Python numexpr.evaluate方法代码示例

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def evaluate(self):
        """Run the engine on the expression

        This method performs alignment which is necessary no matter what engine
        is being used, thus its implementation is in the base class.

        Returns
        -------
        obj : object
            The result of the passed expression.
        """
        if not self._is_aligned:
            self.result_type, self.aligned_axes = _align(self.expr.terms)

        # make sure no names in resolvers and locals/globals clash
        res = self._evaluate()
        return _reconstruct_object(self.result_type, res, self.aligned_axes,
                                   self.expr.terms.return_type) 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def _evaluate(self):
        import numexpr as ne

        # convert the expression to a valid numexpr expression
        s = self.convert()

        try:
            env = self.expr.env
            scope = env.full_scope
            truediv = scope['truediv']
            _check_ne_builtin_clash(self.expr)
            return ne.evaluate(s, local_dict=scope, truediv=truediv)
        except KeyError as e:
            # python 3 compat kludge
            try:
                msg = e.message
            except AttributeError:
                msg = compat.text_type(e)
            raise UndefinedVariableError(msg) 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def evaluate(op, op_str, a, b, use_numexpr=True,
             **eval_kwargs):
    """ evaluate and return the expression of the op on a and b

        Parameters
        ----------

        op :    the actual operand
        op_str: the string version of the op
        a :     left operand
        b :     right operand
        use_numexpr : whether to try to use numexpr (default True)
        """

    use_numexpr = use_numexpr and _bool_arith_check(op_str, a, b)
    if use_numexpr:
        return _evaluate(op, op_str, a, b, **eval_kwargs)
    return _evaluate_standard(op, op_str, a, b) 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def planck_f(f, T):
    """Planck law expressed in frequency.

    If more than 10⁵ resulting radiances, uses numexpr.

    :param f: Frequency.  Quantity in [Hz]
    :param T: Temperature.  Quantity in [K]
    """
#    try:
#        f = f.astype(numpy.float64)
#    except AttributeError:
#        pass
    if (f.size * T.size) > 1e5:
        return numexpr.evaluate("(2 * h * f**3) / (c**2) * "
                                "1 / (exp((h*f)/(k*T)) - 1)") * (
                                    radiance_units["si"])
    return ((2 * ureg.h * f**3) / (ureg.c ** 2) *
            1 / (numpy.exp(((ureg.h * f) / (ureg.k * T)).to("1")) - 1)).to(
                ureg.W / (ureg.m**2 * ureg.sr * ureg.Hz)) 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def execute(cls, ctx, op):
        chunk = op.outputs[0]
        inputs = as_same_device([ctx[c.key] for c in op.inputs], device=op.device)
        for c, i in zip(op.inputs, inputs):
            exec('V_' + c.key + ' = i')
        expr = _evaluate(chunk)
        if cls._lock is not None:
            cls._lock.acquire()
        try:
            res = ne.evaluate(expr)
        finally:
            if cls._lock is not None:
                cls._lock.release()
        res = _maybe_keepdims(chunk, res)
        if chunk.ndim == 0 and res.ndim == 1 and res.size == 0:
            res = res.dtype.type(0)
        ctx[chunk.key] = res 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def fields(x,y,z, kx, ky, kz, B0):
    k1 =  -B0*kx/ky
    k2 = -B0*kz/ky
    kx_x = kx*x
    ky_y = ky*y
    kz_z = kz*z
    cosx = np.cos(kx_x)
    sinhy = np.sinh(ky_y)
    cosz = np.cos(kz_z)
    Bx = k1*np.sin(kx_x)*sinhy*cosz #// here kx is only real
    By = B0*cosx*np.cosh(ky_y)*cosz
    Bz = k2*cosx*sinhy*np.sin(kz_z)
    #Bx = ne.evaluate("k1*sin(kx*x)*sinhy*cosz")
    #By = ne.evaluate("B0*cosx*cosh(ky*y)*cosz")
    #Bz = ne.evaluate("k2*cosx*sinhy*sin(kz*z)")
    return Bx, By, Bz 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def sym_kernel(self, ijk2, hxyz):
        i2 = ijk2[0]
        j2 = ijk2[1]
        k2 = ijk2[2]
        hx = hxyz[0]
        hy = hxyz[1]
        hz = hxyz[2]
        x = hx*np.r_[0:i2+1] - hx/2
        y = hy*np.r_[0:j2+1] - hy/2
        z = hz*np.r_[0:k2+1] - hz/2
        x, y, z = np.ix_(x, y, z)
        r = np.sqrt(x*x + y*y + z*z)
        if ne_flag:
            IG = ne.evaluate("(-x*x*0.5*arctan(y*z/(x*r)) + y*z*log(x+r) - y*y*0.5*arctan(z*x/(y*r)) + z*x*log(y+r) - z*z*0.5*arctan(x*y/(z*r)) + x*y*log(z+r))")
        else:
            IG = (-x*x*0.5*np.arctan(y*z/(x*r)) + y*z*np.log(x+r)
                  -y*y*0.5*np.arctan(z*x/(y*r)) + z*x*np.log(y+r)
                  -z*z*0.5*np.arctan(x*y/(z*r)) + x*y*np.log(z+r))

        kern = (IG[1:i2+1, 1:j2+1, 1:k2+1] - IG[0:i2, 1:j2+1, 1:k2+1]
               -IG[1:i2+1, 0:j2, 1:k2+1] + IG[0:i2, 0:j2, 1:k2+1]
               -IG[1:i2+1, 1:j2+1, 0:k2] + IG[0:i2, 1:j2+1, 0:k2]
               +IG[1:i2+1, 0:j2, 0:k2] - IG[0:i2, 0:j2, 0:k2])

        return kern 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def numexpr_apply(self, X, R, T):
        x, px, y, py, tau, dp = np.copy((X[0], X[1], X[2], X[3], X[4], X[5]))
        R00, R01, R02, R03, R04, R05 = R[0, 0], R[0, 1], R[0, 2], R[0, 3], R[0, 4], R[0, 5]
        R10, R11, R12, R13, R14, R15 = R[1, 0], R[1, 1], R[1, 2], R[1, 3], R[1, 4], R[1, 5]
        R20, R21, R22, R23, R24, R25 = R[2, 0], R[2, 1], R[2, 2], R[2, 3], R[2, 4], R[2, 5]
        R30, R31, R32, R33, R34, R35 = R[3, 0], R[3, 1], R[3, 2], R[3, 3], R[3, 4], R[3, 5]
        R40, R41, R42, R43, R44, R45 = R[4, 0], R[4, 1], R[4, 2], R[4, 3], R[4, 4], R[4, 5]
        R50, R51, R52, R53, R54, R55 = R[5, 0], R[5, 1], R[5, 2], R[5, 3], R[5, 4], R[5, 5]

        T000, T001, T005, T011, T015, T055, T022, T023, T033 = T[0, 0, 0], T[0, 0, 1], T[0, 0, 5], T[0, 1, 1], T[0, 1, 5], T[0, 5, 5], T[0, 2, 2],T[0, 2, 3], T[0, 3, 3]
        T100, T101, T105, T111, T115, T155, T122, T123, T133 = T[1, 0, 0], T[1, 0, 1], T[1, 0, 5], T[1, 1, 1], T[1, 1, 5], T[1, 5, 5], T[1, 2, 2],T[1, 2, 3], T[1, 3, 3]
        T202, T203, T212, T213, T225, T235 = T[2, 0, 2],  T[2, 0, 3],  T[2, 1, 2],  T[2, 1, 3], T[2, 2, 5], T[2, 3, 5]
        T302, T303, T312, T313, T325, T335 = T[3, 0, 2],  T[3, 0, 3],  T[3, 1, 2],  T[3, 1, 3], T[3, 2, 5], T[3, 3, 5]
        T400, T401, T405, T411, T415, T455, T422, T423, T433 = T[4, 0, 0], T[4, 0, 1], T[4, 0, 5], T[4, 1, 1], T[4, 1, 5], T[4, 5, 5], T[4, 2, 2], T[4, 2, 3], T[4, 3, 3]

        X[0] = ne.evaluate('R00 * x + R01 * px + R02 * y + R03 * py + R04 * tau + R05 * dp + T000 * x*x + T001 * x*px + T005 * x*dp + T011 * px*px + T015 * px*dp + T055 * dp*dp + T022 * y*y + T023 * y*py + T033 * py*py')
        X[1] = ne.evaluate('R10 * x + R11 * px + R12 * y + R13 * py + R14 * tau + R15 * dp + T100 * x*x + T101 * x*px + T105 * x*dp + T111 * px*px + T115 * px*dp + T155 * dp*dp + T122 * y*y + T123 * y*py + T133 * py*py')
        X[2] = ne.evaluate('R20 * x + R21 * px + R22 * y + R23 * py + R24 * tau + R25 * dp + T202 * x*y + T203 * x*py + T212 * y*px + T213 * px*py + T225 * y*dp + T235 * py*dp')
        X[3] = ne.evaluate('R30 * x + R31 * px + R32 * y + R33 * py + R34 * tau + R35 * dp + T302 * x*y + T303 * x*py + T312 * y*px + T313 * px*py + T325 * y*dp + T335 * py*dp')
        X[4] = ne.evaluate('R40 * x + R41 * px + R42 * y + R43 * py + R44 * tau + R45 * dp + T400 * x*x + T401 * x*px + T405 * x*dp + T411 * px*px + T415 * px*dp + T455 * dp*dp + T422 * y*y + T423 * y*py + T433 * py*py')  # + U5666*dp2*dp    # third order 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def evaluate(self):
        """Run the engine on the expression

        This method performs alignment which is necessary no matter what engine
        is being used, thus its implementation is in the base class.

        Returns
        -------
        obj : object
            The result of the passed expression.
        """
        if not self._is_aligned:
            self.result_type, self.aligned_axes = _align(self.expr.terms)

        # make sure no names in resolvers and locals/globals clash
        self.pre_evaluate()
        res = self._evaluate()
        return _reconstruct_object(self.result_type, res, self.aligned_axes,
                                   self.expr.terms.return_type) 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def _evaluate(self):
        import numexpr as ne

        # add the resolvers to locals
        self.expr.add_resolvers_to_locals()

        # convert the expression to a valid numexpr expression
        s = self.convert()

        try:
            return ne.evaluate(s, local_dict=self.expr.env.locals,
                               global_dict=self.expr.env.globals,
                               truediv=self.expr.truediv)
        except KeyError as e:
            # python 3 compat kludge
            try:
                msg = e.message
            except AttributeError:
                msg = compat.text_type(e)
            raise UndefinedVariableError(msg) 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def evaluate(op, op_str, a, b, raise_on_error=False, use_numexpr=True,
             **eval_kwargs):
    """ evaluate and return the expression of the op on a and b

        Parameters
        ----------

        op :    the actual operand
        op_str: the string version of the op
        a :     left operand
        b :     right operand
        raise_on_error : pass the error to the higher level if indicated
                         (default is False), otherwise evaluate the op with and
                         return the results
        use_numexpr : whether to try to use numexpr (default True)
        """

    if use_numexpr:
        return _evaluate(op, op_str, a, b, raise_on_error=raise_on_error,
                         **eval_kwargs)
    return _evaluate_standard(op, op_str, a, b, raise_on_error=raise_on_error) 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def where(cond, a, b, raise_on_error=False, use_numexpr=True):
    """ evaluate the where condition cond on a and b

        Parameters
        ----------

        cond : a boolean array
        a :    return if cond is True
        b :    return if cond is False
        raise_on_error : pass the error to the higher level if indicated
                         (default is False), otherwise evaluate the op with and
                         return the results
        use_numexpr : whether to try to use numexpr (default True)
        """

    if use_numexpr:
        return _where(cond, a, b, raise_on_error=raise_on_error)
    return _where_standard(cond, a, b, raise_on_error=raise_on_error) 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def dominant_screen_color(initial_color, func_bounds=lambda: None):
    """
    https://stackoverflow.com/questions/50899692/most-dominant-color-in-rgb-image-opencv-numpy-python
    """
    monitor = get_monitor_bounds(func_bounds)
    if "full" in monitor:
        screenshot = getScreenAsImage()
    else:
        screenshot = getRectAsImage(str2list(monitor, int))

    downscale_width, downscale_height = screenshot.width // 4, screenshot.height // 4
    screenshot = screenshot.resize((downscale_width, downscale_height), Image.HAMMING)

    a = np.array(screenshot)
    a2D = a.reshape(-1, a.shape[-1])
    col_range = (256, 256, 256)  # generically : a2D.max(0)+1
    eval_params = {'a0': a2D[:, 0], 'a1': a2D[:, 1], 'a2': a2D[:, 2],
                   's0': col_range[0], 's1': col_range[1]}
    a1D = ne.evaluate('a0*s0*s1+a1*s0+a2', eval_params)
    color = np.unravel_index(np.bincount(a1D).argmax(), col_range)

    color_hsbk = list(utils.RGBtoHSBK(color, temperature=initial_color[3]))
    # color_hsbk[2] = initial_color[2]  # TODO Decide this
    return color_hsbk 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def normalize_mi_ma(x, mi, ma, clip=False, eps=1e-20, dtype=np.float32):
    if dtype is not None:
        x   = x.astype(dtype,copy=False)
        mi  = dtype(mi) if np.isscalar(mi) else mi.astype(dtype,copy=False)
        ma  = dtype(ma) if np.isscalar(ma) else ma.astype(dtype,copy=False)
        eps = dtype(eps)

    try:
        import numexpr
        x = numexpr.evaluate("(x - mi) / ( ma - mi + eps )")
    except ImportError:
        x =                   (x - mi) / ( ma - mi + eps )

    if clip:
        x = np.clip(x,0,1)

    return x 

# 需要导入模块: import numexpr [as 别名]
# 或者: from numexpr import evaluate [as 别名]
def _compute(self, arrays, dates, assets, mask):
        """
        Compute our stored expression string with numexpr.
        """
        out = full(mask.shape, self.missing_value, dtype=self.dtype)
        # This writes directly into our output buffer.
        numexpr.evaluate(
            self._expr,
            local_dict={
                "x_%d" % idx: array
                for idx, array in enumerate(arrays)
            },
            global_dict={'inf': inf},
            out=out,
        )
        return out