Author: bugman
Date: Fri Nov 27 18:32:47 2009
New Revision: 10006
URL: http://svn.gna.org/viewcvs/relax?rev=10006&view=rev
Log:
Massive simplification of the generic_fns.value.set() function.
This code was previously a mess, spread randomly across the generic code and
specific code. This is
a complete refactorization which will cause breakages for a little while.
Modified:
1.3/generic_fns/value.py
Modified: 1.3/generic_fns/value.py
URL:
http://svn.gna.org/viewcvs/relax/1.3/generic_fns/value.py?rev=10006&r1=10005&r2=10006&view=diff
==============================================================================
--- 1.3/generic_fns/value.py (original)
+++ 1.3/generic_fns/value.py Fri Nov 27 18:32:47 2009
@@ -319,188 +319,46 @@
pipes.test()
# Specific functions.
- return_value = specific_fns.setup.get_specific_fn('return_value',
pipes.get_type())
- set_non_spin_params =
specific_fns.setup.get_specific_fn('set_non_spin_params', pipes.get_type())
-
- # The parameters have been specified.
- if param:
- # Partition the parameters into those which are spin specific and
those which are not.
- spin_params, spin_values, other_params, other_values =
partition_params(val, param)
-
- # Spin specific parameters.
- if spin_params:
- # First test if parameter value already exists, prior to setting
any params.
- if not force:
- # Loop over the spins.
- for spin in spin_loop(spin_id):
- # Skip deselected spins.
- if not spin.select:
- continue
-
- # Loop over the parameters.
- for param in spin_params:
- # Get the value and error.
- temp_value, temp_error = return_value(spin, param)
-
- # Data exists.
- if temp_value != None or temp_error != None:
- raise RelaxValueError((param))
-
- # Loop over the spins.
- for spin in spin_loop(spin_id):
- # Skip deselected spins.
- if not spin.select:
- continue
-
- # Set the individual parameter values.
- for j in xrange(len(spin_params)):
- set_spin_params(value=spin_values[j], error=None,
spin=spin, param=spin_params[j])
-
-
- # All other parameters.
- if other_params:
- set_non_spin_params(value=other_values, param=other_params)
-
-
- # All model parameters (i.e. no parameters have been supplied).
- else:
- # Convert val to a list if necessary.
- if not isinstance(val, list) or not isinstance(val, ndarray):
- val = [val]
-
- # Spin specific models.
- if exists_mol_res_spin_data():
- # Loop over the spins.
- for spin in spin_loop(spin_id):
- # Skip deselected spins.
- if not spin.select:
- continue
-
- # Set the individual parameter values.
- for j in xrange(len(val)):
- set_spin_params(value=val[j], error=None, spin=spin,
param=None)
-
- # Set the non-spin specific parameters.
- set_non_spin_params(value=val, param=param)
+ default_value = specific_fns.setup.get_specific_fn('default_value',
pipes.get_type())
+ get_param_names = specific_fns.setup.get_specific_fn('get_param_names',
pipes.get_type())
+ set_param_values =
specific_fns.setup.get_specific_fn('set_param_values', pipes.get_type())
+
+ # Convert numpy arrays to lists, if necessary.
+ if isinstance(val, ndarray):
+ val = val.tolist()
+
+ # Invalid combinations.
+ if (isinstance(val, float) or isinstance(val, int)) and param == None:
+ raise RelaxError("The combination of a single value '%s' without
specifying the parameter name is invalid." % val)
+ if isinstance(val, list) and isinstance(param, str):
+ raise RelaxError("Invalid combination: When multiple values '%s'
are specified, either no parameters or a list of parameters must by supplied
rather than the single parameter '%s'." % (val, param))
+
+ # Get the parameter list if needed.
+ if param == None:
+ param = get_param_names()
+
+ # Convert the param to a list if needed.
+ if not isinstance(param, list):
+ param = [param]
+
+ # Convert the value to a list if needed.
+ if (isinstance(val, float) or isinstance(val, int)):
+ val = [val] * len(param)
+
+ # Default values.
+ if val == None:
+ # Loop over the parameters, getting the default values.
+ val = []
+ for i in range(len(param)):
+ val.append(default_value(param[i]))
+
+ # Set the parameter values.
+ for i in range(len(param)):
+ set_param_value(param=param[i], val=val[i], spin_id=spin_id,
force=force)
# Reset all minimisation statistics.
if reset:
minimise.reset_min_stats()
-
-
-def set_spin_params(value=None, error=None, param=None, scaling=1.0,
spin=None):
- """Set spin specific parameter values.
-
- @param value: The value to change the parameter to.
- @type value: float or str
- @param error: The error value associated with the parameter, also to
be set.
- @type error: float or str
- @param param: The name of the parameter to change.
- @type param: str
- @param scaling: The scaling factor for the value or error parameters.
- @type scaling: float
- @param spin: The SpinContainer object.
- @type spin: SpinContainer
- """
-
- # Specific functions.
- data_init = specific_fns.setup.get_specific_fn('data_init',
pipes.get_type())
- default_value = specific_fns.setup.get_specific_fn('default_value',
pipes.get_type())
- return_data_name =
specific_fns.setup.get_specific_fn('return_data_name', pipes.get_type())
- set_update = specific_fns.setup.get_specific_fn('set_update',
pipes.get_type())
-
-
- # Setting the model parameters prior to minimisation.
- #####################################################
-
- if param == None:
- # The values are supplied by the user:
- if value:
- # Test if the length of the value array is equal to the length
of the parameter array.
- if len(value) != len(spin.params):
- raise RelaxError("The length of " + repr(len(value)) + " of
the value array must be equal to the length of the parameter array, " +
repr(spin.params) + ", for spin " + repr(spin.num) + " " + spin.name + ".")
-
- # Default values.
- else:
- # Set 'value' to an empty array.
- value = []
-
- # Loop over the parameters.
- for i in xrange(len(spin.params)):
- value.append(default_value(spin.params[i]))
-
- # Loop over the parameters.
- for i in xrange(len(spin.params)):
- # Get the object.
- object_name = return_data_name(spin.params[i])
- if not object_name:
- raise RelaxError("The data type " + repr(spin.params[i]) + "
does not exist.")
-
- # Initialise all data if it doesn't exist.
- if not hasattr(spin, object_name):
- data_init(spin)
-
- # Set the value.
- if value[i] == None:
- setattr(spin, object_name, None)
- else:
- # Catch parameters with string values.
- try:
- value[i] = float(value[i]) * scaling
- except ValueError:
- pass
-
- # Set the attribute.
- setattr(spin, object_name, value[i])
-
-
- # Individual data type.
- #######################
-
- else:
- # Get the object.
- object_name = return_data_name(param)
- if not object_name:
- raise RelaxError("The data type " + repr(param) + " does not
exist.")
-
- # Initialise all data if it doesn't exist.
- if not hasattr(spin, object_name):
- data_init(spin)
-
- # Default value.
- if value == None:
- value = default_value(object_name)
-
- # No default value, hence the parameter cannot be set.
- if value == None:
- raise RelaxParamSetError(param)
-
- # Set the value.
- if value == None:
- setattr(spin, object_name, None)
- else:
- # Catch parameters with string values.
- try:
- value = float(value) * scaling
- except ValueError:
- pass
-
- # Set the attribute.
- setattr(spin, object_name, value)
-
- # Set the error.
- if error != None:
- # Catch parameters with string values.
- try:
- error = float(error) * scaling
- except ValueError:
- pass
-
- # Set the attribute.
- setattr(spin, object_name+'_err', error)
-
- # Update the other parameters if necessary.
- set_update(param=param, spin=spin)
def write(param=None, file=None, dir=None, force=False, return_value=None):
r10006 - /1.3/generic_fns/value.py