Author: bugman
Date: Thu Jul 24 14:37:10 2008
New Revision: 6948
URL: http://svn.gna.org/viewcvs/relax?rev=6948&view=rev
Log:
Fixed the model-free optimisation code to use the dchi2_element and
d2chi2_element functions.
Modified:
1.3/maths_fns/mf.py
Modified: 1.3/maths_fns/mf.py
URL:
http://svn.gna.org/viewcvs/relax/1.3/maths_fns/mf.py?rev=6948&r1=6947&r2=6948&view=diff
==============================================================================
--- 1.3/maths_fns/mf.py (original)
+++ 1.3/maths_fns/mf.py Thu Jul 24 14:37:10 2008
@@ -655,7 +655,7 @@
data.create_dri[m](data, m, data.remap_table[m],
data.get_dr1, params, j)
# Calculate the chi-squared gradient.
- data.dchi2[j] = dchi2(data.relax_data, data.ri, data.dri[j],
data.errors)
+ data.dchi2[j] = dchi2_element(data.relax_data, data.ri,
data.dri[j], data.errors)
# Diagonal scaling.
if self.scaling_flag:
@@ -716,7 +716,7 @@
data.create_dri[m](data, m, data.remap_table[m],
data.get_dr1, params, j)
# Calculate the chi-squared gradient.
- data.dchi2[j] = dchi2(data.relax_data, data.ri, data.dri[j],
data.errors)
+ data.dchi2[j] = dchi2_element(data.relax_data, data.ri,
data.dri[j], data.errors)
# Diagonal scaling.
if self.scaling_flag:
@@ -792,7 +792,7 @@
data.create_dri[m](data, m, data.remap_table[m],
data.get_dr1, params, j)
# Calculate the chi-squared gradient.
- data.dchi2[j] = dchi2(data.relax_data, data.ri, data.dri[j],
data.errors)
+ data.dchi2[j] = dchi2_element(data.relax_data, data.ri,
data.dri[j], data.errors)
# Index for the construction of the global generic model-free
gradient.
index = self.diff_data.num_params
@@ -873,7 +873,7 @@
data.create_dri[m](data, m, data.remap_table[m],
data.get_dr1, params, j)
# Calculate the chi-squared gradient.
- data.dchi2[j] = dchi2(data.relax_data, data.ri, data.dri[j],
data.errors)
+ data.dchi2[j] = dchi2_element(data.relax_data, data.ri,
data.dri[j], data.errors)
# Index for the construction of the global generic model-free
gradient.
index = self.diff_data.num_params
@@ -932,7 +932,7 @@
data.create_d2ri[m](data, m, data.remap_table[m],
data.get_d2r1, params, j, k)
# Calculate the chi-squared Hessian.
- data.d2chi2[j, k] = data.d2chi2[k, j] =
d2chi2(data.relax_data, data.ri, data.dri[j], data.dri[k], data.d2ri[j, k],
data.errors)
+ data.d2chi2[j, k] = data.d2chi2[k, j] =
d2chi2_element(data.relax_data, data.ri, data.dri[j], data.dri[k],
data.d2ri[j, k], data.errors)
# Diagonal scaling.
if self.scaling_flag:
@@ -985,7 +985,7 @@
data.create_d2ri[m](data, m, data.remap_table[m],
data.get_d2r1, params, j, k)
# Calculate the chi-squared Hessian.
- data.d2chi2[j, k] = data.d2chi2[k, j] =
d2chi2(data.relax_data, data.ri, data.dri[j], data.dri[k], data.d2ri[j, k],
data.errors)
+ data.d2chi2[j, k] = data.d2chi2[k, j] =
d2chi2_element(data.relax_data, data.ri, data.dri[j], data.dri[k],
data.d2ri[j, k], data.errors)
# Diagonal scaling.
if self.scaling_flag:
@@ -1056,7 +1056,7 @@
data.create_d2ri[m](data, m,
data.remap_table[m], data.get_d2r1, params, j, k)
# Calculate the chi-squared Hessian.
- data.d2chi2[j, k] = data.d2chi2[k, j] =
d2chi2(data.relax_data, data.ri, data.dri[j], data.dri[k], data.d2ri[j, k],
data.errors)
+ data.d2chi2[j, k] = data.d2chi2[k, j] =
d2chi2_element(data.relax_data, data.ri, data.dri[j], data.dri[k],
data.d2ri[j, k], data.errors)
# Pure diffusion parameter part of the global generic model-free
Hessian.
self.total_d2chi2 = self.total_d2chi2 + data.d2chi2
@@ -1130,7 +1130,7 @@
data.create_d2ri[m](data, m,
data.remap_table[m], data.get_d2r1, params, j, k)
# Calculate the chi-squared Hessian.
- data.d2chi2[j, k] = data.d2chi2[k, j] =
d2chi2(data.relax_data, data.ri, data.dri[j], data.dri[k], data.d2ri[j, k],
data.errors)
+ data.d2chi2[j, k] = data.d2chi2[k, j] =
d2chi2_element(data.relax_data, data.ri, data.dri[j], data.dri[k],
data.d2ri[j, k], data.errors)
# Index for the construction of the global generic model-free
Hessian.
index = self.diff_data.num_params
r6948 - /1.3/maths_fns/mf.py