\ciRddlZddlZddlmZddlmZmZmZm Z m Z m Z m Z ddl mcmZddlmZmZdZedgdd dd ddd d Zd ZdZdZdZedgddddZdS)N)betainc)xp_capabilitiesxp_ravelarray_namespace xp_promote xp_device_length_nonmaskedis_torch)_broadcast_arrays _contains_nanc x t|||}|||jdst d|||jdst d|=|||jdst dt |||d|\}}}||t |}|j}|du} t|j|j} | r!t|}t|}d }nVtj |st||krd } t | || ks|| krd } t | t|}hd } || vrd | } t | hd} ||| vrd|d} t | t||d|}|dvrd} t | |j|d krDt|j}d||<|||j|t |}|1|||d}t'||f|\}}d}n|||\}}|d k}|||d}t-j|||jd}|||d}||||}||||}t'|||||f|\}}}}}|dur"|j|dkrd} t | ||j|dkn|}|||d}|||d}||n|||d}||n|||d}t;|d|d}||n||z }|rf||}|dkr||d}n|||ddz }| ||}||d kd}t-j|||jd}||rE||d}t-j|||j}nZ||rE|dkr?||d}t-j||d }|||t |}|dk|d kz||z}||r?||d}t-j||d }|||||||| | |||f S)!N)integral real floatingz`x` must have real dtype.rz"`p` must have real floating dtype.z`weights` must have real dtype.T)force_floatingxpdevicerz"`axis` must be an integer or None.z7`axis` is not compatible with the shapes of the inputs.>hazen_lowerlinear_higherweibull_nearest _midpoint inverted_cdf round_inward round_nearest round_outwardmedian_unbiasednormal_unbiasedclosest_observationaveraged_inverted_cdfinterpolated_inverted_cdf harrell-davisz`method` must be one of >rrrrrrrr$z `method='z'` does not support `weights`.) xp_omit_okayr>FNTz/If specified, `keepdims` must be True or False.dtyperF)axisstabler)r)keepdimscopyzD`keepdims` may be False only if the length of `p` along `axis` is 1.)rr- propagater$?)!risdtypeasarrayr( ValueErrorrrmaxndimrnpiterableintr shapelistfullnansortr broadcast_arrays count_nonzeroxpxatsetinfargsorttake_along_axismoveaxisr isnananyastype)xpmethodr) nan_policyr-weightsrr( axis_noner7messagemethods no_weights contains_nansr;y n_zero_weight i_zero_weighti_ynnansnan_outn_intp_masks e/var/www/html/mdtn/previsions/meteo_cartes/venv/lib/python3.11/site-packages/scipy/stats/_quantile.py _quantile_ivr`sy Aw ' 'B ::bjjmm)+H I I64555 ::bjjmm)? ; ;?=>>> Ozz"**W--35RSS :;;;q!WTbIIIMAq' 1Yq\\ **A GE I qvqv  D " QKK QKK T  "c$ii4//6!!! $,,DD5LLK!!! t99DAAAG W6W66!!!DDDJv33DfDDD!!!!!ZdrJJJM***C!!!  wt}QW d GGE26y||G D D GGADG / / !Qd3331 ((G44 7 A ((TD(QQ F1IIm $ ( (d ( ; ;jjej44   q#D  1 1$$Wc$==,= 7C /d-<-<-<)1gsM E  2 2X!!!'/'7 ""XH AtR  A AtR  A ggbkk'4.L.LG&3&;]]++mT2>> !RB666A  "M(9A'xx{{  $ $ffTf++GG((B(FFFE %''AffQ!V"f--Gq'""&&qwr{33A 66'?? ' 14 ((Aq'""&&rv..AA VVD\\ 'f77 14 ((Aq$##A&&A 1E)A,, 77A!eA ! ,F vvf~~' JJqtJ $ $ F1f   ! !# & & q&$ H y$ 55)z dask.arrayzNo take_along_axis yet.F) skip_backendsjax_jitrr1)rNr)rOr-rPc <t|||||||}|\ }}}}}}} } } } }} |dvrt||| ||| }nB|dvrt||| | }n+|dvrt||| || }nt ||| || }t j|| | j}| r*|r(d| dz z|j z}| ||}d}| |d|}|s| ||}|j dkr|d n|S) a$ Compute the p-th quantile of the data along the specified axis. Parameters ---------- x : array_like of real numbers Data array. p : array_like of float Probability or sequence of probabilities of the quantiles to compute. Values must be between 0 and 1 (inclusive). While `numpy.quantile` can only compute quantiles according to the Cartesian product of the first two arguments, this function enables calculation of quantiles at different probabilities for each axis slice by following broadcasting rules like those of `scipy.stats` reducing functions. See `axis`, `keepdims`, and the examples. method : str, default: 'linear' The method to use for estimating the quantile. The available options, numbered as they appear in [1]_, are: 1. 'inverted_cdf' 2. 'averaged_inverted_cdf' 3. 'closest_observation' 4. 'interpolated_inverted_cdf' 5. 'hazen' 6. 'weibull' 7. 'linear' (default) 8. 'median_unbiased' 9. 'normal_unbiased' 'harrell-davis' is also available to compute the quantile estimate according to [2]_. 'round_outward', 'round_inward', and 'round_nearest' are available for use in trimming and winsorizing data. See Notes for details. axis : int or None, default: 0 Axis along which the quantiles are computed. ``None`` ravels both `x` and `p` before performing the calculation, without checking whether the original shapes were compatible. As in other `scipy.stats` functions, a positive integer `axis` is resolved after prepending 1s to the shape of `x` or `p` as needed until the two arrays have the same dimensionality. When providing `x` and `p` with different dimensionality, consider using negative `axis` integers for clarity. nan_policy : str, default: 'propagate' Defines how to handle NaNs in the input data `x`. - ``propagate``: if a NaN is present in the axis slice (e.g. row) along which the statistic is computed, the corresponding slice of the output will contain NaN(s). - ``omit``: NaNs will be omitted when performing the calculation. If insufficient data remains in the axis slice along which the statistic is computed, the corresponding slice of the output will contain NaN(s). - ``raise``: if a NaN is present, a ``ValueError`` will be raised. If NaNs are present in `p`, a ``ValueError`` will be raised. keepdims : bool, optional Consider the case in which `x` is 1-D and `p` is a scalar: the quantile is a reducing statistic, and the default behavior is to return a scalar. If `keepdims` is set to True, the axis will not be reduced away, and the result will be a 1-D array with one element. The general case is more subtle, since multiple quantiles may be requested for each axis-slice of `x`. For instance, if both `x` and `p` are 1-D and ``p.size > 1``, no axis can be reduced away; there must be an axis to contain the number of quantiles given by ``p.size``. Therefore: - By default, the axis will be reduced away if possible (i.e. if there is exactly one element of `p` per axis-slice of `x`). - If `keepdims` is set to True, the axis will not be reduced away. - If `keepdims` is set to False, the axis will be reduced away if possible, and an error will be raised otherwise. weights : array_like of finite, non-negative real numbers Frequency weights; e.g., for counting number weights, ``quantile(x, p, weights=weights)`` is equivalent to ``quantile(np.repeat(x, weights), p)``. Values other than finite counting numbers are accepted, but may not have valid statistical interpretations. Not compatible with ``method='harrell-davis'`` or those that begin with ``'round_'``. Returns ------- quantile : scalar or ndarray The resulting quantile(s). The dtype is the result dtype of `x` and `p`. See Also -------- numpy.quantile :ref:`outliers` Notes ----- Given a sample `x` from an underlying distribution, `quantile` provides a nonparametric estimate of the inverse cumulative distribution function. By default, this is done by interpolating between adjacent elements in ``y``, a sorted copy of `x`:: (1-g)*y[j] + g*y[j+1] where the index ``j`` and coefficient ``g`` are the integral and fractional components of ``p * (n-1)``, and ``n`` is the number of elements in the sample. This is a special case of Equation 1 of H&F [1]_. More generally, - ``j = (p*n + m - 1) // 1``, and - ``g = (p*n + m - 1) % 1``, where ``m`` may be defined according to several different conventions. The preferred convention may be selected using the ``method`` parameter: =============================== =============== =============== ``method`` number in H&F ``m`` =============================== =============== =============== ``interpolated_inverted_cdf`` 4 ``0`` ``hazen`` 5 ``1/2`` ``weibull`` 6 ``p`` ``linear`` (default) 7 ``1 - p`` ``median_unbiased`` 8 ``p/3 + 1/3`` ``normal_unbiased`` 9 ``p/4 + 3/8`` =============================== =============== =============== Note that indices ``j`` and ``j + 1`` are clipped to the range ``0`` to ``n - 1`` when the results of the formula would be outside the allowed range of non-negative indices. When ``j`` is clipped to zero, ``g`` is set to zero as well. The ``-1`` in the formulas for ``j`` and ``g`` accounts for Python's 0-based indexing. The table above includes only the estimators from [1]_ that are continuous functions of probability `p` (estimators 4-9). SciPy also provides the three discontinuous estimators from [1]_ (estimators 1-3), where ``j`` is defined as above, ``m`` is defined as follows, and ``g`` is ``0`` when ``index = p*n + m - 1`` is less than ``0`` and otherwise is defined below. 1. ``inverted_cdf``: ``m = 0`` and ``g = int(index - j > 0)`` 2. ``averaged_inverted_cdf``: ``m = 0`` and ``g = (1 + int(index - j > 0)) / 2`` 3. ``closest_observation``: ``m = -1/2`` and ``g = 1 - int((index == j) & (j%2 == 1))`` Note that for methods ``inverted_cdf`` and ``averaged_inverted_cdf``, only the relative proportions of tied observations (and relative weights) affect the results; for all other methods, the total number of observations (and absolute weights) matter. A different strategy for computing quantiles from [2]_, ``method='harrell-davis'``, uses a weighted combination of all elements. The weights are computed as: .. math:: w_{n, i} = I_{i/n}(a, b) - I_{(i - 1)/n}(a, b) where :math:`n` is the number of elements in the sample, :math:`i` are the indices :math:`1, 2, ..., n-1, n` of the sorted elements, :math:`a = p (n + 1)`, :math:`b = (1 - p)(n + 1)`, :math:`p` is the probability of the quantile, and :math:`I` is the regularized, lower incomplete beta function (`scipy.special.betainc`). ``method='round_nearest'`` is equivalent to indexing ``y[j]``, where:: j = int(np.round(p*n) if p < 0.5 else np.round(n*p - 1)) This is useful when winsorizing data: replacing ``p*n`` of the most extreme observations with the next most extreme observation. ``method='round_outward'`` adjusts the direction of rounding to winsorize fewer elements:: j = int(np.floor(p*n) if p < 0.5 else np.ceil(n*p - 1)) and ``method='round_inward'`` rounds to winsorize more elements:: j = int(np.ceil(p*n) if p < 0.5 else np.floor(n*p - 1)) These methods are also useful for trimming data: removing ``p*n`` of the most extreme observations. See :ref:`outliers` for example applications. Examples -------- >>> import numpy as np >>> from scipy import stats >>> x = np.asarray([[10, 8, 7, 5, 4], ... [0, 1, 2, 3, 5]]) Take the median of each row. >>> stats.quantile(x, 0.5, axis=-1) array([7., 2.]) Take a different quantile for each row. >>> stats.quantile(x, [[0.25], [0.75]], axis=-1, keepdims=True) array([[5.], [3.]]) Take multiple quantiles for each row. >>> stats.quantile(x, [0.25, 0.75], axis=-1) array([[5., 8.], [1., 3.]]) Take different quantiles for each row. >>> p = np.asarray([[0.25, 0.75], ... [0.5, 1.0]]) >>> stats.quantile(x, p, axis=-1) array([[5., 8.], [2., 5.]]) Take different quantiles for each column. >>> stats.quantile(x.T, p.T, axis=0) array([[5., 2.], [8., 5.]]) References ---------- .. [1] R. J. Hyndman and Y. Fan, "Sample quantiles in statistical packages," The American Statistician, 50(4), pp. 361-365, 1996 .. [2] Harrell, Frank E., and C. E. Davis. "A new distribution-free quantile estimator." Biometrika 69.3 (1982): 635-640. > rrrrrr r!r"r#>r$>rrrrr&r&r0r+r)r` _quantile_hf _quantile_hd _quantile_bc_quantile_winsorrBrCrDr>r;reshaperHsqueezer7)rLrMrNr)rOr-rPtemprVrZrQr7r^rresr;s r_quantilerospP 1fdJ' J JD04.Q64X 4" CCC1aFGR88 $ $ $1aB'' A A A1aFB//q!Q33 &f   ! !"& ) )CXdQh#)+jje$$ ++c2t $ $C )jj4j((h!mm3r77,rac ztddddd|d|z |dz dz|dz dz }||}|||z|z}|dz} | dz } n||d } |||j } || | jdd d z} || | dz d } || z|z}t| |d} t| |dz d} || |j || |j } } |dz}|dkr ||dk|j }n^|dkr&d||dk|j zdz }n2|dkr,d||dk| dzdkz|j z }|dvrA||}tj ||dk d}tj || dk d}| | d|dz } | | d|dz } d|z |||| |jd z||||| |jd zzS)Nrgr2r&gUUUUUU?g?) rr"r!r#rrrrr r0r+)r(rerightsiderr"r!>rr!r"g)dictcumulative_sumr4int64 broadcast_tor;rG_xp_searchsortedrKr(rBrCrDclip)rVrMrZrNrPrmsmjgjp1jcumulative_weightsr] total_weightgs r_rgrgs 1A4()aAcCi1s D D DB 6 A UQYAg !G..wR.@@ 1BH --'9'?'Dt'KLL))*||dz z}|dkr||||||jd||||||jdzdz S|dkr||}n7|dkr||}n|dkr||}|||||jdS) Nr&rr0r+rvrrr)rGrKrryrr)rVrMrZrNrijks r_riris a!eB ""1bii bh&G&Gb"QQ$$Q "''"++rx(H(Hr$RRSVWX X 8   HHRLL 9   GGBKK :   HHRLL  a1bh!7!7b  A AAra)rbleft)rurc |t||n|}|dj}||jdk}t ||fd|\}}|jdk}|st |rN|r|r||dn|}||||}|||d}|S| |j dt| }|j ddkr|S| | |dd } || |j } |d kr|jn|j} t#t%t'j|j ddzD]\} || zd z} ||| d}| ||}||| | } |||| }]|| |||dd d| }|dkr5|| ||j d|n|}|||d}|S)Nr&rr0)r)rrfrtFr.rTr,rrvr+rs)rr4r(r7r r rk searchsortedrKr=r;rrArIrz less_equallessranger:mathlog2rGrmin)rLrVrurxp_default_inty_0dx_1doutrrZrcomparercx0rs r_r{r{st #%*A   "BZZ]](N ::a==  "D aV" 4 4 4DAq 6Q;D x||"& 74 7BJJq"   aooao..ii^%i88  9Q<<00Awr{a "((1++B>>A 17##A#v~~bmm27G3ty--..2 3 3 UqL   12  . . GArNN HHQ1   HHQ1   ((771bffQR$f??@@!Q G GC59W__"((288A;; S 1 1 1#C ))Ce) 4 4C Jra)rnumpyr8 scipy.specialrscipy._lib._array_apirrrrrr r scipy._lib.array_api_extra_libarray_api_extrarBscipy.stats._axis_nan_policyr r r`rorgrhrjrir{rfrar_rs !!!!!!)((((((((IIIIIIIIv5v5v5r IJ   %A+PTB-B-B-B-  B-J$L$L$LN###"BBBBBB" IJKKK#)d&&&&LK&&&ra