Author: bugman
Date: Wed Oct 12 13:36:16 2011
New Revision: 14829
URL: http://svn.gna.org/viewcvs/relax?rev=14829&view=rev
Log:
Updated the N-state model analysis code to handle the per-alignment tensor
fixed flag.
This allows subsets of tensors to be optimised while the others are fixed.
Modified:
1.3/specific_fns/n_state_model.py
Modified: 1.3/specific_fns/n_state_model.py
URL:
http://svn.gna.org/viewcvs/relax/1.3/specific_fns/n_state_model.py?rev=14829&r1=14828&r2=14829&view=diff
==============================================================================
--- 1.3/specific_fns/n_state_model.py (original)
+++ 1.3/specific_fns/n_state_model.py Wed Oct 12 13:36:16 2011
@@ -38,6 +38,7 @@
import arg_check
from float import isNaN, isInf
import generic_fns
+from generic_fns.align_tensor import all_tensors_fixed
from generic_fns.mol_res_spin import return_spin, spin_loop
from generic_fns import pcs, pipes, rdc
from generic_fns.structure.cones import Iso_cone
@@ -89,11 +90,15 @@
param_vector = []
# A RDC or PCS data type requires the alignment tensors to be at the
start of the parameter vector (unless the tensors are fixed).
- if ('rdc' in data_types or 'pcs' in data_types) and not
(hasattr(cdp.align_tensors, 'fixed') and cdp.align_tensors.fixed):
+ if ('rdc' in data_types or 'pcs' in data_types) and not
all_tensors_fixed():
# Loop over the alignments, adding the alignment tensor
parameters to the parameter vector.
for i in xrange(len(cdp.align_tensors)):
# No alignment ID, so skip the tensor as it will not be
optimised.
if cdp.align_tensors[i].name not in cdp.align_ids:
+ continue
+
+ # Fixed tensor.
+ if cdp.align_tensors[i].fixed:
continue
# Add the parameters.
@@ -176,12 +181,23 @@
# Starting point of the populations.
pop_start = 0
- if ('rdc' in data_types or 'pcs' in data_types) and not
(hasattr(cdp.align_tensors, 'fixed') and cdp.align_tensors.fixed):
- pop_start = pop_start + 5*len(cdp.align_ids)
-
- # The alignment parameters.
- for i in range(5*len(cdp.align_ids)):
- scaling_matrix[i, i] = 1.0
+ if ('rdc' in data_types or 'pcs' in data_types) and not
all_tensors_fixed():
+ # Loop over the alignments.
+ tensor_num = 0
+ for i in range(len(cdp.align_tensors)):
+ # Fixed tensor.
+ if cdp.align_tensors[i].fixed:
+ continue
+
+ # Add the 5 alignment parameters.
+ pop_start = pop_start + 5
+
+ # The alignment parameters.
+ for j in range(5):
+ scaling_matrix[5*tensor_num + j, 5*tensor_num + j] = 1.0
+
+ # Increase the tensor number.
+ tensor_num += 1
# Loop over the populations, and set the scaling factor.
if cdp.model in ['2-domain', 'population']:
@@ -476,31 +492,39 @@
raise RelaxNoModelError
# Unpack and strip off the alignment tensor parameters.
- if ('rdc' in data_types or 'pcs' in data_types) and not
(hasattr(cdp.align_tensors, 'fixed') and cdp.align_tensors.fixed):
+ if ('rdc' in data_types or 'pcs' in data_types) and not
all_tensors_fixed():
# Loop over the alignments, adding the alignment tensor
parameters to the tensor data container.
+ tensor_num = 0
for i in xrange(len(cdp.align_tensors)):
# No alignment ID, so skip the tensor as it will not be
optimised.
if cdp.align_tensors[i].name not in cdp.align_ids:
continue
+ # Fixed tensor.
+ if cdp.align_tensors[i].fixed:
+ continue
+
# Normal tensors.
if sim_index == None:
- cdp.align_tensors[i].Axx = param_vector[5*i]
- cdp.align_tensors[i].Ayy = param_vector[5*i+1]
- cdp.align_tensors[i].Axy = param_vector[5*i+2]
- cdp.align_tensors[i].Axz = param_vector[5*i+3]
- cdp.align_tensors[i].Ayz = param_vector[5*i+4]
+ cdp.align_tensors[i].Axx = param_vector[5*tensor_num]
+ cdp.align_tensors[i].Ayy = param_vector[5*tensor_num+1]
+ cdp.align_tensors[i].Axy = param_vector[5*tensor_num+2]
+ cdp.align_tensors[i].Axz = param_vector[5*tensor_num+3]
+ cdp.align_tensors[i].Ayz = param_vector[5*tensor_num+4]
# Monte Carlo simulated tensors.
else:
- cdp.align_tensors[i].Axx_sim[sim_index] =
param_vector[5*i]
- cdp.align_tensors[i].Ayy_sim[sim_index] =
param_vector[5*i+1]
- cdp.align_tensors[i].Axy_sim[sim_index] =
param_vector[5*i+2]
- cdp.align_tensors[i].Axz_sim[sim_index] =
param_vector[5*i+3]
- cdp.align_tensors[i].Ayz_sim[sim_index] =
param_vector[5*i+4]
+ cdp.align_tensors[i].Axx_sim[sim_index] =
param_vector[5*tensor_num]
+ cdp.align_tensors[i].Ayy_sim[sim_index] =
param_vector[5*tensor_num+1]
+ cdp.align_tensors[i].Axy_sim[sim_index] =
param_vector[5*tensor_num+2]
+ cdp.align_tensors[i].Axz_sim[sim_index] =
param_vector[5*tensor_num+3]
+ cdp.align_tensors[i].Ayz_sim[sim_index] =
param_vector[5*tensor_num+4]
+
+ # Increase the tensor number.
+ tensor_num += 1
# Create a new parameter vector without the tensors.
- param_vector = param_vector[5*len(cdp.align_ids):]
+ param_vector = param_vector[5*tensor_num:]
# Monte Carlo simulation data structures.
if sim_index != None:
@@ -673,8 +697,15 @@
# Starting point of the populations.
pop_start = 0
- if ('rdc' in data_types or 'pcs' in data_types) and not
(hasattr(cdp.align_tensors, 'fixed') and cdp.align_tensors.fixed):
- pop_start = pop_start + 5*len(cdp.align_ids)
+ if ('rdc' in data_types or 'pcs' in data_types) and not
all_tensors_fixed():
+ # Loop over the alignments.
+ for i in range(len(cdp.align_tensors)):
+ # Fixed tensor.
+ if cdp.align_tensors[i].fixed:
+ continue
+
+ # Add 5 parameters.
+ pop_start += 5
# Initialisation (0..j..m).
A = []
@@ -1316,15 +1347,19 @@
num = 0
# Alignment tensor params.
- if ('rdc' in data_types or 'pcs' in data_types) and not
(hasattr(cdp.align_tensors, 'fixed') and cdp.align_tensors.fixed):
+ if ('rdc' in data_types or 'pcs' in data_types) and not
all_tensors_fixed():
# Loop over the alignments.
for i in xrange(len(cdp.align_tensors)):
# No alignment ID, so skip the tensor as it is not part of
the parameter set.
if cdp.align_tensors[i].name not in cdp.align_ids:
continue
+ # Fixed tensor.
+ if cdp.align_tensors[i].fixed:
+ continue
+
# Add 5 tensor parameters.
- num = num + 5
+ num += 5
# Populations.
if cdp.model in ['2-domain', 'population']:
@@ -1462,7 +1497,7 @@
rdcs, rdc_err, rdc_weight, xh_vect, rdc_dj =
self._minimise_setup_rdcs(sim_index=sim_index)
# Get the fixed tensors.
- if ('rdc' in data_types or 'pcs' in data_types) and
(hasattr(cdp.align_tensors, 'fixed') and cdp.align_tensors.fixed):
+ if ('rdc' in data_types or 'pcs' in data_types) and
all_tensors_fixed():
full_tensors =
self._minimise_setup_fixed_tensors(sim_index=sim_index)
# Get the atomic_positions.
r14829 - /1.3/specific_fns/n_state_model.py