Python _util._validate_counts_vector函数代码示例

def osd(counts):
    """Calculate observed OTUs, singles, and doubles.

    Parameters
    ----------
    counts : 1-D array_like, int
        Vector of counts.

    Returns
    -------
    osd : tuple
        Observed OTUs, singles, and doubles.

    See Also
    --------
    observed_otus
    singles
    doubles

    Notes
    -----
    This is a convenience function used by many of the other measures that rely
    on these three measures.

    """
    counts = _validate_counts_vector(counts)
    return observed_otus(counts), singles(counts), doubles(counts)

def robbins(counts):
    r"""Calculate Robbins' estimator for the probability of unobserved outcomes.

    Robbins' estimator is defined as:

    .. math::

       \frac{F_1}{n+1}

    where :math:`F_1` is the number of singleton OTUs.

    Parameters
    ----------
    counts : 1-D array_like, int
        Vector of counts.

    Returns
    -------
    double
        Robbins' estimate.

    Notes
    -----
    Robbins' estimator is defined in [1]_. The estimate computed here is for
    :math:`n-1` counts, i.e. the x-axis is off by 1.

    References
    ----------
    .. [1] Robbins, H. E (1968). Ann. of Stats. Vol 36, pp. 256-257.

    """
    counts = _validate_counts_vector(counts)
    return singles(counts) / counts.sum()

def goods_coverage(counts):
    r"""Calculate Good's coverage of counts.

    Good's coverage estimator is defined as

    .. math::

       1-\frac{F_1}{N}

    where :math:`F_1` is the number of singleton OTUs and :math:`N` is the
    total number of individuals (sum of abundances for all OTUs).

    Parameters
    ----------
    counts : 1-D array_like, int
        Vector of counts.

    Returns
    -------
    double
        Good's coverage estimator.

    """
    counts = _validate_counts_vector(counts)
    f1 = singles(counts)
    N = counts.sum()
    return 1 - (f1 / N)

def fisher_alpha(counts):
    r"""Calculate Fisher's alpha, a metric of diversity.

    Fisher's alpha is estimated by solving the following equation for
    :math:`\alpha`:

    .. math::

       S=\alpha\ln(1+\frac{N}{\alpha})

    where :math:`S` is the number of OTUs and :math:`N` is the
    total number of individuals in the sample.

    Parameters
    ----------
    counts : 1-D array_like, int
        Vector of counts.

    Returns
    -------
    double
        Fisher's alpha.

    Raises
    ------
    RuntimeError
        If the optimizer fails to converge (error > 1.0).

    Notes
    -----
    The implementation here is based on the description given in the SDR-IV
    online manual [1]_. Uses ``scipy.optimize.minimize_scalar`` to find
    Fisher's alpha.

    References
    ----------
    .. [1] http://www.pisces-conservation.com/sdrhelp/index.html

    """
    counts = _validate_counts_vector(counts)
    n = counts.sum()
    s = observed_otus(counts)

    def f(alpha):
        return (alpha * np.log(1 + (n / alpha)) - s) ** 2

    # Temporarily silence RuntimeWarnings (invalid and division by zero) during
    # optimization in case invalid input is provided to the objective function
    # (e.g. alpha=0).
    orig_settings = np.seterr(divide='ignore', invalid='ignore')
    try:
        alpha = minimize_scalar(f).x
    finally:
        np.seterr(**orig_settings)

    if f(alpha) > 1.0:
        raise RuntimeError("Optimizer failed to converge (error > 1.0), so "
                           "could not compute Fisher's alpha.")
    return alpha

def chao1(counts, bias_corrected=True):
    r"""Calculate chao1 richness estimator.

    Uses the bias-corrected version unless `bias_corrected` is ``False`` *and*
    there are both singletons and doubletons.

    Parameters
    ----------
    counts : 1-D array_like, int
        Vector of counts.
    bias_corrected : bool, optional
        Indicates whether or not to use the bias-corrected version of the
        equation. If ``False`` *and* there are both singletons and doubletons,
        the uncorrected version will be used. The biased-corrected version will
        be used otherwise.

    Returns
    -------
    double
        Computed chao1 richness estimator.

    See Also
    --------
    chao1_ci

    Notes
    -----
    The uncorrected version is based on Equation 6 in [1]_:

    .. math::

       chao1=S_{obs}+\frac{F_1^2}{2F_2}

    where :math:`F_1` and :math:`F_2` are the count of singletons and
    doubletons, respectively.

    The bias-corrected version is defined as

    .. math::

       chao1=S_{obs}+\frac{F_1(F_1-1)}{2(F_2+1)}

    References
    ----------
    .. [1] Chao, A. 1984. Non-parametric estimation of the number of classes in
       a population. Scandinavian Journal of Statistics 11, 265-270.

    """
    counts = _validate_counts_vector(counts)
    o, s, d = osd(counts)

    if not bias_corrected and s and d:
        return o + s ** 2 / (d * 2)
    else:
        return o + s * (s - 1) / (2 * (d + 1))

def chao1_ci(counts, bias_corrected=True, zscore=1.96):
    """Calculate chao1 confidence interval.

    Parameters
    ----------
    counts : 1-D array_like, int
        Vector of counts.
    bias_corrected : bool, optional
        Indicates whether or not to use the bias-corrected version of the
        equation. If ``False`` *and* there are both singletons and doubletons,
        the uncorrected version will be used. The biased-corrected version will
        be used otherwise.
    zscore : scalar, optional
        Score to use for confidence. Default of 1.96 is for a 95% confidence
        interval.

    Returns
    -------
    tuple
        chao1 confidence interval as ``(lower_bound, upper_bound)``.

    See Also
    --------
    chao1

    Notes
    -----
    The implementation here is based on the equations in the EstimateS manual
    [1]_. Different equations are employed to calculate the chao1 variance and
    confidence interval depending on `bias_corrected` and the presence/absence
    of singletons and/or doubletons.

    Specifically, the following EstimateS equations are used:

    1. No singletons, Equation 14.
    2. Singletons but no doubletons, Equations 7, 13.
    3. Singletons and doubletons, ``bias_corrected=True``, Equations 6, 13.
    4. Singletons and doubletons, ``bias_corrected=False``, Equations 5, 13.

    References
    ----------
    .. [1] http://viceroy.eeb.uconn.edu/estimates/

    """
    counts = _validate_counts_vector(counts)
    o, s, d = osd(counts)
    if s:
        chao = chao1(counts, bias_corrected)
        chaovar = _chao1_var(counts, bias_corrected)
        return _chao_confidence_with_singletons(chao, o, chaovar, zscore)
    else:
        n = counts.sum()
        return _chao_confidence_no_singletons(n, o, zscore)

    def test_validate_counts_vector(self):
        # python list
        obs = _validate_counts_vector([0, 2, 1, 3])
        npt.assert_array_equal(obs, np.array([0, 2, 1, 3]))
        self.assertEqual(obs.dtype, int)

        # numpy array (no copy made)
        data = np.array([0, 2, 1, 3])
        obs = _validate_counts_vector(data)
        npt.assert_array_equal(obs, data)
        self.assertEqual(obs.dtype, int)
        self.assertTrue(obs is data)

        # single element
        obs = _validate_counts_vector([42])
        npt.assert_array_equal(obs, np.array([42]))
        self.assertEqual(obs.dtype, int)
        self.assertEqual(obs.shape, (1,))

        # suppress casting to int
        obs = _validate_counts_vector([42.2, 42.1, 0], suppress_cast=True)
        npt.assert_array_equal(obs, np.array([42.2, 42.1, 0]))
        self.assertEqual(obs.dtype, float)

        # all zeros
        obs = _validate_counts_vector([0, 0, 0])
        npt.assert_array_equal(obs, np.array([0, 0, 0]))
        self.assertEqual(obs.dtype, int)

        # all zeros (single value)
        obs = _validate_counts_vector([0])
        npt.assert_array_equal(obs, np.array([0]))
        self.assertEqual(obs.dtype, int)

def esty_ci(counts):
    r"""Calculate Esty's CI.

    Esty's CI is defined as

    .. math::

       F_1/N \pm z\sqrt{W}

    where :math:`F_1` is the number of singleton OTUs, :math:`N` is the total
    number of individuals (sum of abundances for all OTUs), and :math:`z` is a
    constant that depends on the targeted confidence and based on the normal
    distribution.

    :math:`W` is defined as

    .. math::

       \frac{F_1(N-F_1)+2NF_2}{N^3}

    where :math:`F_2` is the number of doubleton OTUs.

    Parameters
    ----------
    counts : 1-D array_like, int
        Vector of counts.

    Returns
    -------
    tuple
        Esty's confidence interval as ``(lower_bound, upper_bound)``.

    Notes
    -----
    Esty's CI is defined in [1]_. :math:`z` is hardcoded for a 95% confidence
    interval.

    References
    ----------
    .. [1] Esty, W. W. (1983). "A normal limit law for a nonparametric
       estimator of the coverage of a random sample". Ann Statist 11: 905-912.

    """
    counts = _validate_counts_vector(counts)

    f1 = singles(counts)
    f2 = doubles(counts)
    n = counts.sum()
    z = 1.959963985
    W = (f1 * (n - f1) + 2 * n * f2) / (n ** 3)

    return f1 / n - z * np.sqrt(W), f1 / n + z * np.sqrt(W)

def _setup_faith_pd(counts, otu_ids, tree, validate, single_sample):
    if validate:
        if single_sample:
            # only validate count if operating in single sample mode, they
            # will have already been validated otherwise
            counts = _validate_counts_vector(counts)
            _validate_otu_ids_and_tree(counts, otu_ids, tree)
        else:
            _validate_otu_ids_and_tree(counts[0], otu_ids, tree)

    counts_by_node, tree_index, branch_lengths = _vectorize_counts_and_tree(counts, otu_ids, tree)

    return counts_by_node, branch_lengths

def lladser_ci(counts, r, alpha=0.95, f=10, ci_type='ULCL'):
    """Calculate single CI of the conditional uncovered probability.

    Parameters
    ----------
    counts : 1-D array_like, int
        Vector of counts.
    r : int
        Number of new colors that are required for the next prediction.
    alpha : float, optional
        Desired confidence level.
    f : float, optional
        Ratio between upper and lower bound.
    ci_type : {'ULCL', 'ULCU', 'U', 'L'}
        Type of confidence interval. If ``'ULCL'``, upper and lower bounds with
        conservative lower bound. If ``'ULCU'``, upper and lower bounds with
        conservative upper bound. If ``'U'``, upper bound only, lower bound
        fixed to 0.0. If ``'L'``, lower bound only, upper bound fixed to 1.0.

    Returns
    -------
    tuple
        Confidence interval as ``(lower_bound, upper_bound)``.

    See Also
    --------
    lladser_pe

    Notes
    -----
    This function is just a wrapper around the full CI estimator described
    in Theorem 2 (iii) in [1]_, intended to be called for a single best CI
    estimate on a complete sample.

    References
    ----------
    .. [1] Lladser, Gouet, and Reeder, "Extrapolation of Urn Models via
       Poissonization: Accurate Measurements of the Microbial Unknown" PLoS
       2011.

    """
    counts = _validate_counts_vector(counts)
    sample = _expand_counts(counts)
    np.random.shuffle(sample)

    try:
        ci = list(_lladser_ci_series(sample, r, alpha, f, ci_type))[-1]
    except IndexError:
        ci = (np.nan, np.nan)

    return ci

    def test_validate_counts_vector_invalid_input(self):
        # wrong dtype
        with self.assertRaises(TypeError):
            _validate_counts_vector([0, 2, 1.2, 3])

        # wrong number of dimensions (2-D)
        with self.assertRaises(ValueError):
            _validate_counts_vector([[0, 2, 1, 3], [4, 5, 6, 7]])

        # wrong number of dimensions (scalar)
        with self.assertRaises(ValueError):
            _validate_counts_vector(1)

        # negative values
        with self.assertRaises(ValueError):
            _validate_counts_vector([0, 0, 2, -1, 3])

def mcintosh_d(counts):
    r"""Calculate McIntosh dominance index D.

    McIntosh dominance index D is defined as:

    .. math::

       D = \frac{N - U}{N - \sqrt{N}}

    where :math:`N` is the total number of individuals in the sample and
    :math:`U` is defined as:

    .. math::

       U = \sqrt{\sum{{n_i}^2}}

    where :math:`n_i` is the number of individuals in the :math:`i^{\text{th}}`
    OTU.

    Parameters
    ----------
    counts : 1-D array_like, int
        Vector of counts.

    Returns
    -------
    double
        McIntosh dominance index D.

    See Also
    --------
    mcintosh_e

    Notes
    -----
    The index was proposed in [1]_. The implementation here is based on the
    description given in the SDR-IV online manual [2]_.

    References
    ----------
    .. [1] McIntosh, R. P. 1967 An index of diversity and the relation of
       certain concepts to diversity. Ecology 48, 1115-1126.
    .. [2] http://www.pisces-conservation.com/sdrhelp/index.html

    """
    counts = _validate_counts_vector(counts)
    u = np.sqrt((counts * counts).sum())
    n = counts.sum()
    return (n - u) / (n - np.sqrt(n))

def kempton_taylor_q(counts, lower_quantile=0.25, upper_quantile=0.75):
    """Calculate Kempton-Taylor Q index of alpha diversity.

    Estimates the slope of the cumulative abundance curve in the interquantile
    range. By default, uses lower and upper quartiles, rounding inwards.

    Parameters
    ----------
    counts : 1-D array_like, int
        Vector of counts.
    lower_quantile : float, optional
        Lower bound of the interquantile range. Defaults to lower quartile.
    upper_quantile : float, optional
        Upper bound of the interquantile range. Defaults to upper quartile.

    Returns
    -------
    double
        Kempton-Taylor Q index of alpha diversity.

    Notes
    -----
    The index is defined in [1]_. The implementation here is based on the
    description given in the SDR-IV online manual [2]_.

    The implementation provided here differs slightly from the results given in
    Magurran 1998. Specifically, we have 14 in the numerator rather than 15.
    Magurran recommends counting half of the OTUs with the same # counts as the
    point where the UQ falls and the point where the LQ falls, but the
    justification for this is unclear (e.g. if there were a very large # OTUs
    that just overlapped one of the quantiles, the results would be
    considerably off). Leaving the calculation as-is for now, but consider
    changing.

    References
    ----------
    .. [1] Kempton, R. A. and Taylor, L. R. (1976) Models and statistics for
       species diversity. Nature, 262, 818-820.
    .. [2] http://www.pisces-conservation.com/sdrhelp/index.html

    """
    counts = _validate_counts_vector(counts)
    n = len(counts)
    lower = int(np.ceil(n * lower_quantile))
    upper = int(n * upper_quantile)
    sorted_counts = np.sort(counts)
    return (upper - lower) / np.log(sorted_counts[upper] /
                                    sorted_counts[lower])

def observed_otus(counts):
    """Calculate the number of distinct OTUs.

    Parameters
    ----------
    counts : 1-D array_like, int
        Vector of counts.

    Returns
    -------
    int
        Distinct OTU count.

    """
    counts = _validate_counts_vector(counts)
    return (counts != 0).sum()

def singles(counts):
    """Calculate number of single occurrences (singletons).

    Parameters
    ----------
    counts : 1-D array_like, int
        Vector of counts.

    Returns
    -------
    int
        Singleton count.

    """
    counts = _validate_counts_vector(counts)
    return (counts == 1).sum()