HARKinterpolation.py

'''
Custom interpolation methods for representing approximations to functions.
It also includes wrapper classes to enforce standard methods across classes.
Each interpolation class must have a distance() method that compares itself to
another instance; this is used in HARKcore's solve() method to check for solution
convergence.  The interpolator classes currently in this module inherit their
distance method from HARKobject.
'''

import warnings
import numpy as np
from scipy.interpolate import UnivariateSpline
from HARKcore import HARKobject
from copy import deepcopy

def _isscalar(x):
    '''
    Check whether x is if a scalar type, or 0-dim.
    
    Parameters
    ----------
    x : anything
        An input to be checked for scalar-ness.
        
    Returns
    -------
    is_scalar : boolean
        True if the input is a scalar, False otherwise.
    '''
    return np.isscalar(x) or hasattr(x, 'shape') and x.shape == ()


class HARKinterpolator1D(HARKobject):
    '''
    A wrapper class for 1D interpolation methods in HARK.
    '''
    distance_criteria = []
    
    def __call__(self,x):
        '''
        Evaluates the interpolated function at the given input.
        
        Parameters
        ----------
        x : np.array or float
            Real values to be evaluated in the interpolated function.
        
        Returns
        -------
        y : np.array or float
            The interpolated function evaluated at x: y = f(x), with the same
            shape as x.
        '''
        z = np.asarray(x)
        return (self._evaluate(z.flatten())).reshape(z.shape)
        
    def derivative(self,x):
        '''
        Evaluates the derivative of the interpolated function at the given input.
        
        Parameters
        ----------
        x : np.array or float
            Real values to be evaluated in the interpolated function.
        
        Returns
        -------
        dydx : np.array or float
            The interpolated function's first derivative evaluated at x:
            dydx = f'(x), with the same shape as x.
        '''
        z = np.asarray(x)
        return (self._der(z.flatten())).reshape(z.shape)
        
    def eval_with_derivative(self,x):
        '''
        Evaluates the interpolated function and its derivative at the given input.
        
        Parameters
        ----------
        x : np.array or float
            Real values to be evaluated in the interpolated function.
        
        Returns
        -------
        y : np.array or float
            The interpolated function evaluated at x: y = f(x), with the same
            shape as x.
        dydx : np.array or float
            The interpolated function's first derivative evaluated at x:
            dydx = f'(x), with the same shape as x.
        '''
        z = np.asarray(x)
        y, dydx = self._evalAndDer(z.flatten())
        return y.reshape(z.shape), dydx.reshape(z.shape)
        
    def _evaluate(self,x):
        '''
        Interpolated function evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()
        
    def _der(self,x):
        '''
        Interpolated function derivative evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()
        
    def _evalAndDer(self,x):
        '''
        Interpolated function and derivative evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()

       
class HARKinterpolator2D(HARKobject):
    '''
    A wrapper class for 2D interpolation methods in HARK.
    '''
    distance_criteria = []
    
    def __call__(self,x,y):
        '''
        Evaluates the interpolated function at the given input.
        
        Parameters
        ----------
        x : np.array or float
            Real values to be evaluated in the interpolated function.
        y : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as x.
        
        Returns
        -------
        fxy : np.array or float
            The interpolated function evaluated at x,y: fxy = f(x,y), with the
            same shape as x and y.
        '''
        xa = np.asarray(x)
        ya = np.asarray(y)
        return (self._evaluate(xa.flatten(),ya.flatten())).reshape(xa.shape)
        
    def derivativeX(self,x,y):
        '''
        Evaluates the partial derivative of interpolated function with respect
        to x (the first argument) at the given input.
        
        Parameters
        ----------
        x : np.array or float
            Real values to be evaluated in the interpolated function.
        y : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as x.
        
        Returns
        -------
        dfdx : np.array or float
            The derivative of the interpolated function with respect to x, eval-
            uated at x,y: dfdx = f_x(x,y), with the same shape as x and y.
        '''
        xa = np.asarray(x)
        ya = np.asarray(y)
        return (self._derX(xa.flatten(),ya.flatten())).reshape(xa.shape)
        
    def derivativeY(self,x,y):
        '''
        Evaluates the partial derivative of interpolated function with respect
        to y (the second argument) at the given input.
        
        Parameters
        ----------
        x : np.array or float
            Real values to be evaluated in the interpolated function.
        y : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as x.
        
        Returns
        -------
        dfdy : np.array or float
            The derivative of the interpolated function with respect to y, eval-
            uated at x,y: dfdx = f_y(x,y), with the same shape as x and y.
        '''
        xa = np.asarray(x)
        ya = np.asarray(y)
        return (self._derY(xa.flatten(),ya.flatten())).reshape(xa.shape)
        
    def _evaluate(self,x,y):
        '''
        Interpolated function evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()
        
    def _derX(self,x,y):
        '''
        Interpolated function x-derivative evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()
        
    def _derY(self,x,y):
        '''
        Interpolated function y-derivative evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()

        
class HARKinterpolator3D(HARKobject):
    '''
    A wrapper class for 3D interpolation methods in HARK.
    '''
    distance_criteria = []
    
    def __call__(self,x,y,z):
        '''
        Evaluates the interpolated function at the given input.
        
        Parameters
        ----------
        x : np.array or float
            Real values to be evaluated in the interpolated function.
        y : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as x.
        z : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as x.
        
        Returns
        -------
        fxyz : np.array or float
            The interpolated function evaluated at x,y,z: fxyz = f(x,y,z), with
            the same shape as x, y, and z.
        '''
        xa = np.asarray(x)
        ya = np.asarray(y)
        za = np.asarray(z)
        return (self._evaluate(xa.flatten(),ya.flatten(),za.flatten())).reshape(xa.shape)
        
    def derivativeX(self,x,y,z):
        '''
        Evaluates the partial derivative of the interpolated function with respect
        to x (the first argument) at the given input.
        
        Parameters
        ----------
        x : np.array or float
            Real values to be evaluated in the interpolated function.
        y : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as x.
        z : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as x.
        
        Returns
        -------
        dfdx : np.array or float
            The derivative with respect to x of the interpolated function evaluated
            at x,y,z: dfdx = f_x(x,y,z), with the same shape as x, y, and z.
        '''
        xa = np.asarray(x)
        ya = np.asarray(y)
        za = np.asarray(z)
        return (self._derX(xa.flatten(),ya.flatten(),za.flatten())).reshape(xa.shape)
        
    def derivativeY(self,x,y,z):
        '''
        Evaluates the partial derivative of the interpolated function with respect
        to y (the second argument) at the given input.
        
        Parameters
        ----------
        x : np.array or float
            Real values to be evaluated in the interpolated function.
        y : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as x.
        z : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as x.
        
        Returns
        -------
        dfdy : np.array or float
            The derivative with respect to y of the interpolated function evaluated
            at x,y,z: dfdy = f_y(x,y,z), with the same shape as x, y, and z.
        '''
        xa = np.asarray(x)
        ya = np.asarray(y)
        za = np.asarray(z)
        return (self._derY(xa.flatten(),ya.flatten(),za.flatten())).reshape(xa.shape)
        
    def derivativeZ(self,x,y,z):
        '''
        Evaluates the partial derivative of the interpolated function with respect
        to z (the third argument) at the given input.
        
        Parameters
        ----------
        x : np.array or float
            Real values to be evaluated in the interpolated function.
        y : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as x.
        z : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as x.
        
        Returns
        -------
        dfdz : np.array or float
            The derivative with respect to z of the interpolated function evaluated
            at x,y,z: dfdz = f_z(x,y,z), with the same shape as x, y, and z.
        '''
        xa = np.asarray(x)
        ya = np.asarray(y)
        za = np.asarray(z)
        return (self._derZ(xa.flatten(),ya.flatten(),za.flatten())).reshape(xa.shape)
        
    def _evaluate(self,x,y,z):
        '''
        Interpolated function evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()
        
    def _derX(self,x,y,z):
        '''
        Interpolated function x-derivative evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()
        
    def _derY(self,x,y,z):
        '''
        Interpolated function y-derivative evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()
        
    def _derZ(self,x,y,z):
        '''
        Interpolated function y-derivative evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()
        
        
class HARKinterpolator4D(HARKobject):
    '''
    A wrapper class for 4D interpolation methods in HARK.
    '''
    distance_criteria = []
    
    def __call__(self,w,x,y,z):
        '''
        Evaluates the interpolated function at the given input.
        
        Parameters
        ----------
        w : np.array or float
            Real values to be evaluated in the interpolated function.
        x : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        y : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        z : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        
        Returns
        -------
        fwxyz : np.array or float
            The interpolated function evaluated at w,x,y,z: fwxyz = f(w,x,y,z),
            with the same shape as w, x, y, and z.
        '''
        wa = np.asarray(w)
        xa = np.asarray(x)
        ya = np.asarray(y)
        za = np.asarray(z)
        return (self._evaluate(wa.flatten(),xa.flatten(),ya.flatten(),za.flatten())).reshape(wa.shape)

    def derivativeW(self,w,x,y,z):
        '''
        Evaluates the partial derivative with respect to w (the first argument)
        of the interpolated function at the given input.
        
        Parameters
        ----------
        w : np.array or float
            Real values to be evaluated in the interpolated function.
        x : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        y : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        z : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        
        Returns
        -------
        dfdw : np.array or float
            The derivative with respect to w of the interpolated function eval-
            uated at w,x,y,z: dfdw = f_w(w,x,y,z), with the same shape as inputs.
        '''
        wa = np.asarray(w)
        xa = np.asarray(x)
        ya = np.asarray(y)
        za = np.asarray(z)
        return (self._derW(wa.flatten(),xa.flatten(),ya.flatten(),za.flatten())).reshape(wa.shape)

    def derivativeX(self,w,x,y,z):
        '''
        Evaluates the partial derivative with respect to x (the second argument)
        of the interpolated function at the given input.
        
        Parameters
        ----------
        w : np.array or float
            Real values to be evaluated in the interpolated function.
        x : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        y : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        z : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        
        Returns
        -------
        dfdx : np.array or float
            The derivative with respect to x of the interpolated function eval-
            uated at w,x,y,z: dfdx = f_x(w,x,y,z), with the same shape as inputs.
        '''
        wa = np.asarray(w)
        xa = np.asarray(x)
        ya = np.asarray(y)
        za = np.asarray(z)
        return (self._derX(wa.flatten(),xa.flatten(),ya.flatten(),za.flatten())).reshape(wa.shape)
        
    def derivativeY(self,w,x,y,z):
        '''
        Evaluates the partial derivative with respect to y (the third argument)
        of the interpolated function at the given input.
        
        Parameters
        ----------
        w : np.array or float
            Real values to be evaluated in the interpolated function.
        x : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        y : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        z : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        
        Returns
        -------
        dfdy : np.array or float
            The derivative with respect to y of the interpolated function eval-
            uated at w,x,y,z: dfdy = f_y(w,x,y,z), with the same shape as inputs.
        '''
        wa = np.asarray(w)
        xa = np.asarray(x)
        ya = np.asarray(y)
        za = np.asarray(z)
        return (self._derY(wa.flatten(),xa.flatten(),ya.flatten(),za.flatten())).reshape(wa.shape)

    def derivativeZ(self,w,x,y,z):
        '''
        Evaluates the partial derivative with respect to z (the fourth argument)
        of the interpolated function at the given input.
        
        Parameters
        ----------
        w : np.array or float
            Real values to be evaluated in the interpolated function.
        x : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        y : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        z : np.array or float
            Real values to be evaluated in the interpolated function; must be
            the same size as w.
        
        Returns
        -------
        dfdz : np.array or float
            The derivative with respect to z of the interpolated function eval-
            uated at w,x,y,z: dfdz = f_z(w,x,y,z), with the same shape as inputs.
        '''
        wa = np.asarray(w)
        xa = np.asarray(x)
        ya = np.asarray(y)
        za = np.asarray(z)
        return (self._derZ(wa.flatten(),xa.flatten(),ya.flatten(),za.flatten())).reshape(wa.shape)
        
    def _evaluate(self,w,x,y,z):
        '''
        Interpolated function evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()
        
    def _derW(self,w,x,y,z):
        '''
        Interpolated function w-derivative evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()
        
    def _derX(self,w,x,y,z):
        '''
        Interpolated function w-derivative evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()
        
    def _derY(self,w,x,y,z):
        '''
        Interpolated function w-derivative evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()
        
    def _derZ(self,w,x,y,z):
        '''
        Interpolated function w-derivative evaluator, to be defined in subclasses.
        '''
        raise NotImplementedError()
        
        
class ConstantFunction(HARKobject):
    '''
    A class for representing trivial functions that return the same real output for any input.  This
    is convenient for models where an object might be a (non-trivial) function, but in some variations
    that object is just a constant number.  Rather than needing to make a (Bi/Tri/Quad)-
    LinearInterpolation with trivial state grids and the same f_value in every entry, ConstantFunction
    allows the user to quickly make a constant/trivial function.  This comes up, e.g., in models
    with endogenous pricing of insurance contracts; a contract's premium might depend on some state
    variables of the individual, but in some variations the premium of a contract is just a number.
    '''
    convergence_criteria = ['value']
    
    def __init__(self,value):
        '''
        Make a new ConstantFunction object.
        
        Parameters
        ----------
        value : float
            The constant value that the function returns.
        
        Returns
        -------
        None
        '''
        self.value = float(value)
        
    def __call__(self,*args):
        '''
        Evaluate the constant function.  The first input must exist and should be an array.
        Returns an array of identical shape to args[0] (if it exists).
        '''
        if len(args) > 0: # If there is at least one argument, return appropriately sized array
            if _isscalar(args[0]):
                return self.value
            else:    
                shape = args[0].shape
                return self.value*np.ones(shape)
        else: # Otherwise, return a single instance of the constant value
            return self.value
        
    def derivative(self,*args):
        '''
        Evaluate the derivative of the function.  The first input must exist and should be an array.
        Returns an array of identical shape to args[0] (if it exists).  This is an array of zeros.
        '''
        if len(args) > 0:
            if _isscalar(args[0]):
                return 0.0
            else:
                shape = args[0].shape
                return np.zeros(shape)
        else:
            return 0.0
        
    # All other derivatives are also zero everywhere, so these methods just point to derivative    
    derivativeX = derivative
    derivativeY = derivative
    derivativeZ = derivative
    derivativeW = derivative
    derivativeXX= derivative


class CubicInterp(HARKinterpolator1D):
    '''
    An interpolating function using piecewise cubic splines.  Matches level and
    slope of 1D function at gridpoints, smoothly interpolating in between.
    Extrapolation above highest gridpoint approaches a limiting linear function
    if desired (linear extrapolation also enabled.)
    '''
    distance_criteria = ['x_list','y_list','dydx_list']
    
    def __init__(self,x_list,y_list,dydx_list,intercept_limit=None,slope_limit=None,lower_extrap=False):
        '''
        The interpolation constructor to make a new cubic spline interpolation.
        
        Parameters
        ----------
        x_list : np.array
            List of x values composing the grid.
        y_list : np.array
            List of y values, representing f(x) at the points in x_list.
        dydx_list : np.array
            List of dydx values, representing f'(x) at the points in x_list
        intercept_limit : float
            Intercept of limiting linear function.
        slope_limit : float
            Slope of limiting linear function.
        lower_extrap : boolean
            Indicator for whether lower extrapolation is allowed.  False means
            f(x) = NaN for x < min(x_list); True means linear extrapolation.
            
        Returns
        -------
        new instance of CubicInterp
            
        NOTE: When no input is given for the limiting linear function, linear
        extrapolation is used above the highest gridpoint.        
        '''
        self.x_list = np.asarray(x_list)
        self.y_list = np.asarray(y_list)
        self.dydx_list = np.asarray(dydx_list)
        self.n = len(x_list)
        
        # Define lower extrapolation as linear function (or just NaN)
        if lower_extrap:
            self.coeffs = [[y_list[0],dydx_list[0],0,0]]
        else:
            self.coeffs = [[np.nan,np.nan,np.nan,np.nan]]

        # Calculate interpolation coefficients on segments mapped to [0,1]
        for i in xrange(self.n-1):
           x0 = x_list[i]
           y0 = y_list[i]
           x1 = x_list[i+1]
           y1 = y_list[i+1]
           Span = x1 - x0
           dydx0 = dydx_list[i]*Span
           dydx1 = dydx_list[i+1]*Span
           
           temp = [y0, dydx0, 3*(y1 - y0) - 2*dydx0 - dydx1, 2*(y0 - y1) + dydx0 + dydx1];
           self.coeffs.append(temp)

        # Calculate extrapolation coefficients as a decay toward limiting function y = mx+b
        if slope_limit is None and intercept_limit is None:
            slope_limit = dydx_list[-1]
            intercept_limit = y_list[-1] - slope_limit*x_list[-1]
        gap = slope_limit*x1 + intercept_limit - y1
        slope = slope_limit - dydx_list[self.n-1]
        if (gap != 0) and (slope <= 0):
            temp = [intercept_limit, slope_limit, gap, slope/gap]
        elif slope > 0:
            temp = [intercept_limit, slope_limit, 0, 0] # fixing a problem when slope is positive
        else:
            temp = [intercept_limit, slope_limit, gap, 0]
        self.coeffs.append(temp)
        self.coeffs = np.array(self.coeffs)

    def _evaluate(self,x):
        '''
        Returns the level of the interpolated function at each value in x.  Only
        called internally by HARKinterpolator1D.__call__ (etc).
        '''
        if _isscalar(x):
            pos = np.searchsorted(self.x_list,x)
            if pos == 0:
                y = self.coeffs[0,0] + self.coeffs[0,1]*(x - self.x_list[0])
            elif (pos < self.n):
                alpha = (x - self.x_list[pos-1])/(self.x_list[pos] - self.x_list[pos-1])
                y = self.coeffs[pos,0] + alpha*(self.coeffs[pos,1] + alpha*(self.coeffs[pos,2] + alpha*self.coeffs[pos,3]))
            else:
                alpha = x - self.x_list[self.n-1]
                y = self.coeffs[pos,0] + x*self.coeffs[pos,1] - self.coeffs[pos,2]*np.exp(alpha*self.coeffs[pos,3])
        else:
            m = len(x)
            pos = np.searchsorted(self.x_list,x)
            y = np.zeros(m)
            if y.size > 0:
                out_bot   = pos == 0
                out_top   = pos == self.n
                in_bnds   = np.logical_not(np.logical_or(out_bot, out_top))
                
                # Do the "in bounds" evaluation points
                i = pos[in_bnds]
                coeffs_in = self.coeffs[i,:]
                alpha = (x[in_bnds] - self.x_list[i-1])/(self.x_list[i] - self.x_list[i-1])
                y[in_bnds] = coeffs_in[:,0] + alpha*(coeffs_in[:,1] + alpha*(coeffs_in[:,2] + alpha*coeffs_in[:,3]))
                
                # Do the "out of bounds" evaluation points
                y[out_bot] = self.coeffs[0,0] + self.coeffs[0,1]*(x[out_bot] - self.x_list[0])
                alpha = x[out_top] - self.x_list[self.n-1]
                y[out_top] = self.coeffs[self.n,0] + x[out_top]*self.coeffs[self.n,1] - self.coeffs[self.n,2]*np.exp(alpha*self.coeffs[self.n,3])                      
        return y

    def _der(self,x):
        '''
        Returns the first derivative of the interpolated function at each value
        in x. Only called internally by HARKinterpolator1D.derivative (etc).
        '''
        if _isscalar(x):
            pos = np.searchsorted(self.x_list,x)
            if pos == 0:
                dydx = self.coeffs[0,1]
            elif (pos < self.n):
                alpha = (x - self.x_list[pos-1])/(self.x_list[pos] - self.x_list[pos-1])
                dydx = (self.coeffs[pos,1] + alpha*(2*self.coeffs[pos,2] + alpha*3*self.coeffs[pos,3]))/(self.x_list[pos] - self.x_list[pos-1])
            else:
                alpha = x - self.x_list[self.n-1]
                dydx = self.coeffs[pos,1] - self.coeffs[pos,2]*self.coeffs[pos,3]*np.exp(alpha*self.coeffs[pos,3])
        else:
            m = len(x)
            pos = np.searchsorted(self.x_list,x)
            dydx = np.zeros(m)
            if dydx.size > 0:
                out_bot   = pos == 0
                out_top   = pos == self.n
                in_bnds   = np.logical_not(np.logical_or(out_bot, out_top))
                
                # Do the "in bounds" evaluation points
                i = pos[in_bnds]
                coeffs_in = self.coeffs[i,:]
                alpha = (x[in_bnds] - self.x_list[i-1])/(self.x_list[i] - self.x_list[i-1])
                dydx[in_bnds] = (coeffs_in[:,1] + alpha*(2*coeffs_in[:,2] + alpha*3*coeffs_in[:,3]))/(self.x_list[i] - self.x_list[i-1])
                
                # Do the "out of bounds" evaluation points
                dydx[out_bot] = self.coeffs[0,1]
                alpha = x[out_top] - self.x_list[self.n-1]
                dydx[out_top] = self.coeffs[self.n,1] - self.coeffs[self.n,2]*self.coeffs[self.n,3]*np.exp(alpha*self.coeffs[self.n,3])
        return dydx


    def _evalAndDer(self,x):
        '''
        Returns the level and first derivative of the function at each value in
        x.  Only called internally by HARKinterpolator1D.eval_and_der (etc).
        '''
        if _isscalar(x):
            pos = np.searchsorted(self.x_list,x)
            if pos == 0:
                y = self.coeffs[0,0] + self.coeffs[0,1]*(x - self.x_list[0])
                dydx = self.coeffs[0,1]
            elif (pos < self.n):
                alpha = (x - self.x_list[pos-1])/(self.x_list[pos] - self.x_list[pos-1])
                y = self.coeffs[pos,0] + alpha*(self.coeffs[pos,1] + alpha*(self.coeffs[pos,2] + alpha*self.coeffs[pos,3]))
                dydx = (self.coeffs[pos,1] + alpha*(2*self.coeffs[pos,2] + alpha*3*self.coeffs[pos,3]))/(self.x_list[pos] - self.x_list[pos-1])
            else:
                alpha = x - self.x_list[self.n-1]
                y = self.coeffs[pos,0] + x*self.coeffs[pos,1] - self.coeffs[pos,2]*np.exp(alpha*self.coeffs[pos,3])
                dydx = self.coeffs[pos,1] - self.coeffs[pos,2]*self.coeffs[pos,3]*np.exp(alpha*self.coeffs[pos,3])
        else:
            m = len(x)
            pos = np.searchsorted(self.x_list,x)
            y = np.zeros(m)
            dydx = np.zeros(m)
            if y.size > 0:
                out_bot   = pos == 0
                out_top   = pos == self.n
                in_bnds   = np.logical_not(np.logical_or(out_bot, out_top))
                
                # Do the "in bounds" evaluation points
                i = pos[in_bnds]
                coeffs_in = self.coeffs[i,:]
                alpha = (x[in_bnds] - self.x_list[i-1])/(self.x_list[i] - self.x_list[i-1])
                y[in_bnds] = coeffs_in[:,0] + alpha*(coeffs_in[:,1] + alpha*(coeffs_in[:,2] + alpha*coeffs_in[:,3]))
                dydx[in_bnds] = (coeffs_in[:,1] + alpha*(2*coeffs_in[:,2] + alpha*3*coeffs_in[:,3]))/(self.x_list[i] - self.x_list[i-1])
                
                # Do the "out of bounds" evaluation points
                y[out_bot] = self.coeffs[0,0] + self.coeffs[0,1]*(x[out_bot] - self.x_list[0])
                dydx[out_bot] = self.coeffs[0,1]
                alpha = x[out_top] - self.x_list[self.n-1]
                y[out_top] = self.coeffs[self.n,0] + x[out_top]*self.coeffs[self.n,1] - self.coeffs[self.n,2]*np.exp(alpha*self.coeffs[self.n,3])
                dydx[out_top] = self.coeffs[self.n,1] - self.coeffs[self.n,2]*self.coeffs[self.n,3]*np.exp(alpha*self.coeffs[self.n,3])
        return y, dydx


class LinearInterp(HARKinterpolator1D):
    '''
    A slight extension of scipy.interpolate's UnivariateSpline for linear inter-
    polation.  Allows for linear or decay extrapolation (approaching a limiting
    linear function from below).
    
    '''
    def __init__(self,x_list,y_list,intercept_limit=None,slope_limit=None,lower_extrap=False):
        '''
        The interpolation constructor to make a new linear spline interpolation.
        
        Parameters
        ----------
        x_list : np.array
            List of x values composing the grid.
        y_list : np.array
            List of y values, representing f(x) at the points in x_list.
        intercept_limit : float
            Intercept of limiting linear function.
        slope_limit : float
            Slope of limiting linear function.
        lower_extrap : boolean
            Indicator for whether lower extrapolation is allowed.  False means
            f(x) = NaN for x < min(x_list); True means linear extrapolation.
            
        Returns
        -------
        new instance of LinearInterp
            
        NOTE: When no input is given for the limiting linear function, linear
        extrapolation is used above the highest gridpoint.        
        '''
        # Make the basic linear spline interpolation
        self.x_list = x_list
        self.y_list = y_list
        self.function = UnivariateSpline(x_list,y_list,k=1,s=0)
        self.lower_extrap = lower_extrap
        self.distance_criteria = ['x_list','y_list']
        
        # Make a decay extrapolation
        if intercept_limit is not None and slope_limit is not None:
            slope_at_top = self.function(x_list[-1],1)
            level_diff = intercept_limit + slope_limit*x_list[-1] - y_list[-1]
            slope_diff = slope_limit - slope_at_top
            self.decay_extrap_A = level_diff
            self.decay_extrap_B = -slope_diff/level_diff
            self.intercept_limit = intercept_limit
            self.slope_limit = slope_limit
            self.decay_extrap = True
        else:
            self.decay_extrap = False
        
    def _evaluate(self,x):
        '''
        Returns the level of the interpolated function at each value in x.  Only
        called internally by HARKinterpolator1D.__call__ (etc).
        '''
        out = self.function(x)
        if not self.lower_extrap:
            below_lower_bound = x < self.function._data[0][0]
            out[below_lower_bound] = np.nan
        if self.decay_extrap:
            above_upper_bound = x > self.function._data[0][-1]
            x_temp = x[above_upper_bound] - self.function._data[0][-1]
            out[above_upper_bound] = self.intercept_limit + self.slope_limit*x[above_upper_bound] - self.decay_extrap_A*np.exp(-self.decay_extrap_B*x_temp)
        return out
        
    def _der(self,x):
        '''
        Returns the first derivative of the interpolated function at each value
        in x. Only called internally by HARKinterpolator1D.derivative (etc).
        '''
        out = self.function(x,1)
        if not self.lower_extrap:
            below_lower_bound = x < self.function._data[0][0]
            out[below_lower_bound] = np.nan
        if self.decay_extrap:
            above_upper_bound = x > self.function._data[0][-1]
            x_temp = x[above_upper_bound] - self.function._data[0][-1]
            out[above_upper_bound] = self.slope_limit + self.decay_extrap_B*self.decay_extrap_A*np.exp(-self.decay_extrap_B*x_temp)
        return out
        
    def _evalAndDer(self,x):
        '''
        Returns the level and first derivative of the function at each value in
        x.  Only called internally by HARKinterpolator1D.eval_and_der (etc).
        '''
        out1 = self.function(x)
        out2 = self.function(x,1)
        if not self.lower_extrap:
            below_lower_bound = x < self.function._data[0][0]
            out1[below_lower_bound] = np.nan
            out2[below_lower_bound] = np.nan
        if self.decay_extrap:
            above_upper_bound = x > self.function._data[0][-1]
            x_temp = x[above_upper_bound] - self.function._data[0][-1]
            out1[above_upper_bound] = self.intercept_limit + self.slope_limit*x[above_upper_bound] - self.decay_extrap_A*np.exp(-self.decay_extrap_B*x_temp)
            out2[above_upper_bound] = self.slope_limit + self.decay_extrap_B*self.decay_extrap_A*np.exp(-self.decay_extrap_B*x_temp)
        return out1, out2
        
        
class BilinearInterp(HARKinterpolator2D):
    '''
    Bilinear full (or tensor) grid interpolation of a function f(x,y).
    '''
    def __init__(self,f_values,x_list,y_list,xSearchFunc=None,ySearchFunc=None):
        '''
        Constructor to make a new bilinear interpolation.
        
        Parameters
        ----------
        f_values : numpy.array
            An array of size (x_n,y_n) such that f_values[i,j] = f(x_list[i],y_list[j])
        x_list : numpy.array
            An array of x values, with length designated x_n.
        y_list : numpy.array
            An array of y values, with length designated y_n.
        xSearchFunc : function
            An optional function that returns the reference location for x values:
            indices = xSearchFunc(x_list,x).  Default is np.searchsorted
        ySearchFunc : function
            An optional function that returns the reference location for y values:
            indices = ySearchFunc(y_list,y).  Default is np.searchsorted
            
        Returns
        -------
        new instance of BilinearInterp
        '''
        self.f_values = f_values
        self.x_list = x_list
        self.y_list = y_list
        self.x_n = x_list.size
        self.y_n = y_list.size
        if xSearchFunc is None:
            xSearchFunc = np.searchsorted
        if ySearchFunc is None:
            ySearchFunc = np.searchsorted
        self.xSearchFunc = xSearchFunc
        self.ySearchFunc = ySearchFunc
        self.distance_criteria = ['x_list','y_list','f_values']
        
    def _evaluate(self,x,y):
        '''
        Returns the level of the interpolated function at each value in x,y.
        Only called internally by HARKinterpolator2D.__call__ (etc).
        '''
        if _isscalar(x):
            x_pos = max(min(self.xSearchFunc(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(self.ySearchFunc(self.y_list,y),self.y_n-1),1)
        else:
            x_pos = self.xSearchFunc(self.x_list,x)
            x_pos[x_pos < 1] = 1
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = self.ySearchFunc(self.y_list,y)
            y_pos[y_pos < 1] = 1
            y_pos[y_pos > self.y_n-1] = self.y_n-1
        alpha = (x - self.x_list[x_pos-1])/(self.x_list[x_pos] - self.x_list[x_pos-1])
        beta = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
        f = (
             (1-alpha)*(1-beta)*self.f_values[x_pos-1,y_pos-1]
          +  (1-alpha)*beta*self.f_values[x_pos-1,y_pos]
          +  alpha*(1-beta)*self.f_values[x_pos,y_pos-1]
          +  alpha*beta*self.f_values[x_pos,y_pos])
        return f
        
    def _derX(self,x,y):
        '''
        Returns the derivative with respect to x of the interpolated function
        at each value in x,y. Only called internally by HARKinterpolator2D.derivativeX.
        '''
        if _isscalar(x):
            x_pos = max(min(self.xSearchFunc(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(self.ySearchFunc(self.y_list,y),self.y_n-1),1)
        else:
            x_pos = self.xSearchFunc(self.x_list,x)
            x_pos[x_pos < 1] = 1
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = self.ySearchFunc(self.y_list,y)
            y_pos[y_pos < 1] = 1
            y_pos[y_pos > self.y_n-1] = self.y_n-1
        beta = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
        dfdx = (
              ((1-beta)*self.f_values[x_pos,y_pos-1]
            +  beta*self.f_values[x_pos,y_pos]) -
              ((1-beta)*self.f_values[x_pos-1,y_pos-1]
            +  beta*self.f_values[x_pos-1,y_pos]))/(self.x_list[x_pos] - self.x_list[x_pos-1])
        return dfdx
        
    def _derY(self,x,y):
        '''
        Returns the derivative with respect to y of the interpolated function
        at each value in x,y. Only called internally by HARKinterpolator2D.derivativeY.
        '''
        if _isscalar(x):
            x_pos = max(min(self.xSearchFunc(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(self.ySearchFunc(self.y_list,y),self.y_n-1),1)
        else:
            x_pos = self.xSearchFunc(self.x_list,x)
            x_pos[x_pos < 1] = 1
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = self.ySearchFunc(self.y_list,y)
            y_pos[y_pos < 1] = 1
            y_pos[y_pos > self.y_n-1] = self.y_n-1
        alpha = (x - self.x_list[x_pos-1])/(self.x_list[x_pos] - self.x_list[x_pos-1])
        dfdy = (
              ((1-alpha)*self.f_values[x_pos-1,y_pos]
            +  alpha*self.f_values[x_pos,y_pos]) -
              ((1-alpha)*self.f_values[x_pos-1,y_pos]
            +  alpha*self.f_values[x_pos-1,y_pos-1]))/(self.y_list[y_pos] - self.y_list[y_pos-1])
        return dfdy


class TrilinearInterp(HARKinterpolator3D):
    '''
    Trilinear full (or tensor) grid interpolation of a function f(x,y,z).
    '''
    def __init__(self,f_values,x_list,y_list,z_list,xSearchFunc=None,ySearchFunc=None,zSearchFunc=None):
        '''
        Constructor to make a new trilinear interpolation.
        
        Parameters
        ----------
        f_values : numpy.array
            An array of size (x_n,y_n,z_n) such that f_values[i,j,k] =
            f(x_list[i],y_list[j],z_list[k])
        x_list : numpy.array
            An array of x values, with length designated x_n.
        y_list : numpy.array
            An array of y values, with length designated y_n.
        z_list : numpy.array
            An array of z values, with length designated z_n.
        xSearchFunc : function
            An optional function that returns the reference location for x values:
            indices = xSearchFunc(x_list,x).  Default is np.searchsorted
        ySearchFunc : function
            An optional function that returns the reference location for y values:
            indices = ySearchFunc(y_list,y).  Default is np.searchsorted
        zSearchFunc : function
            An optional function that returns the reference location for z values:
            indices = zSearchFunc(z_list,z).  Default is np.searchsorted
            
        Returns
        -------
        new instance of TrilinearInterp
        '''
        self.f_values = f_values
        self.x_list = x_list
        self.y_list = y_list
        self.z_list = z_list
        self.x_n = x_list.size
        self.y_n = y_list.size
        self.z_n = z_list.size
        if xSearchFunc is None:
            xSearchFunc = np.searchsorted
        if ySearchFunc is None:
            ySearchFunc = np.searchsorted
        if zSearchFunc is None:
            zSearchFunc = np.searchsorted
        self.xSearchFunc = xSearchFunc
        self.ySearchFunc = ySearchFunc
        self.zSearchFunc = zSearchFunc
        self.distance_criteria = ['f_values','x_list','y_list','z_list']
        
    def _evaluate(self,x,y,z):
        '''
        Returns the level of the interpolated function at each value in x,y,z.
        Only called internally by HARKinterpolator3D.__call__ (etc).
        '''
        if _isscalar(x):
            x_pos = max(min(self.xSearchFunc(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(self.ySearchFunc(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(self.zSearchFunc(self.z_list,z),self.z_n-1),1)
        else:
            x_pos = self.xSearchFunc(self.x_list,x)
            x_pos[x_pos < 1] = 1
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = self.ySearchFunc(self.y_list,y)
            y_pos[y_pos < 1] = 1
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            z_pos = self.zSearchFunc(self.z_list,z)
            z_pos[z_pos < 1] = 1
            z_pos[z_pos > self.z_n-1] = self.z_n-1
        alpha = (x - self.x_list[x_pos-1])/(self.x_list[x_pos] - self.x_list[x_pos-1])
        beta = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
        gamma = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
        f = (
             (1-alpha)*(1-beta)*(1-gamma)*self.f_values[x_pos-1,y_pos-1,z_pos-1]
          +  (1-alpha)*(1-beta)*gamma*self.f_values[x_pos-1,y_pos-1,z_pos]
          +  (1-alpha)*beta*(1-gamma)*self.f_values[x_pos-1,y_pos,z_pos-1]
          +  (1-alpha)*beta*gamma*self.f_values[x_pos-1,y_pos,z_pos]
          +  alpha*(1-beta)*(1-gamma)*self.f_values[x_pos,y_pos-1,z_pos-1]
          +  alpha*(1-beta)*gamma*self.f_values[x_pos,y_pos-1,z_pos]
          +  alpha*beta*(1-gamma)*self.f_values[x_pos,y_pos,z_pos-1]
          +  alpha*beta*gamma*self.f_values[x_pos,y_pos,z_pos])
        return f
        
    def _derX(self,x,y,z):
        '''
        Returns the derivative with respect to x of the interpolated function
        at each value in x,y,z. Only called internally by HARKinterpolator3D.derivativeX.
        '''
        if _isscalar(x):
            x_pos = max(min(self.xSearchFunc(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(self.ySearchFunc(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(self.zSearchFunc(self.z_list,z),self.z_n-1),1)
        else:
            x_pos = self.xSearchFunc(self.x_list,x)
            x_pos[x_pos < 1] = 1
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = self.ySearchFunc(self.y_list,y)
            y_pos[y_pos < 1] = 1
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            z_pos = self.zSearchFunc(self.z_list,z)
            z_pos[z_pos < 1] = 1
            z_pos[z_pos > self.z_n-1] = self.z_n-1
        beta = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
        gamma = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
        dfdx = (
           (  (1-beta)*(1-gamma)*self.f_values[x_pos,y_pos-1,z_pos-1]
           +  (1-beta)*gamma*self.f_values[x_pos,y_pos-1,z_pos]
           +  beta*(1-gamma)*self.f_values[x_pos,y_pos,z_pos-1]
           +  beta*gamma*self.f_values[x_pos,y_pos,z_pos]) -
           (  (1-beta)*(1-gamma)*self.f_values[x_pos-1,y_pos-1,z_pos-1]
           +  (1-beta)*gamma*self.f_values[x_pos-1,y_pos-1,z_pos]
           +  beta*(1-gamma)*self.f_values[x_pos-1,y_pos,z_pos-1]
           +  beta*gamma*self.f_values[x_pos-1,y_pos,z_pos]))/(self.x_list[x_pos] - self.x_list[x_pos-1])
        return dfdx
        
    def _derY(self,x,y,z):
        '''
        Returns the derivative with respect to y of the interpolated function
        at each value in x,y,z. Only called internally by HARKinterpolator3D.derivativeY.
        '''
        if _isscalar(x):
            x_pos = max(min(self.xSearchFunc(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(self.ySearchFunc(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(self.zSearchFunc(self.z_list,z),self.z_n-1),1)
        else:
            x_pos = self.xSearchFunc(self.x_list,x)
            x_pos[x_pos < 1] = 1
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = self.ySearchFunc(self.y_list,y)
            y_pos[y_pos < 1] = 1
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            z_pos = self.zSearchFunc(self.z_list,z)
            z_pos[z_pos < 1] = 1
            z_pos[z_pos > self.z_n-1] = self.z_n-1
        alpha = (x - self.x_list[x_pos-1])/(self.x_list[x_pos] - self.x_list[x_pos-1])
        gamma = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
        dfdy = (
           (  (1-alpha)*(1-gamma)*self.f_values[x_pos-1,y_pos,z_pos-1]
           +  (1-alpha)*gamma*self.f_values[x_pos-1,y_pos,z_pos]
           +  alpha*(1-gamma)*self.f_values[x_pos,y_pos,z_pos-1]
           +  alpha*gamma*self.f_values[x_pos,y_pos,z_pos]) -
           (  (1-alpha)*(1-gamma)*self.f_values[x_pos-1,y_pos-1,z_pos-1]
           +  (1-alpha)*gamma*self.f_values[x_pos-1,y_pos-1,z_pos]
           +  alpha*(1-gamma)*self.f_values[x_pos,y_pos-1,z_pos-1]
           +  alpha*gamma*self.f_values[x_pos,y_pos-1,z_pos]))/(self.y_list[y_pos] - self.y_list[y_pos-1])
        return dfdy
        
    def _derZ(self,x,y,z):
        '''
        Returns the derivative with respect to z of the interpolated function
        at each value in x,y,z. Only called internally by HARKinterpolator3D.derivativeZ.
        '''
        if _isscalar(x):
            x_pos = max(min(self.xSearchFunc(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(self.ySearchFunc(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(self.zSearchFunc(self.z_list,z),self.z_n-1),1)
        else:
            x_pos = self.xSearchFunc(self.x_list,x)
            x_pos[x_pos < 1] = 1
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = self.ySearchFunc(self.y_list,y)
            y_pos[y_pos < 1] = 1
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            z_pos = self.zSearchFunc(self.z_list,z)
            z_pos[z_pos < 1] = 1
            z_pos[z_pos > self.z_n-1] = self.z_n-1
        alpha = (x - self.x_list[x_pos-1])/(self.x_list[x_pos] - self.x_list[x_pos-1])
        beta = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
        dfdz = (
           (  (1-alpha)*(1-beta)*self.f_values[x_pos-1,y_pos-1,z_pos]
           +  (1-alpha)*beta*self.f_values[x_pos-1,y_pos,z_pos]
           +  alpha*(1-beta)*self.f_values[x_pos,y_pos-1,z_pos]
           +  alpha*beta*self.f_values[x_pos,y_pos,z_pos]) -
           (  (1-alpha)*(1-beta)*self.f_values[x_pos-1,y_pos-1,z_pos-1]
           +  (1-alpha)*beta*self.f_values[x_pos-1,y_pos,z_pos-1]
           +  alpha*(1-beta)*self.f_values[x_pos,y_pos-1,z_pos-1]
           +  alpha*beta*self.f_values[x_pos,y_pos,z_pos-1]))/(self.z_list[z_pos] - self.z_list[z_pos-1])
        return dfdz
        

class QuadlinearInterp(HARKinterpolator4D):
    '''
    Quadlinear full (or tensor) grid interpolation of a function f(w,x,y,z).
    '''
    def __init__(self,f_values,w_list,x_list,y_list,z_list,wSearchFunc=None,xSearchFunc=None,ySearchFunc=None,zSearchFunc=None):
        '''
        Constructor to make a new quadlinear interpolation.
        
        Parameters
        ----------
        f_values : numpy.array
            An array of size (w_n,x_n,y_n,z_n) such that f_values[i,j,k,l] =
            f(w_list[i],x_list[j],y_list[k],z_list[l])
        w_list : numpy.array
            An array of x values, with length designated w_n.
        x_list : numpy.array
            An array of x values, with length designated x_n.
        y_list : numpy.array
            An array of y values, with length designated y_n.
        z_list : numpy.array
            An array of z values, with length designated z_n.
        wSearchFunc : function
            An optional function that returns the reference location for w values:
            indices = wSearchFunc(w_list,w).  Default is np.searchsorted
        xSearchFunc : function
            An optional function that returns the reference location for x values:
            indices = xSearchFunc(x_list,x).  Default is np.searchsorted
        ySearchFunc : function
            An optional function that returns the reference location for y values:
            indices = ySearchFunc(y_list,y).  Default is np.searchsorted
        zSearchFunc : function
            An optional function that returns the reference location for z values:
            indices = zSearchFunc(z_list,z).  Default is np.searchsorted
            
        Returns
        -------
        new instance of QuadlinearInterp
        '''
        self.f_values = f_values
        self.w_list = w_list
        self.x_list = x_list
        self.y_list = y_list
        self.z_list = z_list
        self.w_n = w_list.size
        self.x_n = x_list.size
        self.y_n = y_list.size
        self.z_n = z_list.size
        if wSearchFunc is None:
            wSearchFunc = np.searchsorted
        if xSearchFunc is None:
            xSearchFunc = np.searchsorted
        if ySearchFunc is None:
            ySearchFunc = np.searchsorted
        if zSearchFunc is None:
            zSearchFunc = np.searchsorted
        self.wSearchFunc = wSearchFunc
        self.xSearchFunc = xSearchFunc
        self.ySearchFunc = ySearchFunc
        self.zSearchFunc = zSearchFunc
        self.distance_criteria = ['f_values','w_list','x_list','y_list','z_list']
        
    def _evaluate(self,w,x,y,z):
        '''
        Returns the level of the interpolated function at each value in x,y,z.
        Only called internally by HARKinterpolator4D.__call__ (etc).
        '''
        if _isscalar(w):
            w_pos = max(min(self.wSearchFunc(self.w_list,w),self.w_n-1),1)
            x_pos = max(min(self.xSearchFunc(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(self.ySearchFunc(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(self.zSearchFunc(self.z_list,z),self.z_n-1),1)
        else:
            w_pos = self.wSearchFunc(self.w_list,w)
            w_pos[w_pos < 1] = 1
            w_pos[w_pos > self.w_n-1] = self.w_n-1
            x_pos = self.xSearchFunc(self.x_list,x)
            x_pos[x_pos < 1] = 1
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = self.ySearchFunc(self.y_list,y)
            y_pos[y_pos < 1] = 1
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            z_pos = self.zSearchFunc(self.z_list,z)
            z_pos[z_pos < 1] = 1
            z_pos[z_pos > self.z_n-1] = self.z_n-1
        i = w_pos # for convenience
        j = x_pos
        k = y_pos
        l = z_pos
        alpha = (w - self.w_list[i-1])/(self.w_list[i] - self.w_list[i-1])
        beta = (x - self.x_list[j-1])/(self.x_list[j] - self.x_list[j-1])
        gamma = (y - self.y_list[k-1])/(self.y_list[k] - self.y_list[k-1])
        delta = (z - self.z_list[l-1])/(self.z_list[l] - self.z_list[l-1])
        f = (
             (1-alpha)*((1-beta)*((1-gamma)*(1-delta)*self.f_values[i-1,j-1,k-1,l-1]
           + (1-gamma)*delta*self.f_values[i-1,j-1,k-1,l]
           + gamma*(1-delta)*self.f_values[i-1,j-1,k,l-1]
           + gamma*delta*self.f_values[i-1,j-1,k,l])
           + beta*((1-gamma)*(1-delta)*self.f_values[i-1,j,k-1,l-1]
           + (1-gamma)*delta*self.f_values[i-1,j,k-1,l]
           + gamma*(1-delta)*self.f_values[i-1,j,k,l-1]
           + gamma*delta*self.f_values[i-1,j,k,l]))
           + alpha*((1-beta)*((1-gamma)*(1-delta)*self.f_values[i,j-1,k-1,l-1]
           + (1-gamma)*delta*self.f_values[i,j-1,k-1,l]
           + gamma*(1-delta)*self.f_values[i,j-1,k,l-1]
           + gamma*delta*self.f_values[i,j-1,k,l])
           + beta*((1-gamma)*(1-delta)*self.f_values[i,j,k-1,l-1]
           + (1-gamma)*delta*self.f_values[i,j,k-1,l]
           + gamma*(1-delta)*self.f_values[i,j,k,l-1]
           + gamma*delta*self.f_values[i,j,k,l])))       
        return f
        
    def _derW(self,w,x,y,z):
        '''
        Returns the derivative with respect to w of the interpolated function
        at each value in w,x,y,z. Only called internally by HARKinterpolator4D.derivativeW.
        '''
        if _isscalar(w):
            w_pos = max(min(self.wSearchFunc(self.w_list,w),self.w_n-1),1)
            x_pos = max(min(self.xSearchFunc(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(self.ySearchFunc(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(self.zSearchFunc(self.z_list,z),self.z_n-1),1)
        else:
            w_pos = self.wSearchFunc(self.w_list,w)
            w_pos[w_pos < 1] = 1
            w_pos[w_pos > self.w_n-1] = self.w_n-1
            x_pos = self.xSearchFunc(self.x_list,x)
            x_pos[x_pos < 1] = 1
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = self.ySearchFunc(self.y_list,y)
            y_pos[y_pos < 1] = 1
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            z_pos = self.zSearchFunc(self.z_list,z)
            z_pos[z_pos < 1] = 1
            z_pos[z_pos > self.z_n-1] = self.z_n-1
        i = w_pos # for convenience
        j = x_pos
        k = y_pos
        l = z_pos
        beta = (x - self.x_list[j-1])/(self.x_list[j] - self.x_list[j-1])
        gamma = (y - self.y_list[k-1])/(self.y_list[k] - self.y_list[k-1])
        delta = (z - self.z_list[l-1])/(self.z_list[l] - self.z_list[l-1])
        dfdw = (
          (  (1-beta)*(1-gamma)*(1-delta)*self.f_values[i,j-1,k-1,l-1]
           + (1-beta)*(1-gamma)*delta*self.f_values[i,j-1,k-1,l]
           + (1-beta)*gamma*(1-delta)*self.f_values[i,j-1,k,l-1]
           + (1-beta)*gamma*delta*self.f_values[i,j-1,k,l]
           + beta*(1-gamma)*(1-delta)*self.f_values[i,j,k-1,l-1]
           + beta*(1-gamma)*delta*self.f_values[i,j,k-1,l]
           + beta*gamma*(1-delta)*self.f_values[i,j,k,l-1]
           + beta*gamma*delta*self.f_values[i,j,k,l] ) - 
          (  (1-beta)*(1-gamma)*(1-delta)*self.f_values[i-1,j-1,k-1,l-1]
           + (1-beta)*(1-gamma)*delta*self.f_values[i-1,j-1,k-1,l]
           + (1-beta)*gamma*(1-delta)*self.f_values[i-1,j-1,k,l-1]
           + (1-beta)*gamma*delta*self.f_values[i-1,j-1,k,l]
           + beta*(1-gamma)*(1-delta)*self.f_values[i-1,j,k-1,l-1]
           + beta*(1-gamma)*delta*self.f_values[i-1,j,k-1,l]
           + beta*gamma*(1-delta)*self.f_values[i-1,j,k,l-1]
           + beta*gamma*delta*self.f_values[i-1,j,k,l] )
              )/(self.w_list[i] - self.w_list[i-1])
        return dfdw
        
    def _derX(self,w,x,y,z):
        '''
        Returns the derivative with respect to x of the interpolated function
        at each value in w,x,y,z. Only called internally by HARKinterpolator4D.derivativeX.
        '''
        if _isscalar(w):
            w_pos = max(min(self.wSearchFunc(self.w_list,w),self.w_n-1),1)
            x_pos = max(min(self.xSearchFunc(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(self.ySearchFunc(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(self.zSearchFunc(self.z_list,z),self.z_n-1),1)
        else:
            w_pos = self.wSearchFunc(self.w_list,w)
            w_pos[w_pos < 1] = 1
            w_pos[w_pos > self.w_n-1] = self.w_n-1
            x_pos = self.xSearchFunc(self.x_list,x)
            x_pos[x_pos < 1] = 1
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = self.ySearchFunc(self.y_list,y)
            y_pos[y_pos < 1] = 1
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            z_pos = self.zSearchFunc(self.z_list,z)
            z_pos[z_pos < 1] = 1
            z_pos[z_pos > self.z_n-1] = self.z_n-1
        i = w_pos # for convenience
        j = x_pos
        k = y_pos
        l = z_pos
        alpha = (w - self.w_list[i-1])/(self.w_list[i] - self.w_list[i-1])
        gamma = (y - self.y_list[k-1])/(self.y_list[k] - self.y_list[k-1])
        delta = (z - self.z_list[l-1])/(self.z_list[l] - self.z_list[l-1])
        dfdx = (
          (  (1-alpha)*(1-gamma)*(1-delta)*self.f_values[i-1,j,k-1,l-1]
           + (1-alpha)*(1-gamma)*delta*self.f_values[i-1,j,k-1,l]
           + (1-alpha)*gamma*(1-delta)*self.f_values[i-1,j,k,l-1]
           + (1-alpha)*gamma*delta*self.f_values[i-1,j,k,l]
           + alpha*(1-gamma)*(1-delta)*self.f_values[i,j,k-1,l-1]
           + alpha*(1-gamma)*delta*self.f_values[i,j,k-1,l]
           + alpha*gamma*(1-delta)*self.f_values[i,j,k,l-1]
           + alpha*gamma*delta*self.f_values[i,j,k,l] ) - 
          (  (1-alpha)*(1-gamma)*(1-delta)*self.f_values[i-1,j-1,k-1,l-1]
           + (1-alpha)*(1-gamma)*delta*self.f_values[i-1,j-1,k-1,l]
           + (1-alpha)*gamma*(1-delta)*self.f_values[i-1,j-1,k,l-1]
           + (1-alpha)*gamma*delta*self.f_values[i-1,j-1,k,l]
           + alpha*(1-gamma)*(1-delta)*self.f_values[i,j-1,k-1,l-1]
           + alpha*(1-gamma)*delta*self.f_values[i,j-1,k-1,l]
           + alpha*gamma*(1-delta)*self.f_values[i,j-1,k,l-1]
           + alpha*gamma*delta*self.f_values[i,j-1,k,l] )
              )/(self.x_list[j] - self.x_list[j-1])
        return dfdx
        
    def _derY(self,w,x,y,z):
        '''
        Returns the derivative with respect to y of the interpolated function
        at each value in w,x,y,z. Only called internally by HARKinterpolator4D.derivativeY.
        '''
        if _isscalar(w):
            w_pos = max(min(self.wSearchFunc(self.w_list,w),self.w_n-1),1)
            x_pos = max(min(self.xSearchFunc(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(self.ySearchFunc(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(self.zSearchFunc(self.z_list,z),self.z_n-1),1)
        else:
            w_pos = self.wSearchFunc(self.w_list,w)
            w_pos[w_pos < 1] = 1
            w_pos[w_pos > self.w_n-1] = self.w_n-1
            x_pos = self.xSearchFunc(self.x_list,x)
            x_pos[x_pos < 1] = 1
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = self.ySearchFunc(self.y_list,y)
            y_pos[y_pos < 1] = 1
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            z_pos = self.zSearchFunc(self.z_list,z)
            z_pos[z_pos < 1] = 1
            z_pos[z_pos > self.z_n-1] = self.z_n-1
        i = w_pos # for convenience
        j = x_pos
        k = y_pos
        l = z_pos
        alpha = (w - self.w_list[i-1])/(self.w_list[i] - self.w_list[i-1])
        beta = (x - self.x_list[j-1])/(self.x_list[j] - self.x_list[j-1])
        delta = (z - self.z_list[l-1])/(self.z_list[l] - self.z_list[l-1])
        dfdy = (
          (  (1-alpha)*(1-beta)*(1-delta)*self.f_values[i-1,j-1,k,l-1]
           + (1-alpha)*(1-beta)*delta*self.f_values[i-1,j-1,k,l]
           + (1-alpha)*beta*(1-delta)*self.f_values[i-1,j,k,l-1]
           + (1-alpha)*beta*delta*self.f_values[i-1,j,k,l]
           + alpha*(1-beta)*(1-delta)*self.f_values[i,j-1,k,l-1]
           + alpha*(1-beta)*delta*self.f_values[i,j-1,k,l]
           + alpha*beta*(1-delta)*self.f_values[i,j,k,l-1]
           + alpha*beta*delta*self.f_values[i,j,k,l] ) - 
          (  (1-alpha)*(1-beta)*(1-delta)*self.f_values[i-1,j-1,k-1,l-1]
           + (1-alpha)*(1-beta)*delta*self.f_values[i-1,j-1,k-1,l]
           + (1-alpha)*beta*(1-delta)*self.f_values[i-1,j,k-1,l-1]
           + (1-alpha)*beta*delta*self.f_values[i-1,j,k-1,l]
           + alpha*(1-beta)*(1-delta)*self.f_values[i,j-1,k-1,l-1]
           + alpha*(1-beta)*delta*self.f_values[i,j-1,k-1,l]
           + alpha*beta*(1-delta)*self.f_values[i,j,k-1,l-1]
           + alpha*beta*delta*self.f_values[i,j,k-1,l] )
              )/(self.y_list[k] - self.y_list[k-1])
        return dfdy
        
    def _derZ(self,w,x,y,z):
        '''
        Returns the derivative with respect to z of the interpolated function
        at each value in w,x,y,z. Only called internally by HARKinterpolator4D.derivativeZ.
        '''
        if _isscalar(w):
            w_pos = max(min(self.wSearchFunc(self.w_list,w),self.w_n-1),1)
            x_pos = max(min(self.xSearchFunc(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(self.ySearchFunc(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(self.zSearchFunc(self.z_list,z),self.z_n-1),1)
        else:
            w_pos = self.wSearchFunc(self.w_list,w)
            w_pos[w_pos < 1] = 1
            w_pos[w_pos > self.w_n-1] = self.w_n-1
            x_pos = self.xSearchFunc(self.x_list,x)
            x_pos[x_pos < 1] = 1
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = self.ySearchFunc(self.y_list,y)
            y_pos[y_pos < 1] = 1
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            z_pos = self.zSearchFunc(self.z_list,z)
            z_pos[z_pos < 1] = 1
            z_pos[z_pos > self.z_n-1] = self.z_n-1
        i = w_pos # for convenience
        j = x_pos
        k = y_pos
        l = z_pos
        alpha = (w - self.w_list[i-1])/(self.w_list[i] - self.w_list[i-1])
        beta = (x - self.x_list[j-1])/(self.x_list[j] - self.x_list[j-1])
        gamma = (y - self.y_list[k-1])/(self.y_list[k] - self.y_list[k-1])
        dfdz = (
          (  (1-alpha)*(1-beta)*(1-gamma)*self.f_values[i-1,j-1,k-1,l]
           + (1-alpha)*(1-beta)*gamma*self.f_values[i-1,j-1,k,l]
           + (1-alpha)*beta*(1-gamma)*self.f_values[i-1,j,k-1,l]
           + (1-alpha)*beta*gamma*self.f_values[i-1,j,k,l]
           + alpha*(1-beta)*(1-gamma)*self.f_values[i,j-1,k-1,l]
           + alpha*(1-beta)*gamma*self.f_values[i,j-1,k,l]
           + alpha*beta*(1-gamma)*self.f_values[i,j,k-1,l]
           + alpha*beta*gamma*self.f_values[i,j,k,l] ) - 
          (  (1-alpha)*(1-beta)*(1-gamma)*self.f_values[i-1,j-1,k-1,l-1]
           + (1-alpha)*(1-beta)*gamma*self.f_values[i-1,j-1,k,l-1]
           + (1-alpha)*beta*(1-gamma)*self.f_values[i-1,j,k-1,l-1]
           + (1-alpha)*beta*gamma*self.f_values[i-1,j,k,l-1]
           + alpha*(1-beta)*(1-gamma)*self.f_values[i,j-1,k-1,l-1]
           + alpha*(1-beta)*gamma*self.f_values[i,j-1,k,l-1]
           + alpha*beta*(1-gamma)*self.f_values[i,j,k-1,l-1]
           + alpha*beta*gamma*self.f_values[i,j,k,l-1] )
              )/(self.z_list[l] - self.z_list[l-1])
        return dfdz
        

class LowerEnvelope(HARKinterpolator1D):
    '''
    The lower envelope of a finite set of 1D functions, each of which can be of
    any class that has the methods __call__, derivative, and eval_with_derivative.
    Generally: it combines HARKinterpolator1Ds. 
    '''
    distance_criteria = ['functions']

    def __init__(self,*functions):
        '''
        Constructor to make a new lower envelope iterpolation.
        
        Parameters
        ----------
        *functions : function
            Any number of real functions; often instances of HARKinterpolator1D
            
        Returns
        -------
        new instance of LowerEnvelope
        '''
        self.functions = []
        for function in functions:
            self.functions.append(function)
        self.funcCount = len(self.functions)

    def _evaluate(self,x):
        '''
        Returns the level of the function at each value in x as the minimum among
        all of the functions.  Only called internally by HARKinterpolator1D.__call__.
        '''
        if _isscalar(x):
            y = np.nanmin([f(x) for f in self.functions])
        else:
            m = len(x)
            fx = np.zeros((m,self.funcCount))
            for j in range(self.funcCount):
                fx[:,j] = self.functions[j](x)
            y = np.nanmin(fx,axis=1)       
        return y

    def _der(self,x):
        '''
        Returns the first derivative of the function at each value in x.  Only
        called internally by HARKinterpolator1D.derivative.
        '''
        y,dydx = self.eval_with_derivative(x)
        return dydx  # Sadly, this is the fastest / most convenient way...

    def _evalAndDer(self,x):
        '''
        Returns the level and first derivative of the function at each value in
        x.  Only called internally by HARKinterpolator1D.eval_and_der.
        '''
        m = len(x)
        fx = np.zeros((m,self.funcCount))
        for j in range(self.funcCount):
            fx[:,j] = self.functions[j](x)
        fx[np.isnan(fx)] = np.inf
        i = np.argmin(fx,axis=1)
        y = fx[np.arange(m),i]
        dydx = np.zeros_like(y)
        for j in range(self.funcCount):
            c = i == j
            dydx[c] = self.functions[j].derivative(x[c])
        return y,dydx


class UpperEnvelope(HARKinterpolator1D):
    '''
    The upper envelope of a finite set of 1D functions, each of which can be of
    any class that has the methods __call__, derivative, and eval_with_derivative.
    Generally: it combines HARKinterpolator1Ds. 
    '''
    distance_criteria = ['functions']

    def __init__(self,*functions):
        '''
        Constructor to make a new upper envelope iterpolation.
        
        Parameters
        ----------
        *functions : function
            Any number of real functions; often instances of HARKinterpolator1D
            
        Returns
        -------
        new instance of UpperEnvelope
        '''
        self.functions = []
        for function in functions:
            self.functions.append(function)
        self.funcCount = len(self.functions)

    def _evaluate(self,x):
        '''
        Returns the level of the function at each value in x as the maximum among
        all of the functions.  Only called internally by HARKinterpolator1D.__call__.
        '''
        if _isscalar(x):
            y = np.nanmax([f(x) for f in self.functions])
        else:
            m = len(x)
            fx = np.zeros((m,self.funcCount))
            for j in range(self.funcCount):
                fx[:,j] = self.functions[j](x)
            y = np.nanmax(fx,axis=1)       
        return y

    def _der(self,x):
        '''
        Returns the first derivative of the function at each value in x.  Only
        called internally by HARKinterpolator1D.derivative.
        '''
        y,dydx = self.eval_with_derivative(x)
        return dydx  # Sadly, this is the fastest / most convenient way...

    def _evalAndDer(self,x):
        '''
        Returns the level and first derivative of the function at each value in
        x.  Only called internally by HARKinterpolator1D.eval_and_der.
        '''
        m = len(x)
        fx = np.zeros((m,self.funcCount))
        for j in range(self.funcCount):
            fx[:,j] = self.functions[j](x)
        fx[np.isnan(fx)] = np.inf
        i = np.argmax(fx,axis=1)
        y = fx[np.arange(m),i]
        dydx = np.zeros_like(y)
        for j in range(self.funcCount):
            c = i == j
            dydx[c] = self.functions[j].derivative(x[c])
        return y,dydx
        
        
class LowerEnvelope2D(HARKinterpolator2D):
    '''
    The lower envelope of a finite set of 2D functions, each of which can be of
    any class that has the methods __call__, derivativeX, and derivativeY.
    Generally: it combines HARKinterpolator2Ds. 
    ''' 

    def __init__(self,*functions):
        '''
        Constructor to make a new lower envelope iterpolation.
        
        Parameters
        ----------
        *functions : function
            Any number of real functions; often instances of HARKinterpolator2D
            
        Returns
        -------
        new instance of LowerEnvelope2D
        '''
        self.functions = []
        for function in functions:
            self.functions.append(function)
        self.funcCount = len(self.functions)
        self.distance_criteria = ['functions']

    def _evaluate(self,x,y):
        '''
        Returns the level of the function at each value in (x,y) as the minimum
        among all of the functions.  Only called internally by
        HARKinterpolator2D.__call__.
        '''
        if _isscalar(x):
            f = np.nanmin([f(x,y) for f in self.functions])
        else:
            m = len(x)
            temp = np.zeros((m,self.funcCount))
            for j in range(self.funcCount):
                temp[:,j] = self.functions[j](x,y)
            f = np.nanmin(temp,axis=1)       
        return f

    def _derX(self,x,y):
        '''
        Returns the first derivative of the function with respect to X at each
        value in (x,y).  Only called internally by HARKinterpolator2D._derX.
        '''
        m = len(x)
        temp = np.zeros((m,self.funcCount))
        for j in range(self.funcCount):
            temp[:,j] = self.functions[j](x,y)
        temp[np.isnan(temp)] = np.inf
        i = np.argmin(temp,axis=1)
        dfdx = np.zeros_like(x)
        for j in range(self.funcCount):
            c = i == j
            dfdx[c] = self.functions[j].derivativeX(x[c],y[c])
        return dfdx
        
    def _derY(self,x,y):
        '''
        Returns the first derivative of the function with respect to Y at each
        value in (x,y).  Only called internally by HARKinterpolator2D._derY.
        '''
        m = len(x)
        temp = np.zeros((m,self.funcCount))
        for j in range(self.funcCount):
            temp[:,j] = self.functions[j](x,y)
        temp[np.isnan(temp)] = np.inf
        i = np.argmin(temp,axis=1)
        y = temp[np.arange(m),i]
        dfdy = np.zeros_like(x)
        for j in range(self.funcCount):
            c = i == j
            dfdy[c] = self.functions[j].derivativeY(x[c],y[c])
        return dfdy
        
    
class LowerEnvelope3D(HARKinterpolator3D):
    '''
    The lower envelope of a finite set of 3D functions, each of which can be of
    any class that has the methods __call__, derivativeX, derivativeY, and
    derivativeZ. Generally: it combines HARKinterpolator2Ds. 
    '''
    distance_criteria = ['functions']

    def __init__(self,*functions):
        '''
        Constructor to make a new lower envelope iterpolation.
        
        Parameters
        ----------
        *functions : function
            Any number of real functions; often instances of HARKinterpolator3D
            
        Returns
        -------
        None
        '''
        self.functions = []
        for function in functions:
            self.functions.append(function)
        self.funcCount = len(self.functions)

    def _evaluate(self,x,y,z):
        '''
        Returns the level of the function at each value in (x,y,z) as the minimum
        among all of the functions.  Only called internally by
        HARKinterpolator3D.__call__.
        '''
        if _isscalar(x):
            f = np.nanmin([f(x,y,z) for f in self.functions])
        else:
            m = len(x)
            temp = np.zeros((m,self.funcCount))
            for j in range(self.funcCount):
                temp[:,j] = self.functions[j](x,y,z)
            f = np.nanmin(temp,axis=1)       
        return f

    def _derX(self,x,y,z):
        '''
        Returns the first derivative of the function with respect to X at each
        value in (x,y,z).  Only called internally by HARKinterpolator3D._derX.
        '''
        m = len(x)
        temp = np.zeros((m,self.funcCount))
        for j in range(self.funcCount):
            temp[:,j] = self.functions[j](x,y,z)
        temp[np.isnan(temp)] = np.inf
        i = np.argmin(temp,axis=1)
        dfdx = np.zeros_like(x)
        for j in range(self.funcCount):
            c = i == j
            dfdx[c] = self.functions[j].derivativeX(x[c],y[c],z[c])
        return dfdx
        
    def _derY(self,x,y,z):
        '''
        Returns the first derivative of the function with respect to Y at each
        value in (x,y,z).  Only called internally by HARKinterpolator3D._derY.
        '''
        m = len(x)
        temp = np.zeros((m,self.funcCount))
        for j in range(self.funcCount):
            temp[:,j] = self.functions[j](x,y,z)
        temp[np.isnan(temp)] = np.inf
        i = np.argmin(temp,axis=1)
        y = temp[np.arange(m),i]
        dfdy = np.zeros_like(x)
        for j in range(self.funcCount):
            c = i == j
            dfdy[c] = self.functions[j].derivativeY(x[c],y[c],z[c])
        return dfdy

    def _derZ(self,x,y,z):
        '''
        Returns the first derivative of the function with respect to Z at each
        value in (x,y,z).  Only called internally by HARKinterpolator3D._derZ.
        '''
        m = len(x)
        temp = np.zeros((m,self.funcCount))
        for j in range(self.funcCount):
            temp[:,j] = self.functions[j](x,y,z)
        temp[np.isnan(temp)] = np.inf
        i = np.argmin(temp,axis=1)
        y = temp[np.arange(m),i]
        dfdz = np.zeros_like(x)
        for j in range(self.funcCount):
            c = i == j
            dfdz[c] = self.functions[j].derivativeZ(x[c],y[c],z[c])
        return dfdz    

class LinearInterpOnInterp1D(HARKinterpolator2D):
    '''
    A 2D interpolator that linearly interpolates among a list of 1D interpolators.
    '''    
    def __init__(self,xInterpolators,y_values):
        '''
        Constructor for the class, generating an approximation to a function of
        the form f(x,y) using interpolations over f(x,y_0) for a fixed grid of
        y_0 values.
        
        Parameters
        ----------
        xInterpolators : [HARKinterpolator1D]
            A list of 1D interpolations over the x variable.  The nth element of
            xInterpolators represents f(x,y_values[n]).
        y_values: numpy.array
            An array of y values equal in length to xInterpolators.
            
        Returns
        -------
        new instance of LinearInterpOnInterp1D
        '''
        self.xInterpolators = xInterpolators
        self.y_list = y_values
        self.y_n = y_values.size
        self.distance_criteria = ['xInterpolators','y_list']
        
    def _evaluate(self,x,y):
        '''
        Returns the level of the interpolated function at each value in x,y.
        Only called internally by HARKinterpolator2D.__call__ (etc).
        '''
        if _isscalar(x):
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            alpha = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
            f = (1-alpha)*self.xInterpolators[y_pos-1](x) + alpha*self.xInterpolators[y_pos](x)
        else:
            m = len(x)
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            f = np.zeros(m) + np.nan
            if y.size > 0:
                for i in xrange(1,self.y_n):
                    c = y_pos == i
                    if np.any(c):
                        alpha = (y[c] - self.y_list[i-1])/(self.y_list[i] - self.y_list[i-1])
                        f[c] = (1-alpha)*self.xInterpolators[i-1](x[c]) + alpha*self.xInterpolators[i](x[c]) 
        return f
        
    def _derX(self,x,y):
        '''
        Returns the derivative with respect to x of the interpolated function
        at each value in x,y. Only called internally by HARKinterpolator2D.derivativeX.
        '''
        if _isscalar(x):
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            alpha = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
            dfdx = (1-alpha)*self.xInterpolators[y_pos-1]._der(x) + alpha*self.xInterpolators[y_pos]._der(x)
        else:
            m = len(x)
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            dfdx = np.zeros(m) + np.nan
            if y.size > 0:
                for i in xrange(1,self.y_n):
                    c = y_pos == i
                    if np.any(c):
                        alpha = (y[c] - self.y_list[i-1])/(self.y_list[i] - self.y_list[i-1])
                        dfdx[c] = (1-alpha)*self.xInterpolators[i-1]._der(x[c]) + alpha*self.xInterpolators[i]._der(x[c])
        return dfdx
        
    def _derY(self,x,y):
        '''
        Returns the derivative with respect to y of the interpolated function
        at each value in x,y. Only called internally by HARKinterpolator2D.derivativeY.
        '''
        if _isscalar(x):
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            dfdy = (self.xInterpolators[y_pos](x) - self.xInterpolators[y_pos-1](x))/(self.y_list[y_pos] - self.y_list[y_pos-1])
        else:
            m = len(x)
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            dfdy = np.zeros(m) + np.nan
            if y.size > 0:
                for i in xrange(1,self.y_n):
                    c = y_pos == i
                    if np.any(c):
                        dfdy[c] = (self.xInterpolators[i](x[c]) - self.xInterpolators[i-1](x[c]))/(self.y_list[i] - self.y_list[i-1])
        return dfdy


class BilinearInterpOnInterp1D(HARKinterpolator3D):
    '''
    A 3D interpolator that bilinearly interpolates among a list of lists of 1D
    interpolators.
    '''
    def __init__(self,xInterpolators,y_values,z_values):
        '''
        Constructor for the class, generating an approximation to a function of
        the form f(x,y,z) using interpolations over f(x,y_0,z_0) for a fixed grid
        of y_0 and z_0 values.
        
        Parameters
        ----------
        xInterpolators : [[HARKinterpolator1D]]
            A list of lists of 1D interpolations over the x variable.  The i,j-th
            element of xInterpolators represents f(x,y_values[i],z_values[j]).
        y_values: numpy.array
            An array of y values equal in length to xInterpolators.
        z_values: numpy.array
            An array of z values equal in length to xInterpolators[0].
            
        Returns
        -------
        new instance of BilinearInterpOnInterp1D
        '''
        self.xInterpolators = xInterpolators
        self.y_list = y_values
        self.y_n = y_values.size
        self.z_list = z_values
        self.z_n = z_values.size
        self.distance_criteria = ['xInterpolators','y_list','z_list']
        
    def _evaluate(self,x,y,z):
        '''
        Returns the level of the interpolated function at each value in x,y,z.
        Only called internally by HARKinterpolator3D.__call__ (etc).
        '''
        if _isscalar(x):
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
            beta = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            f = ((1-alpha)*(1-beta)*self.xInterpolators[y_pos-1][z_pos-1](x)
              + (1-alpha)*beta*self.xInterpolators[y_pos-1][z_pos](x)
              + alpha*(1-beta)*self.xInterpolators[y_pos][z_pos-1](x)
              + alpha*beta*self.xInterpolators[y_pos][z_pos](x))              
        else:
            m = len(x)
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            f = np.zeros(m) + np.nan
            for i in xrange(1,self.y_n):
                for j in xrange(1,self.z_n):
                    c = np.logical_and(i == y_pos, j == z_pos)
                    if np.any(c):
                        alpha = (y[c] - self.y_list[i-1])/(self.y_list[i] - self.y_list[i-1])
                        beta = (z[c] - self.z_list[j-1])/(self.z_list[j] - self.z_list[j-1])
                        f[c] = (
                          (1-alpha)*(1-beta)*self.xInterpolators[i-1][j-1](x[c])
                          + (1-alpha)*beta*self.xInterpolators[i-1][j](x[c])
                          + alpha*(1-beta)*self.xInterpolators[i][j-1](x[c])
                          + alpha*beta*self.xInterpolators[i][j](x[c]))
        return f
        
    def _derX(self,x,y,z):
        '''
        Returns the derivative with respect to x of the interpolated function
        at each value in x,y,z. Only called internally by HARKinterpolator3D.derivativeX.
        '''
        if _isscalar(x):
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
            beta = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            dfdx = ((1-alpha)*(1-beta)*self.xInterpolators[y_pos-1][z_pos-1]._der(x)
              + (1-alpha)*beta*self.xInterpolators[y_pos-1][z_pos]._der(x)
              + alpha*(1-beta)*self.xInterpolators[y_pos][z_pos-1]._der(x)
              + alpha*beta*self.xInterpolators[y_pos][z_pos]._der(x))              
        else:
            m = len(x)
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdx = np.zeros(m) + np.nan
            for i in xrange(1,self.y_n):
                for j in xrange(1,self.z_n):
                    c = np.logical_and(i == y_pos, j == z_pos)
                    if np.any(c):
                        alpha = (y[c] - self.y_list[i-1])/(self.y_list[i] - self.y_list[i-1])
                        beta = (z[c] - self.z_list[j-1])/(self.z_list[j] - self.z_list[j-1])
                        dfdx[c] = (
                          (1-alpha)*(1-beta)*self.xInterpolators[i-1][j-1]._der(x[c])
                          + (1-alpha)*beta*self.xInterpolators[i-1][j]._der(x[c])
                          + alpha*(1-beta)*self.xInterpolators[i][j-1]._der(x[c])
                          + alpha*beta*self.xInterpolators[i][j]._der(x[c]))
        return dfdx
        
    def _derY(self,x,y,z):
        '''
        Returns the derivative with respect to y of the interpolated function
        at each value in x,y,z. Only called internally by HARKinterpolator3D.derivativeY.
        '''
        if _isscalar(x):
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            beta = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            dfdy = (((1-beta)*self.xInterpolators[y_pos][z_pos-1](x) + beta*self.xInterpolators[y_pos][z_pos](x))
                 -  ((1-beta)*self.xInterpolators[y_pos-1][z_pos-1](x) + beta*self.xInterpolators[y_pos-1][z_pos](x)))/(self.y_list[y_pos] - self.y_list[y_pos-1])
        else:
            m = len(x)
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdy = np.zeros(m) + np.nan
            for i in xrange(1,self.y_n):
                for j in xrange(1,self.z_n):
                    c = np.logical_and(i == y_pos, j == z_pos)
                    if np.any(c):
                        beta = (z[c] - self.z_list[j-1])/(self.z_list[j] - self.z_list[j-1])
                        dfdy[c] = (((1-beta)*self.xInterpolators[i][j-1](x[c]) + beta*self.xInterpolators[i][j](x[c]))
                                -  ((1-beta)*self.xInterpolators[i-1][j-1](x[c]) + beta*self.xInterpolators[i-1][j](x[c])))/(self.y_list[i] - self.y_list[i-1])
        return dfdy
        
    def _derZ(self,x,y,z):
        '''
        Returns the derivative with respect to z of the interpolated function
        at each value in x,y,z. Only called internally by HARKinterpolator3D.derivativeZ.
        '''
        if _isscalar(x):
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
            dfdz = (((1-alpha)*self.xInterpolators[y_pos-1][z_pos](x) + alpha*self.xInterpolators[y_pos][z_pos](x))
                 -  ((1-alpha)*self.xInterpolators[y_pos-1][z_pos-1](x) + alpha*self.xInterpolators[y_pos][z_pos-1](x)))/(self.z_list[z_pos] - self.z_list[z_pos-1])
        else:
            m = len(x)
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdz = np.zeros(m) + np.nan
            for i in xrange(1,self.y_n):
                for j in xrange(1,self.z_n):
                    c = np.logical_and(i == y_pos, j == z_pos)
                    if np.any(c):
                        alpha = (y[c] - self.y_list[i-1])/(self.y_list[i] - self.y_list[i-1])
                        dfdz[c] = (((1-alpha)*self.xInterpolators[i-1][j](x[c]) + alpha*self.xInterpolators[i][j](x[c]))
                                -  ((1-alpha)*self.xInterpolators[i-1][j-1](x[c]) + alpha*self.xInterpolators[i][j-1](x[c])))/(self.z_list[j] - self.z_list[j-1])
        return dfdz

             

class TrilinearInterpOnInterp1D(HARKinterpolator4D):
    '''
    A 4D interpolator that trilinearly interpolates among a list of lists of 1D interpolators.
    '''    
    def __init__(self,wInterpolators,x_values,y_values,z_values):
        '''
        Constructor for the class, generating an approximation to a function of
        the form f(w,x,y,z) using interpolations over f(w,x_0,y_0,z_0) for a fixed
        grid of y_0 and z_0 values.
        
        Parameters
        ----------
        wInterpolators : [[[HARKinterpolator1D]]]
            A list of lists of lists of 1D interpolations over the x variable.
            The i,j,k-th element of wInterpolators represents f(w,x_values[i],y_values[j],z_values[k]).
        x_values: numpy.array
            An array of x values equal in length to wInterpolators.
        y_values: numpy.array
            An array of y values equal in length to wInterpolators[0].
        z_values: numpy.array
            An array of z values equal in length to wInterpolators[0][0]
        
        Returns
        -------
        new instance of TrilinearInterpOnInterp1D
        '''
        self.wInterpolators = wInterpolators
        self.x_list = x_values
        self.x_n = x_values.size
        self.y_list = y_values
        self.y_n = y_values.size
        self.z_list = z_values
        self.z_n = z_values.size
        self.distance_criteria = ['wInterpolators','x_list','y_list','z_list']

    def _evaluate(self,w,x,y,z):
        '''
        Returns the level of the interpolated function at each value in w,x,y,z.
        Only called internally by HARKinterpolator4D.__call__ (etc).
        '''
        if _isscalar(w):
            x_pos = max(min(np.searchsorted(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (x - self.x_list[x_pos-1])/(self.x_list[x_pos] - self.x_list[x_pos-1])
            beta = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
            gamma = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            f = (
                (1-alpha)*(1-beta)*(1-gamma)*self.wInterpolators[x_pos-1][y_pos-1][z_pos-1](w)
              + (1-alpha)*(1-beta)*gamma*self.wInterpolators[x_pos-1][y_pos-1][z_pos](w)
              + (1-alpha)*beta*(1-gamma)*self.wInterpolators[x_pos-1][y_pos][z_pos-1](w)
              + (1-alpha)*beta*gamma*self.wInterpolators[x_pos-1][y_pos][z_pos](w)
              + alpha*(1-beta)*(1-gamma)*self.wInterpolators[x_pos][y_pos-1][z_pos-1](w)
              + alpha*(1-beta)*gamma*self.wInterpolators[x_pos][y_pos-1][z_pos](w)
              + alpha*beta*(1-gamma)*self.wInterpolators[x_pos][y_pos][z_pos-1](w)
              + alpha*beta*gamma*self.wInterpolators[x_pos][y_pos][z_pos](w))
        else:
            m = len(x)
            x_pos = np.searchsorted(self.x_list,x)
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            f = np.zeros(m) + np.nan
            for i in xrange(1,self.x_n):
                for j in xrange(1,self.y_n):
                    for k in xrange(1,self.z_n):
                        c = np.logical_and(np.logical_and(i == x_pos, j == y_pos),k == z_pos)
                        if np.any(c):
                            alpha = (x[c] - self.x_list[i-1])/(self.x_list[i] - self.x_list[i-1])
                            beta = (y[c] - self.y_list[j-1])/(self.y_list[j] - self.y_list[j-1])
                            gamma = (z[c] - self.z_list[k-1])/(self.z_list[k] - self.z_list[k-1])
                            f[c] = (
                                (1-alpha)*(1-beta)*(1-gamma)*self.wInterpolators[i-1][j-1][k-1](w[c])
                              + (1-alpha)*(1-beta)*gamma*self.wInterpolators[i-1][j-1][k](w[c])
                              + (1-alpha)*beta*(1-gamma)*self.wInterpolators[i-1][j][k-1](w[c])
                              + (1-alpha)*beta*gamma*self.wInterpolators[i-1][j][k](w[c])
                              + alpha*(1-beta)*(1-gamma)*self.wInterpolators[i][j-1][k-1](w[c])
                              + alpha*(1-beta)*gamma*self.wInterpolators[i][j-1][k](w[c])
                              + alpha*beta*(1-gamma)*self.wInterpolators[i][j][k-1](w[c])
                              + alpha*beta*gamma*self.wInterpolators[i][j][k](w[c]))
        return f
        
    def _derW(self,w,x,y,z):
        '''
        Returns the derivative with respect to w of the interpolated function
        at each value in w,x,y,z. Only called internally by HARKinterpolator4D.derivativeW.
        '''
        if _isscalar(w):
            x_pos = max(min(np.searchsorted(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (x - self.x_list[x_pos-1])/(self.x_list[x_pos] - self.x_list[x_pos-1])
            beta = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
            gamma = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            dfdw = (
                (1-alpha)*(1-beta)*(1-gamma)*self.wInterpolators[x_pos-1][y_pos-1][z_pos-1]._der(w)
              + (1-alpha)*(1-beta)*gamma*self.wInterpolators[x_pos-1][y_pos-1][z_pos]._der(w)
              + (1-alpha)*beta*(1-gamma)*self.wInterpolators[x_pos-1][y_pos][z_pos-1]._der(w)
              + (1-alpha)*beta*gamma*self.wInterpolators[x_pos-1][y_pos][z_pos]._der(w)
              + alpha*(1-beta)*(1-gamma)*self.wInterpolators[x_pos][y_pos-1][z_pos-1]._der(w)
              + alpha*(1-beta)*gamma*self.wInterpolators[x_pos][y_pos-1][z_pos]._der(w)
              + alpha*beta*(1-gamma)*self.wInterpolators[x_pos][y_pos][z_pos-1]._der(w)
              + alpha*beta*gamma*self.wInterpolators[x_pos][y_pos][z_pos]._der(w))
        else:
            m = len(x)
            x_pos = np.searchsorted(self.x_list,x)
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdw = np.zeros(m) + np.nan
            for i in xrange(1,self.x_n):
                for j in xrange(1,self.y_n):
                    for k in xrange(1,self.z_n):
                        c = np.logical_and(np.logical_and(i == x_pos, j == y_pos),k == z_pos)
                        if np.any(c):
                            alpha = (x[c] - self.x_list[i-1])/(self.x_list[i] - self.x_list[i-1])
                            beta = (y[c] - self.y_list[j-1])/(self.y_list[j] - self.y_list[j-1])
                            gamma = (z[c] - self.z_list[k-1])/(self.z_list[k] - self.z_list[k-1])
                            dfdw[c] = (
                                (1-alpha)*(1-beta)*(1-gamma)*self.wInterpolators[i-1][j-1][k-1]._der(w[c])
                              + (1-alpha)*(1-beta)*gamma*self.wInterpolators[i-1][j-1][k]._der(w[c])
                              + (1-alpha)*beta*(1-gamma)*self.wInterpolators[i-1][j][k-1]._der(w[c])
                              + (1-alpha)*beta*gamma*self.wInterpolators[i-1][j][k]._der(w[c])
                              + alpha*(1-beta)*(1-gamma)*self.wInterpolators[i][j-1][k-1]._der(w[c])
                              + alpha*(1-beta)*gamma*self.wInterpolators[i][j-1][k]._der(w[c])
                              + alpha*beta*(1-gamma)*self.wInterpolators[i][j][k-1]._der(w[c])
                              + alpha*beta*gamma*self.wInterpolators[i][j][k]._der(w[c]))
        return dfdw
        
    def _derX(self,w,x,y,z):
        '''
        Returns the derivative with respect to x of the interpolated function
        at each value in w,x,y,z. Only called internally by HARKinterpolator4D.derivativeX.
        '''
        if _isscalar(w):
            x_pos = max(min(np.searchsorted(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            beta = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
            gamma = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            dfdx = (
             ((1-beta)*(1-gamma)*self.wInterpolators[x_pos][y_pos-1][z_pos-1](w)
              + (1-beta)*gamma*self.wInterpolators[x_pos][y_pos-1][z_pos](w)
              + beta*(1-gamma)*self.wInterpolators[x_pos][y_pos][z_pos-1](w)
              + beta*gamma*self.wInterpolators[x_pos][y_pos][z_pos](w)) - 
              ((1-beta)*(1-gamma)*self.wInterpolators[x_pos-1][y_pos-1][z_pos-1](w)
              + (1-beta)*gamma*self.wInterpolators[x_pos-1][y_pos-1][z_pos](w)
              + beta*(1-gamma)*self.wInterpolators[x_pos-1][y_pos][z_pos-1](w)
              + beta*gamma*self.wInterpolators[x_pos-1][y_pos][z_pos](w)))/(self.x_list[x_pos] - self.x_list[x_pos-1])
        else:
            m = len(x)
            x_pos = np.searchsorted(self.x_list,x)
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdx = np.zeros(m) + np.nan
            for i in xrange(1,self.x_n):
                for j in xrange(1,self.y_n):
                    for k in xrange(1,self.z_n):
                        c = np.logical_and(np.logical_and(i == x_pos, j == y_pos),k == z_pos)
                        if np.any(c):
                            beta = (y[c] - self.y_list[j-1])/(self.y_list[j] - self.y_list[j-1])
                            gamma = (z[c] - self.z_list[k-1])/(self.z_list[k] - self.z_list[k-1])
                            dfdx[c] = (
                              ((1-beta)*(1-gamma)*self.wInterpolators[i][j-1][k-1](w[c])
                            + (1-beta)*gamma*self.wInterpolators[i][j-1][k](w[c])
                            + beta*(1-gamma)*self.wInterpolators[i][j][k-1](w[c])
                            + beta*gamma*self.wInterpolators[i][j][k](w[c])) - 
                             ((1-beta)*(1-gamma)*self.wInterpolators[i-1][j-1][k-1](w[c])
                            + (1-beta)*gamma*self.wInterpolators[i-1][j-1][k](w[c])
                            + beta*(1-gamma)*self.wInterpolators[i-1][j][k-1](w[c])
                            + beta*gamma*self.wInterpolators[i-1][j][k](w[c])))/(self.x_list[i] - self.x_list[i-1])
        return dfdx
        
    def _derY(self,w,x,y,z):
        '''
        Returns the derivative with respect to y of the interpolated function
        at each value in w,x,y,z. Only called internally by HARKinterpolator4D.derivativeY.
        '''
        if _isscalar(w):
            x_pos = max(min(np.searchsorted(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (x - self.x_list[x_pos-1])/(self.y_list[x_pos] - self.x_list[x_pos-1])
            gamma = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            dfdy = (
             ((1-alpha)*(1-gamma)*self.wInterpolators[x_pos-1][y_pos][z_pos-1](w)
              + (1-alpha)*gamma*self.wInterpolators[x_pos-1][y_pos][z_pos](w)
              + alpha*(1-gamma)*self.wInterpolators[x_pos][y_pos][z_pos-1](w)
              + alpha*gamma*self.wInterpolators[x_pos][y_pos][z_pos](w)) - 
              ((1-alpha)*(1-gamma)*self.wInterpolators[x_pos-1][y_pos-1][z_pos-1](w)
              + (1-alpha)*gamma*self.wInterpolators[x_pos-1][y_pos-1][z_pos](w)
              + alpha*(1-gamma)*self.wInterpolators[x_pos][y_pos-1][z_pos-1](w)
              + alpha*gamma*self.wInterpolators[x_pos][y_pos-1][z_pos](w)))/(self.y_list[y_pos] - self.y_list[y_pos-1])
        else:
            m = len(x)
            x_pos = np.searchsorted(self.x_list,x)
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdy = np.zeros(m) + np.nan
            for i in xrange(1,self.x_n):
                for j in xrange(1,self.y_n):
                    for k in xrange(1,self.z_n):
                        c = np.logical_and(np.logical_and(i == x_pos, j == y_pos),k == z_pos)
                        if np.any(c):
                            alpha = (x[c] - self.x_list[i-1])/(self.x_list[i] - self.x_list[i-1])
                            gamma = (z[c] - self.z_list[k-1])/(self.z_list[k] - self.z_list[k-1])
                            dfdy[c] = (
                                  ((1-alpha)*(1-gamma)*self.wInterpolators[i-1][j][k-1](w[c])
                                 + (1-alpha)*gamma*self.wInterpolators[i-1][j][k](w[c])
                                 + alpha*(1-gamma)*self.wInterpolators[i][j][k-1](w[c])
                                 + alpha*gamma*self.wInterpolators[i][j][k](w[c])) - 
                                  ((1-alpha)*(1-gamma)*self.wInterpolators[i-1][j-1][k-1](w[c])
                                 + (1-alpha)*gamma*self.wInterpolators[i-1][j-1][k](w[c])
                                 + alpha*(1-gamma)*self.wInterpolators[i][j-1][k-1](w[c])
                                 + alpha*gamma*self.wInterpolators[i][j-1][k](w[c])))/(self.y_list[j] - self.y_list[j-1])
        return dfdy
        
    def _derZ(self,w,x,y,z):
        '''
        Returns the derivative with respect to z of the interpolated function
        at each value in w,x,y,z. Only called internally by HARKinterpolator4D.derivativeZ.
        '''
        if _isscalar(w):
            x_pos = max(min(np.searchsorted(self.x_list,x),self.x_n-1),1)
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (x - self.x_list[x_pos-1])/(self.y_list[x_pos] - self.x_list[x_pos-1])
            beta = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
            dfdz = (
             ((1-alpha)*(1-beta)*self.wInterpolators[x_pos-1][y_pos-1][z_pos](w)
              + (1-alpha)*beta*self.wInterpolators[x_pos-1][y_pos][z_pos](w)
              + alpha*(1-beta)*self.wInterpolators[x_pos][y_pos-1][z_pos](w)
              + alpha*beta*self.wInterpolators[x_pos][y_pos][z_pos](w)) - 
              ((1-alpha)*(1-beta)*self.wInterpolators[x_pos-1][y_pos-1][z_pos-1](w)
              + (1-alpha)*beta*self.wInterpolators[x_pos-1][y_pos][z_pos-1](w)
              + alpha*(1-beta)*self.wInterpolators[x_pos][y_pos-1][z_pos-1](w)
              + alpha*beta*self.wInterpolators[x_pos][y_pos][z_pos-1](w)))/(self.z_list[z_pos] - self.z_list[z_pos-1])
        else:
            m = len(x)
            x_pos = np.searchsorted(self.x_list,x)
            x_pos[x_pos > self.x_n-1] = self.x_n-1
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdz = np.zeros(m) + np.nan
            for i in xrange(1,self.x_n):
                for j in xrange(1,self.y_n):
                    for k in xrange(1,self.z_n):
                        c = np.logical_and(np.logical_and(i == x_pos, j == y_pos),k == z_pos)
                        if np.any(c):
                            alpha = (x[c] - self.x_list[i-1])/(self.x_list[i] - self.x_list[i-1])
                            beta = (y[c] - self.y_list[j-1])/(self.y_list[j] - self.y_list[j-1])
                            dfdz[c] = (
                                  ((1-alpha)*(1-beta)*self.wInterpolators[i-1][j-1][k](w[c])
                                 + (1-alpha)*beta*self.wInterpolators[i-1][j][k](w[c])
                                 + alpha*(1-beta)*self.wInterpolators[i][j-1][k](w[c])
                                 + alpha*beta*self.wInterpolators[i][j][k](w[c])) - 
                                  ((1-alpha)*(1-beta)*self.wInterpolators[i-1][j-1][k-1](w[c])
                                 + (1-alpha)*beta*self.wInterpolators[i-1][j][k-1](w[c])
                                 + alpha*(1-beta)*self.wInterpolators[i][j-1][k-1](w[c])
                                 + alpha*beta*self.wInterpolators[i][j][k-1](w[c])))/(self.z_list[k] - self.z_list[k-1])
        return dfdz
        
       
class LinearInterpOnInterp2D(HARKinterpolator3D):
    '''
    A 3D interpolation method that linearly interpolates between "layers" of
    arbitrary 2D interpolations.  Useful for models with two endogenous state
    variables and one exogenous state variable when solving with the endogenous
    grid method.  NOTE: should not be used if an exogenous 3D grid is used, will
    be significantly slower than TrilinearInterp.
    '''    
    def __init__(self,xyInterpolators,z_values):
        '''
        Constructor for the class, generating an approximation to a function of
        the form f(x,y,z) using interpolations over f(x,y,z_0) for a fixed grid
        of z_0 values.
        
        Parameters
        ----------
        xyInterpolators : [HARKinterpolator2D]
            A list of 2D interpolations over the x and y variables.  The nth
            element of xyInterpolators represents f(x,y,z_values[n]).
        z_values: numpy.array
            An array of z values equal in length to xyInterpolators.
            
        Returns
        -------
        new instance of LinearInterpOnInterp2D
        '''
        self.xyInterpolators = xyInterpolators
        self.z_list = z_values
        self.z_n = z_values.size
        self.distance_criteria = ['xyInterpolators','z_list']
        
    def _evaluate(self,x,y,z):
        '''
        Returns the level of the interpolated function at each value in x,y,z.
        Only called internally by HARKinterpolator3D.__call__ (etc).
        '''
        if _isscalar(x):
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            f = (1-alpha)*self.xyInterpolators[z_pos-1](x,y) + alpha*self.xyInterpolators[z_pos](x,y)
        else:
            m = len(x)
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            f = np.zeros(m) + np.nan
            if x.size > 0:
                for i in xrange(1,self.z_n):
                    c = z_pos == i
                    if np.any(c):
                        alpha = (z[c] - self.z_list[i-1])/(self.z_list[i] - self.z_list[i-1])
                        f[c] = (1-alpha)*self.xyInterpolators[i-1](x[c],y[c]) + alpha*self.xyInterpolators[i](x[c],y[c]) 
        return f
        
    def _derX(self,x,y,z):
        '''
        Returns the derivative with respect to x of the interpolated function
        at each value in x,y,z. Only called internally by HARKinterpolator3D.derivativeX.
        '''
        if _isscalar(x):
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            dfdx = (1-alpha)*self.xyInterpolators[z_pos-1].derivativeX(x,y) + alpha*self.xyInterpolators[z_pos].derivativeX(x,y)
        else:
            m = len(x)
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdx = np.zeros(m) + np.nan
            if x.size > 0:
                for i in xrange(1,self.z_n):
                    c = z_pos == i
                    if np.any(c):
                        alpha = (z[c] - self.z_list[i-1])/(self.z_list[i] - self.z_list[i-1])
                        dfdx[c] = (1-alpha)*self.xyInterpolators[i-1].derivativeX(x[c],y[c]) + alpha*self.xyInterpolators[i].derivativeX(x[c],y[c]) 
        return dfdx
        
    def _derY(self,x,y,z):
        '''
        Returns the derivative with respect to y of the interpolated function
        at each value in x,y,z. Only called internally by HARKinterpolator3D.derivativeY.
        '''
        if _isscalar(x):
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            dfdy = (1-alpha)*self.xyInterpolators[z_pos-1].derivativeY(x,y) + alpha*self.xyInterpolators[z_pos].derivativeY(x,y)
        else:
            m = len(x)
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdy = np.zeros(m) + np.nan
            if x.size > 0:
                for i in xrange(1,self.z_n):
                    c = z_pos == i
                    if np.any(c):
                        alpha = (z[c] - self.z_list[i-1])/(self.z_list[i] - self.z_list[i-1])
                        dfdy[c] = (1-alpha)*self.xyInterpolators[i-1].derivativeY(x[c],y[c]) + alpha*self.xyInterpolators[i].derivativeY(x[c],y[c]) 
        return dfdy
        
    def _derZ(self,x,y,z):
        '''
        Returns the derivative with respect to z of the interpolated function
        at each value in x,y,z. Only called internally by HARKinterpolator3D.derivativeZ.
        '''
        if _isscalar(x):
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            dfdz = (self.xyInterpolators[z_pos].derivativeX(x,y) - self.xyInterpolators[z_pos-1].derivativeX(x,y))/(self.z_list[z_pos] - self.z_list[z_pos-1])
        else:
            m = len(x)
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdz = np.zeros(m) + np.nan
            if x.size > 0:
                for i in xrange(1,self.z_n):
                    c = z_pos == i
                    if np.any(c):
                        dfdz[c] = (self.xyInterpolators[i](x[c],y[c]) - self.xyInterpolators[i-1](x[c],y[c]))/(self.z_list[i] - self.z_list[i-1])
        return dfdz

class BilinearInterpOnInterp2D(HARKinterpolator4D):
    '''
    A 4D interpolation method that bilinearly interpolates among "layers" of
    arbitrary 2D interpolations.  Useful for models with two endogenous state
    variables and two exogenous state variables when solving with the endogenous
    grid method.  NOTE: should not be used if an exogenous 4D grid is used, will
    be significantly slower than QuadlinearInterp.
    '''    
    def __init__(self,wxInterpolators,y_values,z_values):
        '''
        Constructor for the class, generating an approximation to a function of
        the form f(w,x,y,z) using interpolations over f(w,x,y_0,z_0) for a fixed
        grid of y_0 and z_0 values.
        
        Parameters
        ----------
        wxInterpolators : [[HARKinterpolator2D]]
            A list of lists of 2D interpolations over the w and x variables.
            The i,j-th element of wxInterpolators represents
            f(w,x,y_values[i],z_values[j]).
        y_values: numpy.array
            An array of y values equal in length to wxInterpolators.
        z_values: numpy.array
            An array of z values equal in length to wxInterpolators[0].
            
        Returns
        -------
        new instance of BilinearInterpOnInterp2D
        '''
        self.wxInterpolators = wxInterpolators
        self.y_list = y_values
        self.y_n = y_values.size
        self.z_list = z_values
        self.z_n = z_values.size
        self.distance_criteria = ['wxInterpolators','y_list','z_list']
        
    def _evaluate(self,w,x,y,z):
        '''
        Returns the level of the interpolated function at each value in x,y,z.
        Only called internally by HARKinterpolator4D.__call__ (etc).
        '''
        if _isscalar(x):
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
            beta = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            f = ((1-alpha)*(1-beta)*self.wxInterpolators[y_pos-1][z_pos-1](w,x)
              + (1-alpha)*beta*self.wxInterpolators[y_pos-1][z_pos](w,x)
              + alpha*(1-beta)*self.wxInterpolators[y_pos][z_pos-1](w,x)
              + alpha*beta*self.wxInterpolators[y_pos][z_pos](w,x))              
        else:
            m = len(x)
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            f = np.zeros(m) + np.nan
            for i in xrange(1,self.y_n):
                for j in xrange(1,self.z_n):
                    c = np.logical_and(i == y_pos, j == z_pos)
                    if np.any(c):
                        alpha = (y[c] - self.y_list[i-1])/(self.y_list[i] - self.y_list[i-1])
                        beta = (z[c] - self.z_list[j-1])/(self.z_list[j] - self.z_list[j-1])
                        f[c] = (
                          (1-alpha)*(1-beta)*self.wxInterpolators[i-1][j-1](w[c],x[c])
                          + (1-alpha)*beta*self.wxInterpolators[i-1][j](w[c],x[c])
                          + alpha*(1-beta)*self.wxInterpolators[i][j-1](w[c],x[c])
                          + alpha*beta*self.wxInterpolators[i][j](w[c],x[c]))
        return f
        
    def _derW(self,w,x,y,z):
        '''
        Returns the derivative with respect to w of the interpolated function
        at each value in w,x,y,z. Only called internally by HARKinterpolator4D.derivativeW.
        '''
        # This may look strange, as we call the derivativeX() method to get the
        # derivative with respect to w, but that's just a quirk of 4D interpolations
        # beginning with w rather than x.  The derivative wrt the first dimension
        # of an element of wxInterpolators is the w-derivative of the main function.
        if _isscalar(x):
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
            beta = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            dfdw = ((1-alpha)*(1-beta)*self.wxInterpolators[y_pos-1][z_pos-1].derivativeX(w,x)
              + (1-alpha)*beta*self.wxInterpolators[y_pos-1][z_pos].derivativeX(w,x)
              + alpha*(1-beta)*self.wxInterpolators[y_pos][z_pos-1].derivativeX(w,x)
              + alpha*beta*self.wxInterpolators[y_pos][z_pos].derivativeX(w,x))              
        else:
            m = len(x)
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdw = np.zeros(m) + np.nan
            for i in xrange(1,self.y_n):
                for j in xrange(1,self.z_n):
                    c = np.logical_and(i == y_pos, j == z_pos)
                    if np.any(c):
                        alpha = (y[c] - self.y_list[i-1])/(self.y_list[i] - self.y_list[i-1])
                        beta = (z[c] - self.z_list[j-1])/(self.z_list[j] - self.z_list[j-1])
                        dfdw[c] = (
                          (1-alpha)*(1-beta)*self.wxInterpolators[i-1][j-1].derivativeX(w[c],x[c])
                          + (1-alpha)*beta*self.wxInterpolators[i-1][j].derivativeX(w[c],x[c])
                          + alpha*(1-beta)*self.wxInterpolators[i][j-1].derivativeX(w[c],x[c])
                          + alpha*beta*self.wxInterpolators[i][j].derivativeX(w[c],x[c]))
        return dfdw
        
    def _derX(self,w,x,y,z):
        '''
        Returns the derivative with respect to x of the interpolated function
        at each value in w,x,y,z. Only called internally by HARKinterpolator4D.derivativeX.
        '''
        # This may look strange, as we call the derivativeY() method to get the
        # derivative with respect to x, but that's just a quirk of 4D interpolations
        # beginning with w rather than x.  The derivative wrt the second dimension
        # of an element of wxInterpolators is the x-derivative of the main function.
        if _isscalar(x):
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
            beta = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            dfdx = ((1-alpha)*(1-beta)*self.wxInterpolators[y_pos-1][z_pos-1].derivativeY(w,x)
              + (1-alpha)*beta*self.wxInterpolators[y_pos-1][z_pos].derivativeY(w,x)
              + alpha*(1-beta)*self.wxInterpolators[y_pos][z_pos-1].derivativeY(w,x)
              + alpha*beta*self.wxInterpolators[y_pos][z_pos].derivativeY(w,x))              
        else:
            m = len(x)
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdx = np.zeros(m) + np.nan
            for i in xrange(1,self.y_n):
                for j in xrange(1,self.z_n):
                    c = np.logical_and(i == y_pos, j == z_pos)
                    if np.any(c):
                        alpha = (y[c] - self.y_list[i-1])/(self.y_list[i] - self.y_list[i-1])
                        beta = (z[c] - self.z_list[j-1])/(self.z_list[j] - self.z_list[j-1])
                        dfdx[c] = (
                          (1-alpha)*(1-beta)*self.wxInterpolators[i-1][j-1].derivativeY(w[c],x[c])
                          + (1-alpha)*beta*self.wxInterpolators[i-1][j].derivativeY(w[c],x[c])
                          + alpha*(1-beta)*self.wxInterpolators[i][j-1].derivativeY(w[c],x[c])
                          + alpha*beta*self.wxInterpolators[i][j].derivativeY(w[c],x[c]))
        return dfdx
        
    def _derY(self,w,x,y,z):
        '''
        Returns the derivative with respect to y of the interpolated function
        at each value in w,x,y,z. Only called internally by HARKinterpolator4D.derivativeY.
        '''
        if _isscalar(x):
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            beta = (z - self.z_list[z_pos-1])/(self.z_list[z_pos] - self.z_list[z_pos-1])
            dfdy = (((1-beta)*self.wxInterpolators[y_pos][z_pos-1](w,x) + beta*self.wxInterpolators[y_pos][z_pos](w,x))
                 -  ((1-beta)*self.wxInterpolators[y_pos-1][z_pos-1](w,x) + beta*self.wxInterpolators[y_pos-1][z_pos](w,x)))/(self.y_list[y_pos] - self.y_list[y_pos-1])
        else:
            m = len(x)
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdy = np.zeros(m) + np.nan
            for i in xrange(1,self.y_n):
                for j in xrange(1,self.z_n):
                    c = np.logical_and(i == y_pos, j == z_pos)
                    if np.any(c):
                        beta = (z[c] - self.z_list[j-1])/(self.z_list[j] - self.z_list[j-1])
                        dfdy[c] = (((1-beta)*self.wxInterpolators[i][j-1](w[c],x[c]) + beta*self.wxInterpolators[i][j](w[c],x[c]))
                                -  ((1-beta)*self.wxInterpolators[i-1][j-1](w[c],x[c]) + beta*self.wxInterpolators[i-1][j](w[c],x[c])))/(self.y_list[i] - self.y_list[i-1])
        return dfdy
        
    def _derZ(self,w,x,y,z):
        '''
        Returns the derivative with respect to z of the interpolated function
        at each value in w,x,y,z. Only called internally by HARKinterpolator4D.derivativeZ.
        '''
        if _isscalar(x):
            y_pos = max(min(np.searchsorted(self.y_list,y),self.y_n-1),1)
            z_pos = max(min(np.searchsorted(self.z_list,z),self.z_n-1),1)
            alpha = (y - self.y_list[y_pos-1])/(self.y_list[y_pos] - self.y_list[y_pos-1])
            dfdz = (((1-alpha)*self.wxInterpolators[y_pos-1][z_pos](w,x) + alpha*self.wxInterpolators[y_pos][z_pos](w,x))
                 -  ((1-alpha)*self.wxInterpolators[y_pos-1][z_pos-1](w,x) + alpha*self.wxInterpolators[y_pos][z_pos-1](w,x)))/(self.z_list[z_pos] - self.z_list[z_pos-1])
        else:
            m = len(x)
            y_pos = np.searchsorted(self.y_list,y)
            y_pos[y_pos > self.y_n-1] = self.y_n-1
            y_pos[y_pos < 1] = 1
            z_pos = np.searchsorted(self.z_list,z)
            z_pos[z_pos > self.z_n-1] = self.z_n-1
            z_pos[z_pos < 1] = 1
            dfdz = np.zeros(m) + np.nan
            for i in xrange(1,self.y_n):
                for j in xrange(1,self.z_n):
                    c = np.logical_and(i == y_pos, j == z_pos)
                    if np.any(c):
                        alpha = (y[c] - self.y_list[i-1])/(self.y_list[i] - self.y_list[i-1])
                        dfdz[c] = (((1-alpha)*self.wxInterpolators[i-1][j](w[c],x[c]) + alpha*self.wxInterpolators[i][j](w[c],x[c]))
                                -  ((1-alpha)*self.wxInterpolators[i-1][j-1](w[c],x[c]) + alpha*self.wxInterpolators[i][j-1](w[c],x[c])))/(self.z_list[j] - self.z_list[j-1])
        return dfdz
        
        
class Curvilinear2DInterp(HARKinterpolator2D):
    '''
    A 2D interpolation method for curvilinear or "warped grid" interpolation, as
    in White (2015).  Used for models with two endogenous states that are solved
    with the endogenous grid method.
    '''
    def __init__(self,f_values,x_values,y_values):
        '''
        Constructor for 2D curvilinear interpolation for a function f(x,y)
        
        Parameters
        ----------
        f_values: numpy.array
            A 2D array of function values such that f_values[i,j] =
            f(x_values[i,j],y_values[i,j]).
        x_values: numpy.array
            A 2D array of x values of the same size as f_values.
        y_values: numpy.array
            A 2D array of y values of the same size as f_values.
            
        Returns
        -------
        new instance of Curvilinear2DInterp
        '''
        self.f_values = f_values
        self.x_values = x_values
        self.y_values = y_values
        my_shape = f_values.shape
        self.x_n = my_shape[0]
        self.y_n = my_shape[1]
        self.updatePolarity()
        self.distance_criteria = ['f_values','x_values','y_values']
        
    def updatePolarity(self):
        '''
        Fills in the polarity attribute of the interpolation, determining whether
        the "plus" (True) or "minus" (False) solution of the system of equations
        should be used for each sector.  Needs to be called in __init__.
        
        Parameters
        ----------
        none
            
        Returns
        -------
        none
        '''       
        # Grab a point known to be inside each sector: the midway point between
        # the lower left and upper right vertex of each sector
        x_temp = 0.5*(self.x_values[0:(self.x_n-1),0:(self.y_n-1)] + self.x_values[1:self.x_n,1:self.y_n])
        y_temp = 0.5*(self.y_values[0:(self.x_n-1),0:(self.y_n-1)] + self.y_values[1:self.x_n,1:self.y_n])
        size = (self.x_n-1)*(self.y_n-1)
        x_temp = np.reshape(x_temp,size)
        y_temp = np.reshape(y_temp,size)
        y_pos = np.tile(np.arange(0,self.y_n-1),self.x_n-1)
        x_pos = np.reshape(np.tile(np.arange(0,self.x_n-1),(self.y_n-1,1)).transpose(),size)
        
        # Set the polarity of all sectors to "plus", then test each sector
        self.polarity = np.ones((self.x_n-1,self.y_n-1),dtype=bool)
        alpha, beta = self.findCoords(x_temp,y_temp,x_pos,y_pos)
        polarity = np.logical_and(
            np.logical_and(alpha > 0, alpha < 1),
            np.logical_and(beta > 0, beta < 1))
        
        # Update polarity: if (alpha,beta) not in the unit square, then that
        # sector must use the "minus" solution instead
        self.polarity = np.reshape(polarity,(self.x_n-1,self.y_n-1))
        
    def findSector(self,x,y):
        '''
        Finds the quadrilateral "sector" for each (x,y) point in the input.
        Only called as a subroutine of _evaluate().
        
        Parameters
        ----------
        x : np.array
            Values whose sector should be found.
        y : np.array
            Values whose sector should be found.  Should be same size as x.
            
        Returns
        -------
        x_pos : np.array
            Sector x-coordinates for each point of the input, of the same size.
        y_pos : np.array
            Sector y-coordinates for each point of the input, of the same size.
        '''
        # Initialize the sector guess
        m = x.size
        x_pos_guess = (np.ones(m)*self.x_n/2).astype(int)
        y_pos_guess = (np.ones(m)*self.y_n/2).astype(int)
        
        # Define a function that checks whether a set of points violates a linear
        # boundary defined by (x_bound_1,y_bound_1) and (x_bound_2,y_bound_2),
        # where the latter is *COUNTER CLOCKWISE* from the former.  Returns
        # 1 if the point is outside the boundary and 0 otherwise.
        violationCheck = lambda x_check,y_check,x_bound_1,y_bound_1,x_bound_2,y_bound_2 : (
            (y_bound_2 - y_bound_1)*x_check - (x_bound_2 - x_bound_1)*y_check > x_bound_1*y_bound_2 - y_bound_1*x_bound_2 ) + 0
        
        # Identify the correct sector for each point to be evaluated
        these = np.ones(m,dtype=bool)
        max_loops = self.x_n + self.y_n
        loops = 0
        while np.any(these) and loops < max_loops:
            # Get coordinates for the four vertices: (xA,yA),...,(xD,yD)
            x_temp = x[these]
            y_temp = y[these]
            xA = self.x_values[x_pos_guess[these],y_pos_guess[these]]
            xB = self.x_values[x_pos_guess[these]+1,y_pos_guess[these]]
            xC = self.x_values[x_pos_guess[these],y_pos_guess[these]+1]
            xD = self.x_values[x_pos_guess[these]+1,y_pos_guess[these]+1]
            yA = self.y_values[x_pos_guess[these],y_pos_guess[these]]
            yB = self.y_values[x_pos_guess[these]+1,y_pos_guess[these]]
            yC = self.y_values[x_pos_guess[these],y_pos_guess[these]+1]
            yD = self.y_values[x_pos_guess[these]+1,y_pos_guess[these]+1]
            
            # Check the "bounding box" for the sector: is this guess plausible?
            move_down = (y_temp < np.minimum(yA,yB)) + 0
            move_right = (x_temp > np.maximum(xB,xD)) + 0
            move_up = (y_temp > np.maximum(yC,yD)) + 0
            move_left = (x_temp < np.minimum(xA,xC)) + 0
            
            # Check which boundaries are violated (and thus where to look next)
            c = (move_down + move_right + move_up + move_left) == 0
            move_down[c] = violationCheck(x_temp[c],y_temp[c],xA[c],yA[c],xB[c],yB[c])
            move_right[c] = violationCheck(x_temp[c],y_temp[c],xB[c],yB[c],xD[c],yD[c])
            move_up[c] = violationCheck(x_temp[c],y_temp[c],xD[c],yD[c],xC[c],yC[c])
            move_left[c] = violationCheck(x_temp[c],y_temp[c],xC[c],yC[c],xA[c],yA[c])
            
            # Update the sector guess based on the violations
            x_pos_next = x_pos_guess[these] - move_left + move_right
            x_pos_next[x_pos_next < 0] = 0
            x_pos_next[x_pos_next > (self.x_n-2)] = self.x_n-2
            y_pos_next = y_pos_guess[these] - move_down + move_up
            y_pos_next[y_pos_next < 0] = 0
            y_pos_next[y_pos_next > (self.y_n-2)] = self.y_n-2
            
            # Check which sectors have not changed, and mark them as complete
            no_move = np.array(np.logical_and(x_pos_guess[these] == x_pos_next, y_pos_guess[these] == y_pos_next))
            x_pos_guess[these] = x_pos_next
            y_pos_guess[these] = y_pos_next
            temp = these.nonzero()
            these[temp[0][no_move]] = False
            
            # Move to the next iteration of the search
            loops += 1
            
        # Return the output
        x_pos = x_pos_guess
        y_pos = y_pos_guess
        return x_pos, y_pos
        
    def findCoords(self,x,y,x_pos,y_pos):
        '''
        Calculates the relative coordinates (alpha,beta) for each point (x,y),
        given the sectors (x_pos,y_pos) in which they reside.  Only called as
        a subroutine of __call__().
        
        Parameters
        ----------
        x : np.array
            Values whose sector should be found.
        y : np.array
            Values whose sector should be found.  Should be same size as x.
        x_pos : np.array
            Sector x-coordinates for each point in (x,y), of the same size.
        y_pos : np.array
            Sector y-coordinates for each point in (x,y), of the same size.
            
        Returns
        -------
        alpha : np.array
            Relative "horizontal" position of the input in their respective sectors.
        beta : np.array
            Relative "vertical" position of the input in their respective sectors.
        '''
        # Calculate relative coordinates in the sector for each point
        xA = self.x_values[x_pos,y_pos]
        xB = self.x_values[x_pos+1,y_pos]
        xC = self.x_values[x_pos,y_pos+1]
        xD = self.x_values[x_pos+1,y_pos+1]
        yA = self.y_values[x_pos,y_pos]
        yB = self.y_values[x_pos+1,y_pos]
        yC = self.y_values[x_pos,y_pos+1]
        yD = self.y_values[x_pos+1,y_pos+1]
        polarity = 2.0*self.polarity[x_pos,y_pos] - 1.0
        a = xA
        b = (xB-xA)
        c = (xC-xA)
        d = (xA-xB-xC+xD)
        e = yA
        f = (yB-yA)
        g = (yC-yA)
        h = (yA-yB-yC+yD)
        denom = (d*g-h*c)
        mu = (h*b-d*f)/denom
        tau = (h*(a-x) - d*(e-y))/denom
        zeta = a - x + c*tau
        eta = b + c*mu + d*tau
        theta = d*mu
        alpha = (-eta + polarity*np.sqrt(eta**2.0 - 4.0*zeta*theta))/(2.0*theta)
        beta = mu*alpha + tau
        
        # Alternate method if there are sectors that are "too regular"
        z = np.logical_or(np.isnan(alpha),np.isnan(beta))   # These points weren't able to identify coordinates
        if np.any(z): 
            these = np.isclose(f/b,(yD-yC)/(xD-xC)) # iso-beta lines have equal slope
            if np.any(these):
                kappa     = f[these]/b[these]
                int_bot   = yA[these] - kappa*xA[these]
                int_top   = yC[these] - kappa*xC[these]
                int_these = y[these] - kappa*x[these]
                beta_temp = (int_these-int_bot)/(int_top-int_bot)
                x_left    = beta_temp*xC[these] + (1.0-beta_temp)*xA[these]
                x_right   = beta_temp*xD[these] + (1.0-beta_temp)*xB[these]
                alpha_temp= (x[these]-x_left)/(x_right-x_left)
                beta[these]  = beta_temp
                alpha[these] = alpha_temp
                
            #print(np.sum(np.isclose(g/c,(yD-yB)/(xD-xB))))
        
        return alpha, beta
        
    def _evaluate(self,x,y):
        '''
        Returns the level of the interpolated function at each value in x,y.
        Only called internally by HARKinterpolator2D.__call__ (etc).
        '''
        x_pos, y_pos = self.findSector(x,y)
        alpha, beta = self.findCoords(x,y,x_pos,y_pos)
        
        # Calculate the function at each point using bilinear interpolation
        f = (
             (1-alpha)*(1-beta)*self.f_values[x_pos,y_pos]
          +  (1-alpha)*beta*self.f_values[x_pos,y_pos+1]
          +  alpha*(1-beta)*self.f_values[x_pos+1,y_pos]
          +  alpha*beta*self.f_values[x_pos+1,y_pos+1])
        return f
        
    def _derX(self,x,y):
        '''
        Returns the derivative with respect to x of the interpolated function
        at each value in x,y. Only called internally by HARKinterpolator2D.derivativeX.
        '''
        x_pos, y_pos = self.findSector(x,y)
        alpha, beta = self.findCoords(x,y,x_pos,y_pos)
        
        # Get four corners data for each point
        xA = self.x_values[x_pos,y_pos]
        xB = self.x_values[x_pos+1,y_pos]
        xC = self.x_values[x_pos,y_pos+1]
        xD = self.x_values[x_pos+1,y_pos+1]
        yA = self.y_values[x_pos,y_pos]
        yB = self.y_values[x_pos+1,y_pos]
        yC = self.y_values[x_pos,y_pos+1]
        yD = self.y_values[x_pos+1,y_pos+1]
        fA = self.f_values[x_pos,y_pos]
        fB = self.f_values[x_pos+1,y_pos]
        fC = self.f_values[x_pos,y_pos+1]
        fD = self.f_values[x_pos+1,y_pos+1]
        
        # Calculate components of the alpha,beta --> x,y delta translation matrix
        alpha_x = (1-beta)*(xB-xA) + beta*(xD-xC)
        alpha_y = (1-beta)*(yB-yA) + beta*(yD-yC)
        beta_x  = (1-alpha)*(xC-xA) + alpha*(xD-xB)
        beta_y  = (1-alpha)*(yC-yA) + alpha*(yD-yB)
        
        # Invert the delta translation matrix into x,y --> alpha,beta
        det = alpha_x*beta_y - beta_x*alpha_y
        x_alpha = beta_y/det
        x_beta  = -alpha_y/det
        #y_alpha = -beta_x/det
        #y_beta  = alpha_x/det
        
        # Calculate the derivative of f w.r.t. alpha and beta
        dfda = (1-beta)*(fB-fA) + beta*(fD-fC)
        dfdb = (1-alpha)*(fC-fA) + alpha*(fD-fB)
        
        # Calculate the derivative with respect to x (and return it)
        dfdx = x_alpha*dfda + x_beta*dfdb
        return dfdx
        
    def _derY(self,x,y):
        '''
        Returns the derivative with respect to y of the interpolated function
        at each value in x,y. Only called internally by HARKinterpolator2D.derivativeX.
        '''
        x_pos, y_pos = self.findSector(x,y)
        alpha, beta = self.findCoords(x,y,x_pos,y_pos)
        
        # Get four corners data for each point
        xA = self.x_values[x_pos,y_pos]
        xB = self.x_values[x_pos+1,y_pos]
        xC = self.x_values[x_pos,y_pos+1]
        xD = self.x_values[x_pos+1,y_pos+1]
        yA = self.y_values[x_pos,y_pos]
        yB = self.y_values[x_pos+1,y_pos]
        yC = self.y_values[x_pos,y_pos+1]
        yD = self.y_values[x_pos+1,y_pos+1]
        fA = self.f_values[x_pos,y_pos]
        fB = self.f_values[x_pos+1,y_pos]
        fC = self.f_values[x_pos,y_pos+1]
        fD = self.f_values[x_pos+1,y_pos+1]
        
        # Calculate components of the alpha,beta --> x,y delta translation matrix
        alpha_x = (1-beta)*(xB-xA) + beta*(xD-xC)
        alpha_y = (1-beta)*(yB-yA) + beta*(yD-yC)
        beta_x  = (1-alpha)*(xC-xA) + alpha*(xD-xB)
        beta_y  = (1-alpha)*(yC-yA) + alpha*(yD-yB)
        
        # Invert the delta translation matrix into x,y --> alpha,beta
        det = alpha_x*beta_y - beta_x*alpha_y
        #x_alpha = beta_y/det
        #x_beta  = -alpha_y/det
        y_alpha = -beta_x/det
        y_beta  = alpha_x/det
        
        # Calculate the derivative of f w.r.t. alpha and beta
        dfda = (1-beta)*(fB-fA) + beta*(fD-fC)
        dfdb = (1-alpha)*(fC-fA) + alpha*(fD-fB)
        
        # Calculate the derivative with respect to x (and return it)
        dfdy = y_alpha*dfda + y_beta*dfdb
        return dfdy
        
        
if __name__ == '__main__':       
    print("Sorry, HARKinterpolation doesn't actually do much on its own.")
    print("To see some examples of its interpolation methods in action, look at any")
    print("of the model modules in /ConsumptionSavingModel.  In the future, running")
    print("this module will show examples of each interpolation class.")
    
    from time import clock
    import matplotlib.pyplot as plt
    
    RNG = np.random.RandomState(123)
    
    if False:
        x = np.linspace(1,20,39)
        y = np.log(x)
        dydx = 1.0/x
        f = CubicInterp(x,y,dydx)
        x_test = np.linspace(0,30,200)
        y_test = f(x_test)
        plt.plot(x_test,y_test)
        plt.show()
    
    if False:
        f = lambda x,y : 3.0*x**2.0 + x*y + 4.0*y**2.0
        dfdx = lambda x,y : 6.0*x + y
        dfdy = lambda x,y : x + 8.0*y
        
        y_list = np.linspace(0,5,100,dtype=float)
        xInterpolators = []
        xInterpolators_alt = []
        for y in y_list:
            this_x_list = np.sort((RNG.rand(100)*5.0))
            this_interpolation = LinearInterp(this_x_list,f(this_x_list,y*np.ones(this_x_list.size)))
            that_interpolation = CubicInterp(this_x_list,f(this_x_list,y*np.ones(this_x_list.size)),dfdx(this_x_list,y*np.ones(this_x_list.size)))
            xInterpolators.append(this_interpolation)
            xInterpolators_alt.append(that_interpolation)
        g = LinearInterpOnInterp1D(xInterpolators,y_list)
        h = LinearInterpOnInterp1D(xInterpolators_alt,y_list)
        
        rand_x = RNG.rand(100)*5.0
        rand_y = RNG.rand(100)*5.0
        z = (f(rand_x,rand_y) - g(rand_x,rand_y))/f(rand_x,rand_y)
        q = (dfdx(rand_x,rand_y) - g.derivativeX(rand_x,rand_y))/dfdx(rand_x,rand_y)
        r = (dfdy(rand_x,rand_y) - g.derivativeY(rand_x,rand_y))/dfdy(rand_x,rand_y)
        #print(z)
        #print(q)
        #print(r)
        
        z = (f(rand_x,rand_y) - g(rand_x,rand_y))/f(rand_x,rand_y)
        q = (dfdx(rand_x,rand_y) - g.derivativeX(rand_x,rand_y))/dfdx(rand_x,rand_y)
        r = (dfdy(rand_x,rand_y) - g.derivativeY(rand_x,rand_y))/dfdy(rand_x,rand_y)
        print(z)
        #print(q)
        #print(r)
    
    
    if False:
        f = lambda x,y,z : 3.0*x**2.0 + x*y + 4.0*y**2.0 - 5*z**2.0 + 1.5*x*z
        dfdx = lambda x,y,z : 6.0*x + y + 1.5*z
        dfdy = lambda x,y,z : x + 8.0*y
        dfdz = lambda x,y,z : -10.0*z + 1.5*x
        
        y_list = np.linspace(0,5,51,dtype=float)
        z_list = np.linspace(0,5,51,dtype=float)
        xInterpolators = []
        for y in y_list:
            temp = []
            for z in z_list:
                this_x_list = np.sort((RNG.rand(100)*5.0))
                this_interpolation = LinearInterp(this_x_list,f(this_x_list,y*np.ones(this_x_list.size),z*np.ones(this_x_list.size)))
                temp.append(this_interpolation)
            xInterpolators.append(deepcopy(temp))
        g = BilinearInterpOnInterp1D(xInterpolators,y_list,z_list)
        
        rand_x = RNG.rand(1000)*5.0
        rand_y = RNG.rand(1000)*5.0
        rand_z = RNG.rand(1000)*5.0
        z = (f(rand_x,rand_y,rand_z) - g(rand_x,rand_y,rand_z))/f(rand_x,rand_y,rand_z)
        q = (dfdx(rand_x,rand_y,rand_z) - g.derivativeX(rand_x,rand_y,rand_z))/dfdx(rand_x,rand_y,rand_z)
        r = (dfdy(rand_x,rand_y,rand_z) - g.derivativeY(rand_x,rand_y,rand_z))/dfdy(rand_x,rand_y,rand_z)
        p = (dfdz(rand_x,rand_y,rand_z) - g.derivativeZ(rand_x,rand_y,rand_z))/dfdz(rand_x,rand_y,rand_z)
        z.sort()
    
    
    
    if False:
        f = lambda w,x,y,z : 4.0*w*z - 2.5*w*x + w*y + 6.0*x*y - 10.0*x*z + 3.0*y*z - 7.0*z + 4.0*x + 2.0*y - 5.0*w
        dfdw = lambda w,x,y,z : 4.0*z - 2.5*x + y - 5.0
        dfdx = lambda w,x,y,z : -2.5*w + 6.0*y - 10.0*z + 4.0
        dfdy = lambda w,x,y,z : w + 6.0*x + 3.0*z + 2.0
        dfdz = lambda w,x,y,z : 4.0*w - 10.0*x + 3.0*y - 7
        
        x_list = np.linspace(0,5,16,dtype=float)
        y_list = np.linspace(0,5,16,dtype=float)
        z_list = np.linspace(0,5,16,dtype=float)
        wInterpolators = []
        for x in x_list:
            temp = []
            for y in y_list:
                temptemp = []
                for z in z_list:
                    this_w_list = np.sort((RNG.rand(16)*5.0))
                    this_interpolation = LinearInterp(this_w_list,f(this_w_list,x*np.ones(this_w_list.size),y*np.ones(this_w_list.size),z*np.ones(this_w_list.size)))
                    temptemp.append(this_interpolation)
                temp.append(deepcopy(temptemp))
            wInterpolators.append(deepcopy(temp))
        g = TrilinearInterpOnInterp1D(wInterpolators,x_list,y_list,z_list)
        
        N = 20000
        rand_w = RNG.rand(N)*5.0
        rand_x = RNG.rand(N)*5.0
        rand_y = RNG.rand(N)*5.0
        rand_z = RNG.rand(N)*5.0
        t_start = clock()
        z = (f(rand_w,rand_x,rand_y,rand_z) - g(rand_w,rand_x,rand_y,rand_z))/f(rand_w,rand_x,rand_y,rand_z)
        q = (dfdw(rand_w,rand_x,rand_y,rand_z) - g.derivativeW(rand_w,rand_x,rand_y,rand_z))/dfdw(rand_w,rand_x,rand_y,rand_z)
        r = (dfdx(rand_w,rand_x,rand_y,rand_z) - g.derivativeX(rand_w,rand_x,rand_y,rand_z))/dfdx(rand_w,rand_x,rand_y,rand_z)
        p = (dfdy(rand_w,rand_x,rand_y,rand_z) - g.derivativeY(rand_w,rand_x,rand_y,rand_z))/dfdy(rand_w,rand_x,rand_y,rand_z)
        s = (dfdz(rand_w,rand_x,rand_y,rand_z) - g.derivativeZ(rand_w,rand_x,rand_y,rand_z))/dfdz(rand_w,rand_x,rand_y,rand_z)
        t_end = clock()
        
        z.sort()
        print(z)
        print(t_end-t_start)
    
    if False:
        f = lambda x,y : 3.0*x**2.0 + x*y + 4.0*y**2.0
        dfdx = lambda x,y : 6.0*x + y
        dfdy = lambda x,y : x + 8.0*y
        
        x_list = np.linspace(0,5,101,dtype=float)
        y_list = np.linspace(0,5,101,dtype=float)
        x_temp,y_temp = np.meshgrid(x_list,y_list,indexing='ij')
        g = BilinearInterp(f(x_temp,y_temp),x_list,y_list)
        
        rand_x = RNG.rand(100)*5.0
        rand_y = RNG.rand(100)*5.0
        z = (f(rand_x,rand_y) - g(rand_x,rand_y))/f(rand_x,rand_y)
        q = (f(x_temp,y_temp) - g(x_temp,y_temp))/f(x_temp,y_temp)
        #print(z)
        #print(q)
        
        
    if False:
        f = lambda x,y,z : 3.0*x**2.0 + x*y + 4.0*y**2.0 - 5*z**2.0 + 1.5*x*z
        dfdx = lambda x,y,z : 6.0*x + y + 1.5*z
        dfdy = lambda x,y,z : x + 8.0*y
        dfdz = lambda x,y,z : -10.0*z + 1.5*x
        
        x_list = np.linspace(0,5,11,dtype=float)
        y_list = np.linspace(0,5,11,dtype=float)
        z_list = np.linspace(0,5,101,dtype=float)
        x_temp,y_temp,z_temp = np.meshgrid(x_list,y_list,z_list,indexing='ij')
        g = TrilinearInterp(f(x_temp,y_temp,z_temp),x_list,y_list,z_list)
        
        rand_x = RNG.rand(1000)*5.0
        rand_y = RNG.rand(1000)*5.0
        rand_z = RNG.rand(1000)*5.0
        z = (f(rand_x,rand_y,rand_z) - g(rand_x,rand_y,rand_z))/f(rand_x,rand_y,rand_z)
        q = (dfdx(rand_x,rand_y,rand_z) - g.derivativeX(rand_x,rand_y,rand_z))/dfdx(rand_x,rand_y,rand_z)
        r = (dfdy(rand_x,rand_y,rand_z) - g.derivativeY(rand_x,rand_y,rand_z))/dfdy(rand_x,rand_y,rand_z)
        p = (dfdz(rand_x,rand_y,rand_z) - g.derivativeZ(rand_x,rand_y,rand_z))/dfdz(rand_x,rand_y,rand_z)
        p.sort()    
        plt.plot(p)
        
        
    if False:
        f = lambda w,x,y,z : 4.0*w*z - 2.5*w*x + w*y + 6.0*x*y - 10.0*x*z + 3.0*y*z - 7.0*z + 4.0*x + 2.0*y - 5.0*w
        dfdw = lambda w,x,y,z : 4.0*z - 2.5*x + y - 5.0
        dfdx = lambda w,x,y,z : -2.5*w + 6.0*y - 10.0*z + 4.0
        dfdy = lambda w,x,y,z : w + 6.0*x + 3.0*z + 2.0
        dfdz = lambda w,x,y,z : 4.0*w - 10.0*x + 3.0*y - 7
        
        w_list = np.linspace(0,5,16,dtype=float)
        x_list = np.linspace(0,5,16,dtype=float)
        y_list = np.linspace(0,5,16,dtype=float)
        z_list = np.linspace(0,5,16,dtype=float)
        w_temp,x_temp,y_temp,z_temp = np.meshgrid(w_list,x_list,y_list,z_list,indexing='ij')
        mySearch = lambda trash,x : np.floor(x/5*32).astype(int)
        g = QuadlinearInterp(f(w_temp,x_temp,y_temp,z_temp),w_list,x_list,y_list,z_list)
        
        N = 1000000
        rand_w = RNG.rand(N)*5.0
        rand_x = RNG.rand(N)*5.0
        rand_y = RNG.rand(N)*5.0
        rand_z = RNG.rand(N)*5.0
        t_start = clock()
        z = (f(rand_w,rand_x,rand_y,rand_z) - g(rand_w,rand_x,rand_y,rand_z))/f(rand_w,rand_x,rand_y,rand_z)
        t_end = clock()
        #print(z)
        print(t_end-t_start)
    
    
    if False:
        f = lambda x,y : 3.0*x**2.0 + x*y + 4.0*y**2.0
        dfdx = lambda x,y : 6.0*x + y
        dfdy = lambda x,y : x + 8.0*y
        
        warp_factor = 0.01
        x_list = np.linspace(0,5,71,dtype=float)
        y_list = np.linspace(0,5,51,dtype=float)
        x_temp,y_temp = np.meshgrid(x_list,y_list,indexing='ij')
        x_adj = x_temp + warp_factor*(RNG.rand(x_list.size,y_list.size) - 0.5)
        y_adj = y_temp + warp_factor*(RNG.rand(x_list.size,y_list.size) - 0.5)
        g = Curvilinear2DInterp(f(x_adj,y_adj),x_adj,y_adj)
        
        rand_x = RNG.rand(1000)*5.0
        rand_y = RNG.rand(1000)*5.0
        t_start = clock()
        z = (f(rand_x,rand_y) - g(rand_x,rand_y))/f(rand_x,rand_y)
        q = (dfdx(rand_x,rand_y) - g.derivativeX(rand_x,rand_y))/dfdx(rand_x,rand_y)
        r = (dfdy(rand_x,rand_y) - g.derivativeY(rand_x,rand_y))/dfdy(rand_x,rand_y)
        t_end = clock()
        z.sort()
        q.sort()
        r.sort()
        #print(z)
        print(t_end-t_start)
        
        
    if False:
        f = lambda x,y,z : 3.0*x**2.0 + x*y + 4.0*y**2.0 - 5*z**2.0 + 1.5*x*z
        dfdx = lambda x,y,z : 6.0*x + y + 1.5*z
        dfdy = lambda x,y,z : x + 8.0*y
        dfdz = lambda x,y,z : -10.0*z + 1.5*x
        
        warp_factor = 0.01
        x_list = np.linspace(0,5,11,dtype=float)
        y_list = np.linspace(0,5,11,dtype=float)
        z_list = np.linspace(0,5,101,dtype=float)
        x_temp,y_temp = np.meshgrid(x_list,y_list,indexing='ij')
        xyInterpolators = []
        for j in range(z_list.size):
            x_adj = x_temp + warp_factor*(RNG.rand(x_list.size,y_list.size) - 0.5)
            y_adj = y_temp + warp_factor*(RNG.rand(x_list.size,y_list.size) - 0.5)
            z_temp = z_list[j]*np.ones(x_adj.shape)
            thisInterp = Curvilinear2DInterp(f(x_adj,y_adj,z_temp),x_adj,y_adj)
            xyInterpolators.append(thisInterp)
        g = LinearInterpOnInterp2D(xyInterpolators,z_list)
        
        N = 1000
        rand_x = RNG.rand(N)*5.0
        rand_y = RNG.rand(N)*5.0
        rand_z = RNG.rand(N)*5.0
        z = (f(rand_x,rand_y,rand_z) - g(rand_x,rand_y,rand_z))/f(rand_x,rand_y,rand_z)
        p = (dfdz(rand_x,rand_y,rand_z) - g.derivativeZ(rand_x,rand_y,rand_z))/dfdz(rand_x,rand_y,rand_z)
        p.sort()
        plt.plot(p)
        
        
    if False:
        f = lambda w,x,y,z : 4.0*w*z - 2.5*w*x + w*y + 6.0*x*y - 10.0*x*z + 3.0*y*z - 7.0*z + 4.0*x + 2.0*y - 5.0*w
        dfdw = lambda w,x,y,z : 4.0*z - 2.5*x + y - 5.0
        dfdx = lambda w,x,y,z : -2.5*w + 6.0*y - 10.0*z + 4.0
        dfdy = lambda w,x,y,z : w + 6.0*x + 3.0*z + 2.0
        dfdz = lambda w,x,y,z : 4.0*w - 10.0*x + 3.0*y - 7
        
        warp_factor = 0.1
        w_list = np.linspace(0,5,16,dtype=float)
        x_list = np.linspace(0,5,16,dtype=float)
        y_list = np.linspace(0,5,16,dtype=float)
        z_list = np.linspace(0,5,16,dtype=float)
        w_temp,x_temp = np.meshgrid(w_list,x_list,indexing='ij')
        wxInterpolators = []
        for i in range(y_list.size):
            temp = []
            for j in range(z_list.size):
                w_adj = w_temp + warp_factor*(RNG.rand(w_list.size,x_list.size) - 0.5)
                x_adj = x_temp + warp_factor*(RNG.rand(w_list.size,x_list.size) - 0.5)
                y_temp = y_list[i]*np.ones(w_adj.shape)
                z_temp = z_list[j]*np.ones(w_adj.shape)
                thisInterp = Curvilinear2DInterp(f(w_adj,x_adj,y_temp,z_temp),w_adj,x_adj)
                temp.append(thisInterp)
            wxInterpolators.append(temp)
        g = BilinearInterpOnInterp2D(wxInterpolators,y_list,z_list)
        
        N = 1000000
        rand_w = RNG.rand(N)*5.0
        rand_x = RNG.rand(N)*5.0
        rand_y = RNG.rand(N)*5.0
        rand_z = RNG.rand(N)*5.0
        
        t_start = clock()
        z = (f(rand_w,rand_x,rand_y,rand_z) - g(rand_w,rand_x,rand_y,rand_z))/f(rand_w,rand_x,rand_y,rand_z)
        t_end = clock()
        z.sort()
        print(z)
        print(t_end-t_start)