How to use the pysal.lib.common.np.array function in pysal

>>> w = np.array([0.5, 0.1, 0.3, 0.8])

    We specify the number of groups for which the weighted mean is computed.

    >>> n = 2

    Applying sum_by_n function

    >>> sum_by_n(d, w, n)
    array([ 5.9, 30. ])

    """
    t = len(d)
    h = t // n #must be floor!
    d = d * w
    return np.array([sum(d[i: i + h]) for i in range(0, t, h)])

def __getCardinalities(self):
        ido = self.w.id_order
        self.wc = np.array([self.w.cardinalities[ido[i]] for i in range(self.n)])
        return self.wc

extra[id][-1] * 1.0 / extra[id][1])
            trp_r = sorted(trp_r)
            if not weighted:
                return r[id], trp_r[0], trp_r[-1]
            else:
                trp_aw = self.aw[triples[0]]
                extra_w = trp_aw[0] + (trp_aw[0] - trp_aw[-
                                                          1]) * (extra[id][-1] * 1.0 / extra[id][1])
                return r[id], trp_r[0], trp_r[-1], self.aw[id], trp_aw[0] + extra_w
        if not weighted:
            lowest, highest = [], []
            for trp in triples:
                trp_r = np.sort(r[list(trp)])
                lowest.append(trp_r[0])
                highest.append(trp_r[-1])
            return r[id], np.median(np.array(lowest)), np.median(np.array(highest))
        if weighted:
            lowest, highest = [], []
            lowest_aw, highest_aw = [], []
            for trp in triples:
                trp_r = r[list(trp)]
                dtype = [('r', '%s' % trp_r.dtype), ('w',
                                                     '%s' % self.aw.dtype)]
                trp_r = np.array(list(zip(trp_r, list(trp))), dtype=dtype)
                trp_r.sort(order='r')
                lowest.append(trp_r['r'][0])
                highest.append(trp_r['r'][-1])
                lowest_aw.append(self.aw[int(round(trp_r['w'][0]))])
                highest_aw.append(self.aw[int(round(trp_r['w'][-1]))])
            wm_lowest = weighted_median(np.array(lowest), np.array(lowest_aw))
            wm_highest = weighted_median(
                np.array(highest), np.array(highest_aw))

>>> d = np.array([5,4,3,1,2])

    Creating another array including weight values for the above integers.
    The median of d will be decided with a consideration to these weight
    values.

    >>> w = np.array([10, 22, 9, 2, 5])

    Applying weighted_median function

    >>> weighted_median(d, w)
    4

    """
    dtype = [('w', '%s' % w.dtype), ('v', '%s' % d.dtype)]
    d_w = np.array(list(zip(w, d)), dtype=dtype)
    d_w.sort(order='v')
    reordered_w = d_w['w'].cumsum()
    cumsum_threshold = reordered_w[-1] * 1.0 / 2
    median_inx = (reordered_w >= cumsum_threshold).nonzero()[0][0]
    if reordered_w[median_inx] == cumsum_threshold and len(d) - 1 > median_inx:
        return np.sort(d)[median_inx:median_inx + 2].mean()
    return np.sort(d)[median_inx]

def __crand(self):
        y = self.y
        rGs = np.zeros((self.n, self.permutations))
        n_1 = self.n - 1
        rid = list(range(n_1))
        prange = list(range(self.permutations))
        k = self.w.max_neighbors + 1
        rids = np.array([np.random.permutation(rid)[0:k] for i in prange])
        ids = np.arange(self.w.n)
        ido = self.w.id_order
        wc = self.__getCardinalities()
        if self.w_transform == "r":
            den = np.array(wc) + self.star
        else:
            den = np.ones(self.w.n)
        for i in range(self.w.n):
            idsi = ids[ids != i]
            np.random.shuffle(idsi)
            yi_star = y[i] * self.star
            wci = wc[i]
            rGs[i] = (y[idsi[rids[:, 0:wci]]]).sum(1) + yi_star
            rGs[i] = (np.array(rGs[i]) / den[i]) / (self.y_sum - (1 - self.star) * y[i])
        self.rGs = rGs

def __search_headbanging_median(self):
        r, tr = self.r, self.tr
        new_r = []
        for k in list(tr.keys()):
            screens = self.__get_screens(
                k, tr[k], weighted=(self.aw is not None))
            new_r.append(self.__get_median_from_screens(screens))
        self.r = np.array(new_r)

return r[id], trp_r[0], trp_r[-1], self.aw[id], trp_aw[0] + extra_w
        if not weighted:
            lowest, highest = [], []
            for trp in triples:
                trp_r = np.sort(r[list(trp)])
                lowest.append(trp_r[0])
                highest.append(trp_r[-1])
            return r[id], np.median(np.array(lowest)), np.median(np.array(highest))
        if weighted:
            lowest, highest = [], []
            lowest_aw, highest_aw = [], []
            for trp in triples:
                trp_r = r[list(trp)]
                dtype = [('r', '%s' % trp_r.dtype), ('w',
                                                     '%s' % self.aw.dtype)]
                trp_r = np.array(list(zip(trp_r, list(trp))), dtype=dtype)
                trp_r.sort(order='r')
                lowest.append(trp_r['r'][0])
                highest.append(trp_r['r'][-1])
                lowest_aw.append(self.aw[int(round(trp_r['w'][0]))])
                highest_aw.append(self.aw[int(round(trp_r['w'][-1]))])
            wm_lowest = weighted_median(np.array(lowest), np.array(lowest_aw))
            wm_highest = weighted_median(
                np.array(highest), np.array(highest_aw))
            triple_members = flatten(triples, unique=False)
            return r[id], wm_lowest, wm_highest, self.aw[id] * len(triples), self.aw[triple_members].sum()

def __crand(self):
        y = self.y
        rGs = np.zeros((self.n, self.permutations))
        n_1 = self.n - 1
        rid = list(range(n_1))
        prange = list(range(self.permutations))
        k = self.w.max_neighbors + 1
        rids = np.array([np.random.permutation(rid)[0:k] for i in prange])
        ids = np.arange(self.w.n)
        ido = self.w.id_order
        wc = self.__getCardinalities()
        if self.w_transform == "r":
            den = np.array(wc) + self.star
        else:
            den = np.ones(self.w.n)
        for i in range(self.w.n):
            idsi = ids[ids != i]
            np.random.shuffle(idsi)
            yi_star = y[i] * self.star
            wci = wc[i]
            rGs[i] = (y[idsi[rids[:, 0:wci]]]).sum(1) + yi_star
            rGs[i] = (np.array(rGs[i]) / den[i]) / (self.y_sum - (1 - self.star) * y[i])
        self.rGs = rGs