def test_unit_box_map():
X = np.random.rand(5, 3)
lb = -1 * np.ones((3, ))
ub = 2 * np.ones((3, ))
X1 = to_unit_box(X, lb, ub)
np.testing.assert_equal(X.shape, X1.shape)
np.testing.assert_almost_equal(
X1, to_unit_box(X, np.atleast_2d(lb), np.atleast_2d(ub)))
assert (X.max() <= 1.0 and X.min() >= 0)
# Try to map back to what we started with
X2 = from_unit_box(X1, lb, ub)
np.testing.assert_equal(X.shape, X2.shape)
np.testing.assert_almost_equal(
X2, from_unit_box(X1, np.atleast_2d(lb), np.atleast_2d(ub)))
np.testing.assert_almost_equal(X2, X)
def test_unit_box_map():
X = np.random.rand(5, 3)
lb = -1 * np.ones((3,))
ub = 2 * np.ones((3,))
X1 = to_unit_box(X, lb, ub)
np.testing.assert_equal(X.shape, X1.shape)
np.testing.assert_almost_equal(
X1, to_unit_box(X, np.atleast_2d(lb), np.atleast_2d(ub)))
assert(X.max() <= 1.0 and X.min() >= 0)
# Try to map back to what we started with
X2 = from_unit_box(X1, lb, ub)
np.testing.assert_equal(X.shape, X2.shape)
np.testing.assert_almost_equal(
X2, from_unit_box(X1, np.atleast_2d(lb), np.atleast_2d(ub)))
np.testing.assert_almost_equal(X2, X)
def predict_std(self, x):
"""Predict standard deviation at points xx.
:param xx: Prediction points, must be of size num_pts x dim or (dim, )
:type xx: numpy.ndarray
:return: Predicted standard deviation, of size num_pts x 1
:rtype: numpy.ndarray
"""
return self.model.predict_std(to_unit_box(x, self.lb, self.ub))
def predict(self, xx):
"""Evaluate the surrogate model at the points xx
:param xx: Prediction points, must be of size num_pts x dim or (dim, )
:type xx: numpy.ndarray
:return: Prediction of size num_pts x 1
:rtype: numpy.ndarray
"""
return self.model.predict(to_unit_box(xx, self.lb, self.ub))
def predict_std(self, x):
"""Predict standard deviation at points xx.
:param xx: Prediction points, must be of size num_pts x dim or (dim, )
:type xx: numpy.ndarray
:return: Predicted standard deviation, of size num_pts x 1
:rtype: numpy.ndarray
"""
return self.model.predict_std(
to_unit_box(x, self.lb, self.ub))
def predict(self, xx):
"""Evaluate the surrogate model at the points xx
:param xx: Prediction points, must be of size num_pts x dim or (dim, )
:type xx: numpy.ndarray
:return: Prediction of size num_pts x 1
:rtype: numpy.ndarray
"""
return self.model.predict(
to_unit_box(xx, self.lb, self.ub))
def deriv(self, x, ds=None):
"""Evaluate the derivative of the rbf interpolant at x
:param x: Point for which we want to compute the MARS gradient
:type x: numpy.array
:param ds: Not used
:type ds: None
:return: Derivative of the MARS interpolant at x
:rtype: numpy.array
"""
return self.model.deriv(to_unit_box(x, self.data), ds)
def evals(self, x, ds=None):
"""Evaluate the capped rbf interpolant at the points xx
:param x: Points where to evaluate, of size npts x dim
:type x: numpy.array
:param ds: Not used
:type ds: None
:return: Values of the MARS interpolant at x, of length npts
:rtype: numpy.array
"""
return self.model.evals(to_unit_box(x, self.data), ds)
def eval(self, x, ds=None):
"""Evaluate the response surface at the point xx
:param x: Point where to evaluate
:type x: numpy.array
:param ds: Not used
:type ds: None
:return: Value of the interpolant at x
:rtype: float
"""
return self.model.eval(to_unit_box(x, self.data), ds)
def add_points(self, xx, fx):
"""Add new function evaluations.
This method SHOULD NOT trigger a new fit, it just updates X and
fX but leaves the original surrogate object intact
:param xx: Points to add
:type xx: numpy.ndarray
:param fx: The function values of the point to add
:type fx: numpy.array or float
"""
super().add_points(xx, fx)
self.model.add_points(to_unit_box(xx, self.lb, self.ub), fx)
def predict_deriv(self, x):
"""Evaluate the derivative of the surrogate model at points xx
Remember the chain rule:
f'(x) = (d/dx) g((x-a)/(b-a)) = g'((x-a)/(b-a)) * 1/(b-a)
:param xx: Prediction points, must be of size num_pts x dim or (dim, )
:type xx: numpy.array
:return: Derivative of the RBF interpolant at xx
:rtype: numpy.array
"""
return self.model.predict_deriv(to_unit_box(
x, self.lb, self.ub)) / (self.ub - self.lb)
def predict_deriv(self, x):
"""Evaluate the derivative of the surrogate model at points xx
Remember the chain rule:
f'(x) = (d/dx) g((x-a)/(b-a)) = g'((x-a)/(b-a)) * 1/(b-a)
:param xx: Prediction points, must be of size num_pts x dim or (dim, )
:type xx: numpy.array
:return: Derivative of the RBF interpolant at xx
:rtype: numpy.array
"""
return self.model.predict_deriv(
to_unit_box(x, self.lb, self.ub)) / (self.ub - self.lb)
def add_points(self, xx, fx):
"""Add new function evaluations.
This method SHOULD NOT trigger a new fit, it just updates X and
fX but leaves the original surrogate object intact
:param xx: Points to add
:type xx: numpy.ndarray
:param fx: The function values of the point to add
:type fx: numpy.array or float
"""
super().add_points(xx, fx)
self.model.add_points(
to_unit_box(xx, self.lb, self.ub), fx)
def add_point(self, xx, fx):
"""Add a new function evaluation
:param xx: Point to add
:type xx: numpy.array
:param fx: The function value of the point to add
:type fx: float
"""
if self.nump >= self.fvalues.shape[0]:
self.fvalues.resize(2 * self.fvalues.shape[0], 1)
self.fvalues[self.nump] = fx
self.nump += 1
self.updated = False
self.model.add_point(to_unit_box(xx, self.data), fx)
