Author: bugman
Date: Tue Sep 22 14:29:24 2009
New Revision: 9556
URL: http://svn.gna.org/viewcvs/relax?rev=9556&view=rev
Log:
Redesigned the relaxation curve fitting grid search for the new design of the
minfx grid search.
Modified:
1.3/specific_fns/relax_fit.py
Modified: 1.3/specific_fns/relax_fit.py
URL:
http://svn.gna.org/viewcvs/relax/1.3/specific_fns/relax_fit.py?rev=9556&r1=9555&r2=9556&view=diff
==============================================================================
--- 1.3/specific_fns/relax_fit.py (original)
+++ 1.3/specific_fns/relax_fit.py Tue Sep 22 14:29:24 2009
@@ -24,6 +24,8 @@
"""The relaxation curve fitting specific code."""
# Python module imports.
+from minfx.generic import generic_minimise
+from minfx.grid import grid
from numpy import array, average, dot, float64, identity, zeros
from numpy.linalg import inv
from re import match, search
@@ -33,7 +35,6 @@
from base_class import Common_functions
from generic_fns import pipes
from generic_fns.mol_res_spin import exists_mol_res_spin_data,
generate_spin_id, return_spin, spin_loop
-from minfx.generic import generic_minimise
from relax_errors import RelaxError, RelaxFuncSetupError, RelaxLenError,
RelaxNoModelError, RelaxNoSequenceError
# C modules.
@@ -440,70 +441,49 @@
raise RelaxError("Cannot run a grid search on a model with zero
parameters.")
# Lower bounds.
- if lower != None:
- if len(lower) != n:
- raise RelaxLenError('lower bounds', n)
+ if lower != None and len(lower) != n:
+ raise RelaxLenError('lower bounds', n)
# Upper bounds.
- if upper != None:
- if len(upper) != n:
- raise RelaxLenError('upper bounds', n)
-
- # Increment.
- if isinstance(inc, list):
- if len(inc) != n:
- raise RelaxLenError('increment', n)
- inc = inc
+ if upper != None and len(upper) != n:
+ raise RelaxLenError('upper bounds', n)
+
+ # Increments.
+ if isinstance(inc, list) and len(inc) != n:
+ raise RelaxLenError('increment', n)
elif isinstance(inc, int):
- temp = []
- for j in xrange(n):
- temp.append(inc)
- inc = temp
+ inc = [inc]*n
# Minimisation options initialisation.
- min_options = []
+ default_bounds = False
+ if not lower:
+ lower = []
+ default_bounds = True
+ if not upper:
+ upper = []
j = 0
# Loop over the parameters.
for i in xrange(len(spin.params)):
# Relaxation rate (from 0 to 20 s^-1).
- if spin.params[i] == 'Rx':
- min_options.append([inc[j], 0.0, 20.0])
+ if spin.params[i] == 'Rx' and default_bounds:
+ lower.append(0.0)
+ upper.append(20.0)
# Intensity
- elif search('^I', spin.params[i]):
+ elif search('^I', spin.params[i]) and default_bounds:
# Find the position of the first time point.
pos = cdp.relax_times.index(min(cdp.relax_times))
- # Scaling.
- min_options.append([inc[j], 0.0,
average(spin.intensities[pos])])
-
- # Increment j.
- j = j + 1
-
- # Set the lower and upper bounds if these are supplied.
- if lower != None:
- for j in xrange(n):
- if lower[j] != None:
- min_options[j][1] = lower[j]
- if upper != None:
- for j in xrange(n):
- if upper[j] != None:
- min_options[j][2] = upper[j]
-
- # Test if the grid is too large.
- grid_size = 1
- for i in xrange(len(min_options)):
- grid_size = grid_size * min_options[i][0]
- if isinstance(grid_size, long):
- raise RelaxError("A grid search of size " + repr(grid_size) + "
is too large.")
-
- # Diagonal scaling of minimisation options.
- for j in xrange(len(min_options)):
- min_options[j][1] = min_options[j][1] / scaling_matrix[j, j]
- min_options[j][2] = min_options[j][2] / scaling_matrix[j, j]
-
- return grid_size, min_options
+ # Defaults.
+ lower.append(0.0)
+ upper.append(average(spin.intensities[pos]))
+
+ # Parameter scaling.
+ lower[i] = lower[i] / scaling_matrix[i, i]
+ upper[i] = upper[i] / scaling_matrix[i, i]
+
+ return inc, lower, upper
def linear_constraints(self, spin=None, scaling_matrix=None):
@@ -642,7 +622,7 @@
# Get the grid search minimisation options.
if match('^[Gg]rid', min_algor):
- grid_size, min_options = self.grid_search_setup(spin=spin,
param_vector=param_vector, lower=lower, upper=upper, inc=inc,
scaling_matrix=scaling_matrix)
+ inc, lower, upper = self.grid_search_setup(spin=spin,
param_vector=param_vector, lower=lower, upper=upper, inc=inc,
scaling_matrix=scaling_matrix)
# Linear constraints.
if constraints:
@@ -660,10 +640,6 @@
string = "Fitting to spin " + repr(spin_id)
print(("\n\n" + string))
print((len(string) * '~'))
-
- # Grid search print out.
- if match('^[Gg]rid', min_algor):
- print(("Unconstrained grid search size: " +
repr(grid_size) + " (constraints may decrease this size).\n"))
# Initialise the function to minimise.
@@ -703,13 +679,37 @@
# Minimisation.
###############
+ # Constrained optimisation.
if constraints:
- results = generic_minimise(func=func, dfunc=dfunc,
d2func=d2func, args=(), x0=param_vector, min_algor=min_algor,
min_options=min_options, func_tol=func_tol, grad_tol=grad_tol,
maxiter=max_iterations, A=A, b=b, full_output=True, print_flag=verbosity)
+ # Grid search.
+ if search('^[Gg]rid', min_algor):
+ results = grid(func=func, args=(), num_incs=inc,
lower=lower, upper=upper, verbosity=verbosity)
+
+ # Minimisation.
+ else:
+ results = generic_minimise(func=func, dfunc=dfunc,
d2func=d2func, args=(), x0=param_vector, min_algor=min_algor,
min_options=min_options, func_tol=func_tol, grad_tol=grad_tol,
maxiter=max_iterations, A=A, b=b, full_output=True, print_flag=verbosity)
+
+ # Unconstrained optimisation.
else:
- results = generic_minimise(func=func, dfunc=dfunc,
d2func=d2func, 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=True, print_flag=verbosity)
+ # Grid search.
+ if search('^[Gg]rid', min_algor):
+ results = grid(func=target.func, args=(),
incs=min_options, verbosity=verbosity)
+
+ # Minimisation.
+ else:
+ results = generic_minimise(func=func, dfunc=dfunc,
d2func=d2func, 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=True, print_flag=verbosity)
+
+ # Unpack the results.
if results == None:
return
- param_vector, chi2, iter_count, f_count, g_count, h_count,
warning = results
+ if search('^[Gg]rid', min_algor):
+ param_vector, chi2, iter_count = results
+ f_count = iter_count
+ g_count = 0.0
+ h_count = 0.0
+ warning = None
+ else:
+ param_vector, chi2, iter_count, f_count, g_count, h_count,
warning = results
# Scaling.
if scaling:
r9556 - /1.3/specific_fns/relax_fit.py