Author: bugman
Date: Fri Aug 28 18:01:25 2009
New Revision: 9412
URL: http://svn.gna.org/viewcvs/relax?rev=9412&view=rev
Log:
Created the 'rigid' frame order model.
Modified:
1.3/maths_fns/frame_order_models.py
1.3/specific_fns/frame_order.py
Modified: 1.3/maths_fns/frame_order_models.py
URL:
http://svn.gna.org/viewcvs/relax/1.3/maths_fns/frame_order_models.py?rev=9412&r1=9411&r2=9412&view=diff
==============================================================================
--- 1.3/maths_fns/frame_order_models.py (original)
+++ 1.3/maths_fns/frame_order_models.py Fri Aug 28 18:01:25 2009
@@ -73,23 +73,24 @@
# Store the initial parameter (as a copy).
self.params = deepcopy(init_params)
+ # The rigid model.
+ if model == 'rigid':
+ # Tensor optimisation.
+ self.__init_tensors(full_tensors, red_tensors, red_errors,
full_in_ref_frame)
+
+ # Alias the target function.
+ self.func = self.func_rigid
+
+
# Isotropic cone model.
if model == 'iso cone':
- # Some checks.
- if red_tensors == None or not len(red_tensors):
- raise RelaxError("The red_tensors argument " +
repr(red_tensors) + " must be supplied.")
- if red_errors == None or not len(red_errors):
- raise RelaxError("The red_errors argument " +
repr(red_errors) + " must be supplied.")
- if full_in_ref_frame == None or not len(full_in_ref_frame):
- raise RelaxError("The full_in_ref_frame argument " +
repr(full_in_ref_frame) + " must be supplied.")
-
# Mix up.
if full_tensors != None and frame_order_2nd != None:
raise RelaxError("Tensors and Frame Order matrices cannot be
supplied together.")
- # Tensor optimisation.
- elif full_tensors != None:
- self.__init_iso_cone(full_tensors, red_tensors, red_errors,
full_in_ref_frame)
+ # Tensor setup.
+ if full_tensors != None:
+ self.__init_tensors(full_tensors, red_tensors, red_errors,
full_in_ref_frame)
# Optimisation to the 2nd degree Frame Order matrix components
directly.
elif frame_order_2nd != None:
@@ -99,8 +100,15 @@
else:
raise RelaxError("The arguments have been incorrectly
supplied, cannot determine the optimisation mode.")
-
- def __init_iso_cone(self, full_tensors, red_tensors, red_errors,
full_in_ref_frame):
+ # The cone axis storage and molecular frame z-axis.
+ self.cone_axis = zeros(3, float64)
+ self.z_axis = array([0, 0, 1], float64)
+
+ # Alias the target function.
+ self.func = self.func_iso_cone
+
+
+ def __init_tensors(self, full_tensors, red_tensors, red_errors,
full_in_ref_frame):
"""Set up isotropic cone optimisation against the alignment tensor
data.
@keyword full_tensors: An array of the {Sxx, Syy, Sxy, Sxz,
Syz} values for all full
@@ -116,10 +124,6 @@
@type red_errors: numpy nx5D, rank-1 float64 array
"""
- # Checks.
- if red_tensors == None:
- raise RelaxError("The reduced tensors have not been supplied.")
-
# Some checks.
if full_tensors == None or not len(full_tensors):
raise RelaxError("The full_tensors argument " +
repr(full_tensors) + " must be supplied.")
@@ -138,19 +142,12 @@
self.red_tensors_bc = zeros(self.num_tensors*5, float64)
self.full_in_ref_frame = full_in_ref_frame
- # The cone axis storage and molecular frame z-axis.
- self.cone_axis = zeros(3, float64)
- self.z_axis = array([0, 0, 1], float64)
-
# The rotation to the Frame Order eigenframe.
self.rot = zeros((3, 3), float64)
self.tensor_3D = zeros((3, 3), float64)
# Initialise the Frame Order matrices.
self.frame_order_2nd = zeros((9, 9), float64)
-
- # Alias the target function.
- self.func = self.func_iso_cone
def __init_iso_cone_elements(self, frame_order_2nd):
@@ -182,11 +179,106 @@
self.func = self.func_iso_cone_elements
+ def func_rigid(self, params):
+ """Target function for rigid model optimisation using the alignment
tensors.
+
+ This function optimises against alignment tensors. The Euler angles
for the tensor rotation
+ are the 3 parameters optimised in this model.
+
+ @param params: The vector of parameter values. These are the
tensor rotation angles
+ {alpha, beta, gamma}.
+ @type params: list of float
+ @return: The chi-squared or SSE value.
+ @rtype: float
+ """
+
+ # Unpack the parameters.
+ alpha, beta, gamma = params
+
+ # Alignment tensor rotation.
+ R_euler_zyz(self.rot, alpha, beta, gamma)
+
+ # Back calculate the rotated tensors.
+ for i in range(self.num_tensors):
+ # Tensor indices.
+ index1 = i*5
+ index2 = i*5+5
+
+ # Convert the original tensor to 3D, rank-2 form.
+ to_tensor(self.tensor_3D, self.full_tensors[index1:index2])
+
+ # Rotate the tensor (normal R.X.RT rotation).
+ if self.full_in_ref_frame[i]:
+ self.tensor_3D = dot(self.rot, dot(self.tensor_3D,
transpose(self.rot)))
+
+ # Rotate the tensor (inverse RT.X.R rotation).
+ else:
+ self.tensor_3D = dot(transpose(self.rot),
dot(self.tensor_3D, self.rot))
+
+ # Convert the tensor back to 5D, rank-1 form, as the
back-calculated reduced tensor.
+ to_5D(self.red_tensors_bc[index1:index2], self.tensor_3D)
+
+ # Return the chi-squared value.
+ return chi2(self.red_tensors, self.red_tensors_bc, self.red_errors)
+
+
def func_iso_cone(self, params):
"""Target function for isotropic cone model optimisation using the
alignment tensors.
This function optimises against alignment tensors. The cone axis
spherical angles theta and
phi and the cone angle theta are the 3 parameters optimised in this
model.
+
+ @param params: The vector of parameter values {alpha, beta, gamma,
theta, phi, theta_cone}
+ where the first 3 are the tensor rotation Euler
angles, the next two are the
+ polar and azimuthal angles of the cone axis
theta_cone is the isotropic cone
+ angle.
+ @type params: list of float
+ @return: The chi-squared or SSE value.
+ @rtype: float
+ """
+
+ # Unpack the parameters.
+ alpha, beta, gamma, theta, phi, theta_cone = params
+
+ # Generate the 2nd degree Frame Order super matrix.
+ self.frame_order_2nd =
compile_2nd_matrix_iso_cone(self.frame_order_2nd, self.rot, self.z_axis,
self.cone_axis, theta, phi, theta_cone)
+
+ # Reduced alignment tensor rotation.
+ R_euler_zyz(self.rot, alpha, beta, gamma)
+
+ # Back calculate the reduced tensors.
+ for i in range(self.num_tensors):
+ # Tensor indices.
+ index1 = i*5
+ index2 = i*5+5
+
+ # Reduce the tensor.
+ reduce_alignment_tensor(self.frame_order_2nd,
self.full_tensors[index1:index2], self.red_tensors_bc[index1:index2])
+
+ # Convert the tensor to 3D, rank-2 form.
+ to_tensor(self.tensor_3D, self.red_tensors_bc[index1:index2])
+
+ # Rotate the tensor (normal R.X.RT rotation).
+ if self.full_in_ref_frame[i]:
+ self.tensor_3D = dot(self.rot, dot(self.tensor_3D,
transpose(self.rot)))
+
+ # Rotate the tensor (inverse RT.X.R rotation).
+ else:
+ self.tensor_3D = dot(transpose(self.rot),
dot(self.tensor_3D, self.rot))
+
+ # Convert the tensor back to 5D, rank-1 form.
+ to_5D(self.red_tensors_bc[index1:index2], self.tensor_3D)
+
+ # Return the chi-squared value.
+ return chi2(self.red_tensors, self.red_tensors_bc, self.red_errors)
+
+
+ def func_iso_cone_elements(self, params):
+ """Target function for isotropic cone model optimisation using the
Frame Order matrix.
+
+ This function optimises by directly matching the elements of the 2nd
degree Frame Order
+ super matrix. The cone axis spherical angles theta and phi and the
cone angle theta are the
+ 3 parameters optimised in this model.
@param params: The vector of parameter values {theta, phi,
theta_cone} where the first two
are the polar and azimuthal angles of the cone axis
theta_cone is the
@@ -197,57 +289,6 @@
"""
# Break up the parameters.
- alpha, beta, gamma, theta, phi, theta_cone = params
-
- # Generate the 2nd degree Frame Order super matrix.
- self.frame_order_2nd =
compile_2nd_matrix_iso_cone(self.frame_order_2nd, self.rot, self.z_axis,
self.cone_axis, theta, phi, theta_cone)
-
- # Reduced alignment tensor rotation into the eigenframe.
- R_euler_zyz(self.rot, alpha, beta, gamma)
-
- # Back calculate the reduced tensors.
- for i in range(self.num_tensors):
- # Tensor indices.
- index1 = i*5
- index2 = i*5+5
-
- # Reduce the tensor.
- reduce_alignment_tensor(self.frame_order_2nd,
self.full_tensors[index1:index2], self.red_tensors_bc[index1:index2])
-
- # Convert the tensor to 3D, rank-2 form.
- to_tensor(self.tensor_3D, self.red_tensors_bc[index1:index2])
-
- # Rotate the tensor (normal R.X.RT rotation).
- if self.full_in_ref_frame[i]:
- self.tensor_3D = dot(self.rot, dot(self.tensor_3D,
transpose(self.rot)))
-
- # Rotate the tensor (inverse RT.X.R rotation).
- else:
- self.tensor_3D = dot(transpose(self.rot),
dot(self.tensor_3D, self.rot))
-
- # Convert the tensor back to 5D, rank-1 form.
- to_5D(self.red_tensors_bc[index1:index2], self.tensor_3D)
-
- # Return the chi-squared value.
- return chi2(self.red_tensors, self.red_tensors_bc, self.red_errors)
-
-
- def func_iso_cone_elements(self, params):
- """Target function for isotropic cone model optimisation using the
Frame Order matrix.
-
- This function optimises by directly matching the elements of the 2nd
degree Frame Order
- super matrix. The cone axis spherical angles theta and phi and the
cone angle theta are the
- 3 parameters optimised in this model.
-
- @param params: The vector of parameter values {theta, phi,
theta_cone} where the first two
- are the polar and azimuthal angles of the cone axis
theta_cone is the
- isotropic cone angle.
- @type params: list of float
- @return: The chi-squared or SSE value.
- @rtype: float
- """
-
- # Break up the parameters.
theta, phi, theta_cone = params
# Generate the 2nd degree Frame Order super matrix.
Modified: 1.3/specific_fns/frame_order.py
URL:
http://svn.gna.org/viewcvs/relax/1.3/specific_fns/frame_order.py?rev=9412&r1=9411&r2=9412&view=diff
==============================================================================
--- 1.3/specific_fns/frame_order.py (original)
+++ 1.3/specific_fns/frame_order.py Fri Aug 28 18:01:25 2009
@@ -158,6 +158,23 @@
if not hasattr(cdp, 'params'):
cdp.params = []
+ # The rigid model.
+ if cdp.model == 'rigid':
+ # Set up the parameter arrays.
+ if not len(cdp.params):
+ cdp.params.append('alpha')
+ cdp.params.append('beta')
+ cdp.params.append('gamma')
+
+ # Initialise the tensor rotation angles.
+ if not hasattr(cdp, 'alpha'):
+ cdp.alpha = 0.0
+ if not hasattr(cdp, 'beta'):
+ cdp.beta = 0.0
+ if not hasattr(cdp, 'gamma'):
+ cdp.gamma = 0.0
+
+
# Isotropic cone model.
if cdp.model == 'iso cone':
# Set up the parameter arrays.
@@ -208,6 +225,30 @@
if isNaN(func):
raise RelaxNaNError('chi-squared')
+ # The rigid model.
+ if cdp.model == 'rigid':
+ # Disassemble the parameter vector.
+ alpha, beta, gamma = param_vector
+
+ # Wrap the Euler angles.
+ alpha = wrap_angles(alpha, 0.0, 2.0*pi)
+ beta = wrap_angles(beta, 0.0, 2.0*pi)
+ gamma = wrap_angles(gamma, 0.0, 2.0*pi)
+
+ # Monte Carlo simulation data structures.
+ if sim_index != None:
+ # Model parameters.
+ cdp.alpha_sim[sim_index] = alpha
+ cdp.beta_sim[sim_index] = beta
+ cdp.gamma_sim[sim_index] = gamma
+
+ # Normal data structures.
+ else:
+ # Model parameters.
+ cdp.alpha = alpha
+ cdp.beta = beta
+ cdp.gamma = gamma
+
# Isotropic cone model.
if cdp.model == 'iso cone':
# Disassemble the parameter vector.
@@ -234,14 +275,6 @@
cdp.phi_axis_sim[sim_index] = phi_axis
cdp.theta_cone_sim[sim_index] = theta_cone
- # Optimisation info.
- cdp.chi2_sim[sim_index] = func
- cdp.iter_sim[sim_index] = iter_count
- cdp.f_count_sim[sim_index] = f_count
- cdp.g_count_sim[sim_index] = g_count
- cdp.h_count_sim[sim_index] = h_count
- cdp.warning_sim[sim_index] = warning
-
# Normal data structures.
else:
# Model parameters.
@@ -252,13 +285,24 @@
cdp.phi_axis = phi_axis
cdp.theta_cone = theta_cone
- # Optimisation info.
- cdp.chi2 = func
- cdp.iter = iter_count
- cdp.f_count = f_count
- cdp.g_count = g_count
- cdp.h_count = h_count
- cdp.warning = warning
+ # Monte Carlo simulation optimisation data structures.
+ if sim_index != None:
+ cdp.chi2_sim[sim_index] = func
+ cdp.iter_sim[sim_index] = iter_count
+ cdp.f_count_sim[sim_index] = f_count
+ cdp.g_count_sim[sim_index] = g_count
+ cdp.h_count_sim[sim_index] = h_count
+ cdp.warning_sim[sim_index] = warning
+
+ # Normal optimisation data structures.
+ else:
+ cdp.chi2 = func
+ cdp.iter = iter_count
+ cdp.f_count = f_count
+ cdp.g_count = g_count
+ cdp.h_count = h_count
+ cdp.warning = warning
+
def back_calc(self):
@@ -630,16 +674,21 @@
if constraints:
raise RelaxError("Constraints are as of yet not implemented.")
+ # The rigid model.
+ if cdp.model == 'rigid':
+ # The initial parameter vector (tensor rotation Euler angles).
+ param_vector = array([cdp.alpha, cdp.beta, cdp.gamma], float64)
+
# Isotropic cone model.
if cdp.model == 'iso cone':
- # The initial parameter vector (the cone axis angles and the
cone angle).
+ # The initial parameter vector (tensor rotation Euler angles,
the cone axis angles and the cone angle).
param_vector = array([cdp.alpha, cdp.beta, cdp.gamma,
cdp.theta_axis, cdp.phi_axis, cdp.theta_cone], float64)
- # Get the data structures for optimisation using the tensors as
base data sets.
- full_tensors, red_tensors, red_tensor_err, full_in_ref_frame =
self.__minimise_setup_tensors(sim_index)
-
- # Set up the optimisation function.
- target = frame_order_models.Frame_order(model=cdp.model,
full_tensors=full_tensors, red_tensors=red_tensors,
red_errors=red_tensor_err, full_in_ref_frame=full_in_ref_frame)
+ # Get the data structures for optimisation using the tensors as base
data sets.
+ full_tensors, red_tensors, red_tensor_err, full_in_ref_frame =
self.__minimise_setup_tensors(sim_index)
+
+ # Set up the optimisation function.
+ target = frame_order_models.Frame_order(model=cdp.model,
full_tensors=full_tensors, red_tensors=red_tensors,
red_errors=red_tensor_err, full_in_ref_frame=full_in_ref_frame)
# Minimisation.
results = generic_minimise(func=target.func, args=(),
x0=param_vector, min_algor=min_algor, min_options=min_options,
func_tol=func_tol, grad_tol=grad_tol, maxiter=max_iterations, full_output=1,
print_flag=verbosity)
@@ -751,7 +800,7 @@
raise RelaxModelError('Frame Order')
# Test if the model name exists.
- if not model in ['iso cone']:
+ if not model in ['rigid', 'iso cone']:
raise RelaxError("The model name " + repr(model) + " is
invalid.")
# Set the model
r9412 - in /1.3: maths_fns/frame_order_models.py specific_fns/frame_order.py