviation about the center. Must be a callable that takes in a masked array and outputs a width estimator. Masked (rejected) pixels are those where:: deviation < (-sigma_lower * stdfunc(deviation)) deviation > (sigma_upper * stdfunc(deviation)) where:: deviation = data - cenfunc(data [,axis=int]) Defaults to the standard deviation (`numpy.std`). axis : int or `None`, optional If not `None`, clip along the given axis. For this case, ``axis`` will be passed on to ``cenfunc`` and ``stdfunc``, which are expected to return an array with the axis dimension removed (like the numpy functions). If `None`, clip over all axes. Defaults to `None`. copy : bool, optional If `True`, the ``data`` array will be copied. If `False`, the returned masked array data will contain the same array as ``data``. Defaults to `True`. Returns ------- filtered_data : `numpy.ma.MaskedArray` A masked array with the same shape as ``data`` input, where the points rejected by the algorithm have been masked. Notes ----- 1. The routine works by calculating:: deviation = data - cenfunc(data [,axis=int]) and then setting a mask for points outside the range:: deviation < (-sigma_lower * stdfunc(deviation)) deviation > (sigma_upper * stdfunc(deviation)) It will iterate a given number of times, or until no further data are rejected. 2. Most numpy functions deal well with masked arrays, but if one would like to have an array with just the good (or bad) values, one can use:: good_only = filtered_data.data[~filtered_data.mask] bad_only = filtered_data.data[filtered_data.mask] However, for multidimensional data, this flattens the array, which may not be what one wants (especially if filtering was done along an axis). See Also -------- SigmaClip Examples -------- This example generates random variates from a Gaussian distribution and returns a masked array in which all points that are more than 2 sample standard deviations from the median are masked:: >>> from astropy.stats import sigma_clip >>> from numpy.random import randn >>> randvar = randn(10000) >>> filtered_data = sigma_clip(randvar, sigma=2, iters=5) This example sigma clips on a similar distribution, but uses 3 sigma relative to the sample *mean*, clips until convergence, and does not copy the data:: >>> from astropy.stats import sigma_clip >>> from numpy.random import randn >>> from numpy import mean >>> randvar = randn(10000) >>> filtered_data = sigma_clip(randvar, sigma=3, iters=None, ... cenfunc=mean, copy=False) This example sigma clips along one axis on a similar distribution (with bad points inserted):: >>> from astropy.stats import sigma_clip >>> from numpy.random import normal >>> from numpy import arange, diag, ones >>> data = arange(5) + normal(0., 0.05, (5, 5)) + diag(ones(5)) >>> filtered_data = sigma_clip(data, sigma=2.3, axis=0) Note that along the other axis, no points would be masked, as the variance is higher. """ sigclip = SigmaClip(sigma=sigma, sigma_lower=sigma_lower, sigma_upper=sigma_upper, iters=iters, cenfunc=cenfunc, stdfunc=stdfunc) return sigclip(data, axis=axis, copy=copy)
[docs]def sigma_clipped_stats(data, mask=None, mask_value=None, sigma=3.0, sigma_lower=None, sigma_upper=None, iters=5, cenfunc=np.ma.median, stdfunc=np.std, std_ddof=0, axis=None): """ Calculate sigma-clipped statistics on the provided data. Parameters ---------- data : array-like Data array or object that can be converted to an array. mask : `numpy.ndarray` (bool), optional A boolean mask with the same shape as ``data``, where a `True` value indicates the corresponding element of ``data`` is masked. Masked pixels are excluded when computing the statistics. mask_value : float, optional A data value (e.g., ``0.0``) that is ignored when computing the statistics. ``mask_value`` will be masked in addition to any input ``mask``. sigma : float, optional The number of standard deviations to use as the lower and upper clipping limit. These limits are overridden by ``sigma_lower`` and ``sigma_upper``, if input. Defaults to 3. sigma_lower : float, optional The number of standard deviations to use as the lower bound for the clipping limit. If `None` then the value of ``sigma`` is used. Defaults to `None`. sigma_upper : float, optional The number of standard deviations to use as the upper bound for the clipping limit. If `None` then the value of ``sigma`` is used. Defaults to `None`. iters : int, optional The number of iterations to perform sigma clipping, or `None` to clip until convergence is achieved (i.e., continue until the last iteration clips nothing) when calculating the statistics. Defaults to 5. cenfunc : callable, optional The function used to compute the center for the clipping. Must be a callable that takes in a masked array and outputs the central value. Defaults to the median (`numpy.ma.median`). stdfunc : callable, optional The function used to compute the standard deviation about the center. Must be a callable that takes in a masked array and outputs a width estimator. Masked (rejected) pixels are those where:: deviation < (-sigma_lower * stdfunc(deviation)) deviation > (sigma_upper * stdfunc(deviation)) where:: deviation = data - cenfunc(data [,axis=int]) Defaults to the standard deviation (`numpy.std`). std_ddof : int, optional The delta degrees of freedom for the standard deviation calculation. The divisor used in the calculation is ``N - std_ddof``, where ``N`` represents the number of elements. The default is zero. axis : int or `None`, optional If not `None`, clip along the given axis. For this case, ``axis`` will be passed on to ``cenfunc`` and ``stdfunc``, which are expected to return an array with the axis dimension removed (like the numpy functions). If `None`, clip over all axes. Defaults to `None`. Returns ------- mean, median, stddev : float The mean, median, and standard deviation of the sigma-clipped data. """ if mask is not None: data = np.ma.MaskedArray(data, mask) if mask_value is not None: data = np.ma.masked_values(data, mask_value) data_clip = sigma_clip(data, sigma=sigma, sigma_lower=sigma_lower, sigma_upper=sigma_upper, iters=iters, cenfunc=cenfunc, stdfunc=stdfunc, axis=axis) mean = np.ma.mean(data_clip, axis=axis) median = np.ma.median(data_clip, axis=axis) std = np.ma.std(data_clip, ddof=std_ddof, axis=axis) if axis is None and np.ma.isMaskedArray(median): # With Numpy 1.10 np.ma.median always return a MaskedArray, even with # one element. So for compatibility with previous versions, we take the # scalar value median = median.item() return mean, median, std