Python numpy.select() Examples

def hsv_to_rgb(hsv):
    # Translated from source of colorsys.hsv_to_rgb
    # h,s should be a numpy arrays with values between 0.0 and 1.0
    # v should be a numpy array with values between 0.0 and 255.0
    # hsv_to_rgb returns an array of uints between 0 and 255.
    rgb = np.empty_like(hsv)
    rgb[..., 3:] = hsv[..., 3:]
    h, s, v = hsv[..., 0], hsv[..., 1], hsv[..., 2]
    i = (h * 6.0).astype('uint8')
    f = (h * 6.0) - i
    p = v * (1.0 - s)
    q = v * (1.0 - s * f)
    t = v * (1.0 - s * (1.0 - f))
    i = i % 6
    conditions = [s == 0.0, i == 1, i == 2, i == 3, i == 4, i == 5]
    rgb[..., 0] = np.select(conditions, [v, q, p, p, t, v], default=v)
    rgb[..., 1] = np.select(conditions, [v, v, v, q, p, p], default=t)
    rgb[..., 2] = np.select(conditions, [v, p, t, v, v, q], default=p)
    return rgb.astype('uint8') 

def gsea_pval(es, esnull):
    """Compute nominal p-value.

    From article (PNAS):
    estimate nominal p-value for S from esnull by using the positive
    or negative portion of the distribution corresponding to the sign
    of the observed ES(S).
    """

    # to speed up, using numpy function to compute pval in parallel.
    condlist = [ es < 0, es >=0]
    choicelist = [(esnull < es.reshape(len(es),1)).sum(axis=1)/ (esnull < 0).sum(axis=1),
                  (esnull >= es.reshape(len(es),1)).sum(axis=1)/ (esnull >= 0).sum(axis=1)]
    pvals = np.select(condlist, choicelist)

    return pvals 

def test_searched_case_column(batting, batting_df):
    t = batting
    df = batting_df
    expr = (
        ibis.case()
        .when(t.RBI < 5, 'really bad team')
        .when(t.teamID == 'PH1', 'ph1 team')
        .else_(t.teamID)
        .end()
    )
    result = expr.execute()
    expected = pd.Series(
        np.select(
            [df.RBI < 5, df.teamID == 'PH1'],
            ['really bad team', 'ph1 team'],
            df.teamID,
        )
    )
    tm.assert_series_equal(result, expected) 

def test_simple_case_column(batting, batting_df):
    t = batting
    df = batting_df
    expr = (
        t.RBI.case()
        .when(5, 'five')
        .when(4, 'four')
        .when(3, 'three')
        .else_('could be good?')
        .end()
    )
    result = expr.execute()
    expected = pd.Series(
        np.select(
            [df.RBI == 5, df.RBI == 4, df.RBI == 3],
            ['five', 'four', 'three'],
            'could be good?',
        )
    )
    tm.assert_series_equal(result, expected) 

def testSelect(self, rng_factory, shapes, dtypes):
    rng = rng_factory()
    n = len(dtypes) - 1
    def args_maker():
      condlist = [rng(shape, onp.bool_) for shape in shapes[:n]]
      choicelist = [rng(shape, dtype)
                    for shape, dtype in zip(shapes[n:-1], dtypes[:n])]
      default = rng(shapes[-1], dtypes[-1])
      return condlist, choicelist, default
    # TODO(phawkins): float32/float64 type mismatches
    def onp_fun(condlist, choicelist, default):
      choicelist = [x if lnp.bfloat16 != lnp.result_type(x)
                    else x.astype(onp.float32) for x in choicelist]
      dtype = lnp.result_type(default, *choicelist)
      return onp.select(condlist,
                        [onp.asarray(x, dtype=dtype) for x in choicelist],
                        onp.asarray(default, dtype=dtype))
    self._CheckAgainstNumpy(onp_fun, lnp.select, args_maker,
                            check_dtypes=False)
    self._CompileAndCheck(lnp.select, args_maker, check_dtypes=True,
                          check_incomplete_shape=True,
                          rtol={onp.float64: 1e-7, onp.complex128: 1e-7}) 

def check_quantifier_results(path):
    """Check quantifier vs results file in case of miscounts"""

    resfile = glob.glob(os.path.join(path,'result*.csv'))[0]
    df = read_results_file(resfile)
    files = glob.glob(os.path.join(path,'miRNAs_expressed_all_samples*.csv'))
    q = pd.read_csv(files[0],sep='\t')
    key='provisional id'
    m=q.merge(df,left_on='#miRNA',right_on=key).drop_duplicates('#miRNA')
    m.sc = m['miRDeep2 score']
    m['err'] = abs(m['read_count']-m['total read count'])
    cols=['#miRNA','total read count','read_count','miRDeep2 score']
    print (m[m.err>400].sort('total read count',ascending=False)[cols])
    m['size'] = np.select([m.sc < 2, m.sc < 3, m.sc < 4], [20,40,50], 80)
    f,ax=plt.subplots(1,1)
    plt.xscale('log')
    plt.yscale('log')
    m.plot(x='total read count',y='read_count', kind='scatter',s=60,alpha=0.6,ax=ax)
    #ax.plot([0, 1], [0, 1], transform=ax.transAxes,color='red',alpha=0.7)
    plt.show()
    return 

def rgb_to_hsv(rgb):
    # Translated from source of colorsys.rgb_to_hsv
    # r,g,b should be a numpy arrays with values between 0 and 255
    # rgb_to_hsv returns an array of floats between 0.0 and 1.0.
    rgb = rgb.astype('float')
    hsv = np.zeros_like(rgb)
    # in case an RGBA array was passed, just copy the A channel
    hsv[..., 3:] = rgb[..., 3:]
    r, g, b = rgb[..., 0], rgb[..., 1], rgb[..., 2]
    maxc = np.max(rgb[..., :3], axis=-1)
    minc = np.min(rgb[..., :3], axis=-1)
    hsv[..., 2] = maxc
    mask = maxc != minc
    hsv[mask, 1] = (maxc - minc)[mask] / maxc[mask]
    rc = np.zeros_like(r)
    gc = np.zeros_like(g)
    bc = np.zeros_like(b)
    rc[mask] = (maxc - r)[mask] / (maxc - minc)[mask]
    gc[mask] = (maxc - g)[mask] / (maxc - minc)[mask]
    bc[mask] = (maxc - b)[mask] / (maxc - minc)[mask]
    hsv[..., 0] = np.select([r == maxc, g == maxc], [bc - gc, 2.0 + rc - bc], default=4.0 + gc - rc)
    hsv[..., 0] = (hsv[..., 0] / 6.0) % 1.0
    return hsv 

def simulate_conditional(self, X):
    """ Draws random samples from the conditional distribution

    Args:
      X: x to be conditioned on when drawing a sample from y ~ p(y|x) - numpy array of shape (n_samples, ndim_x)

    Returns:
      Conditional random samples y drawn from p(y|x) - numpy array of shape (n_samples, ndim_y)
    """
    mean = self.arma_c * (1 - self.arma_a1) + self.arma_a1 * X
    y_ar = self.random_state.normal(loc=mean, scale=self.std, size=X.shape[0])

    mean_jump = mean + self.jump_mean
    y_jump = self.random_state.normal(loc=mean_jump, scale=self.jump_std, size=X.shape[0])

    jump_bernoulli = self.random_state.uniform(size=X.shape[0]) < self.jump_prob

    return X, np.select([jump_bernoulli, np.bitwise_not(jump_bernoulli)], [y_jump, y_ar]) 

def evaluate(x, y, amplitude, x_0, y_0, a, b, theta):
        """Two dimensional Ellipse model function."""

        xx = x - x_0
        yy = y - y_0
        cost = np.cos(theta)
        sint = np.sin(theta)
        numerator1 = (xx * cost) + (yy * sint)
        numerator2 = -(xx * sint) + (yy * cost)
        in_ellipse = (((numerator1 / a) ** 2 + (numerator2 / b) ** 2) <= 1.)
        result = np.select([in_ellipse], [amplitude])

        if isinstance(amplitude, Quantity):
            return Quantity(result, unit=amplitude.unit, copy=False)
        else:
            return result 

def _lazyselect(condlist, choicelist, arrays, default=0):
    """
    Mimic `np.select(condlist, choicelist)`.

    Notice it assumes that all `arrays` are of the same shape, or can be
    broadcasted together.

    All functions in `choicelist` must accept array arguments in the order
    given in `arrays` and must return an array of the same shape as broadcasted
    `arrays`.

    Examples
    --------
    >>> x = np.arange(6)
    >>> np.select([x <3, x > 3], [x**2, x**3], default=0)
    array([  0,   1,   4,   0,  64, 125])

    >>> _lazyselect([x < 3, x > 3], [lambda x: x**2, lambda x: x**3], (x,))
    array([   0.,    1.,    4.,   0.,   64.,  125.])

    >>> a = -np.ones_like(x)
    >>> _lazyselect([x < 3, x > 3],
    ...             [lambda x, a: x**2, lambda x, a: a * x**3],
    ...             (x, a), default=np.nan)
    array([   0.,    1.,    4.,   nan,  -64., -125.])

    """
    arrays = np.broadcast_arrays(*arrays)
    tcode = np.mintypecode([a.dtype.char for a in arrays])
    out = _valarray(np.shape(arrays[0]), value=default, typecode=tcode)
    for index in range(len(condlist)):
        func, cond = choicelist[index], condlist[index]
        if np.all(cond is False):
            continue
        cond, _ = np.broadcast_arrays(cond, arrays[0])
        temp = tuple(np.extract(cond, arr) for arr in arrays)
        np.place(out, cond, func(*temp))
    return out 

def _pdf(self, x, c, d):
        u = 2 / (d - c + 1)

        condlist = [x < c, x <= d, x > d]
        choicelist = [u * x / c, u, u * (1 - x) / (1 - d)]
        return np.select(condlist, choicelist) 

def _cdf(self, x, c, d):
        condlist = [x < c, x <= d, x > d]
        choicelist = [x**2 / c / (d - c + 1),
                      (c + 2 * (x - c)) / (d - c + 1),
                      1 - ((1 - x)**2 / (d - c + 1) / (1 - d))]
        return np.select(condlist, choicelist) 

def _ppf(self, q, c, d):
        qc, qd = self._cdf(c, c, d), self._cdf(d, c, d)
        condlist = [q < qc, q <= qd, q > qd]
        choicelist = [np.sqrt(q * c * (1 + d - c)),
                      0.5 * q * (1 + d - c) + 0.5 * c,
                      1 - np.sqrt((1 - q) * (d - c + 1) * (1 - d))]
        return np.select(condlist, choicelist) 

def _pmf(self, x):
        return np.select([x == k for k in self.xk],
                         [np.broadcast_arrays(p, x)[0] for p in self.pk], 0) 

def apply_slippage(terrain, repose_slope, cell_width):
  delta = util.simple_gradient(terrain) / cell_width
  smoothed = util.gaussian_blur(terrain, sigma=1.5)
  should_smooth = np.abs(delta) > repose_slope
  result = np.select([np.abs(delta) > repose_slope], [smoothed], terrain)
  return result 

def test_cdf(self):
        x = numpy.r_[0:36:100j]
        k = numpy.floor(x)
        out = numpy.select([k >= 30,k >= 5],[1.0,(k-5.0+1)/(30-5.0)],0)
        vals = stats.randint.cdf(x,5,30)
        assert_array_almost_equal(vals, out, decimal=12) 

def execute_searched_case(op, whens, thens, otherwise, **kwargs):
    if otherwise is None:
        otherwise = np.nan
    raw = np.select(whens, thens, otherwise)
    return wrap_case_result(raw, op.to_expr()) 

def execute_simple_case_scalar(op, value, whens, thens, otherwise, **kwargs):
    if otherwise is None:
        otherwise = np.nan
    raw = np.select(np.asarray(whens) == value, thens, otherwise)
    return wrap_case_result(raw, op.to_expr()) 

def execute_simple_case_series(op, value, whens, thens, otherwise, **kwargs):
    if otherwise is None:
        otherwise = np.nan
    raw = np.select([value == when for when in whens], thens, otherwise)
    return wrap_case_result(raw, op.to_expr()) 

def get_reward(self):
        reward = (self.current_entities == self.end_entities)

        # set the True and False values to the values of positive and negative rewards.
        condlist = [reward == True, reward == False]
        choicelist = [self.positive_reward, self.negative_reward]
        reward = np.select(condlist, choicelist)  # [B,]
        return reward 

def test_two_dimentional(dtype):
    """check two dimentation sparsemax case."""
    t = np.linspace(-2, 2, test_obs, dtype=dtype)
    z = np.vstack([t, np.zeros(test_obs, dtype=dtype)]).T

    tf_sparsemax_out = sparsemax(z.astype(dtype)).numpy()

    p0_expected = np.select([t < -1, t <= 1, t > 1], [0, (t + 1) / 2, 1])

    test_utils.assert_allclose_according_to_type(p0_expected, tf_sparsemax_out[:, 0])
    test_utils.assert_allclose_according_to_type(
        1 - p0_expected, tf_sparsemax_out[:, 1]
    )
    assert z.shape == tf_sparsemax_out.shape 

def select(condlist, choicelist, default=0):  # pylint: disable=missing-docstring
  if len(condlist) != len(choicelist):
    msg = 'condlist must have length equal to choicelist ({} vs {})'
    raise ValueError(msg.format(len(condlist), len(choicelist)))
  if not condlist:
    raise ValueError('condlist must be non-empty')
  choices = _promote_dtype(default, *choicelist)
  choicelist = choices[1:]
  output = choices[0]
  # The traversal is in reverse order so we can return the first value in
  # choicelist where condlist is True.
  for cond, choice in zip(condlist[::-1], choicelist[::-1]):
    output = where(cond, choice, output)
  return output 

def _lazyselect(condlist, choicelist, arrays, default=0):
    """
    Mimic `np.select(condlist, choicelist)`.

    Notice it assumes that all `arrays` are of the same shape, or can be
    broadcasted together.

    All functions in `choicelist` must accept array arguments in the order
    given in `arrays` and must return an array of the same shape as broadcasted
    `arrays`.

    Examples
    --------
    >>> x = np.arange(6)
    >>> np.select([x <3, x > 3], [x**2, x**3], default=0)
    array([  0,   1,   4,   0,  64, 125])

    >>> _lazyselect([x < 3, x > 3], [lambda x: x**2, lambda x: x**3], (x,))
    array([   0.,    1.,    4.,   0.,   64.,  125.])

    >>> a = -np.ones_like(x)
    >>> _lazyselect([x < 3, x > 3],
    ...             [lambda x, a: x**2, lambda x, a: a * x**3],
    ...             (x, a), default=np.nan)
    array([   0.,    1.,    4.,   nan,  -64., -125.])

    """
    arrays = np.broadcast_arrays(*arrays)
    tcode = np.mintypecode([a.dtype.char for a in arrays])
    out = _valarray(np.shape(arrays[0]), value=default, typecode=tcode)
    for index in range(len(condlist)):
        func, cond = choicelist[index], condlist[index]
        if np.all(cond is False):
            continue
        cond, _ = np.broadcast_arrays(cond, arrays[0])
        temp = tuple(np.extract(cond, arr) for arr in arrays)
        np.place(out, cond, func(*temp))
    return out 

def _ppf(self, q, c, d):
        qc, qd = self._cdf(c, c, d), self._cdf(d, c, d)
        condlist = [q < qc, q <= qd, q > qd]
        choicelist = [np.sqrt(q * c * (1 + d - c)),
                      0.5 * q * (1 + d - c) + 0.5 * c,
                      1 - np.sqrt((1 - q) * (d - c + 1) * (1 - d))]
        return np.select(condlist, choicelist) 

def _pmf(self, x):
        return np.select([x == k for k in self.xk],
                         [np.broadcast_arrays(p, x)[0] for p in self.pk], 0) 

def test_cdf(self):
        x = numpy.r_[0:36:100j]
        k = numpy.floor(x)
        out = numpy.select([k >= 30, k >= 5], [1.0, (k-5.0+1)/(30-5.0)], 0)
        vals = stats.randint.cdf(x, 5, 30)
        assert_array_almost_equal(vals, out, decimal=12) 

def take(a, indices, axis=None, out=None):
    """Takes elements of an array at specified indices along an axis.

    This is an implementation of "fancy indexing" at single axis.

    This function does not support ``mode`` option.

    Args:
        a (cupy.ndarray): Array to extract elements.
        indices (int or array-like): Indices of elements that this function
            takes.
        axis (int): The axis along which to select indices. The flattened input
            is used by default.
        out (cupy.ndarray): Output array. If provided, it should be of
            appropriate shape and dtype.

    Returns:
        cupy.ndarray: The result of fancy indexing.

    .. seealso:: :func:`numpy.take`

    """
    # TODO(okuta): check type
    return a.take(indices, axis, out) 

def ive(v, z):
    """Exponentially scaled modified Bessel function of the first kind."""
    output = array_ops.reshape(script_ops.py_func(
        lambda v, z: np.select(condlist=[v == 0, v == 1],
                               choicelist=[scipy.special.i0e(z, dtype=z.dtype),
                                           scipy.special.i1e(z, dtype=z.dtype)],
                               default=scipy.special.ive(v, z, dtype=z.dtype)), [v, z], z.dtype),
        ops.convert_to_tensor(array_ops.shape(z), dtype=dtypes.int32))

    def grad(dy):
        return None, dy * (ive(v - 1, z) - ive(v, z) * (v + z) / z)

    return output, grad 

def backward_var(self, grad, index, **kwargs):
        # d arcsin / dx at x = -1, 1 returns 0, not NaN
        a = self.variables[index]
        return np.select([np.abs(a.data) != 1], [grad / np.sqrt(1 - a.data ** 2)]) 

def backward_var(self, grad, index, **kwargs):
        # d arccos / dx at x = -1, 1 returns 0, not NaN
        a = self.variables[index]
        return np.select([np.abs(a.data) != 1], [-grad / np.sqrt(1 - a.data ** 2)])