mailRe: r25519 - /branches/est_par_error/specific_analyses/relax_disp/estimate_r2eff.py


Others Months | Index by Date | Thread Index
>>   [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Header


Content

Posted by Edward d'Auvergne on September 02, 2014 - 09:43:
Hi Troels,

I'm not sure if I can answer your question about the rows/columns of
the Jacobian in the code here.

Regards,

Edward


On 1 September 2014 23:25,  <tlinnet@xxxxxxxxxxxxx> wrote:
Author: tlinnet
Date: Mon Sep  1 23:25:36 2014
New Revision: 25519

URL: http://svn.gna.org/viewcvs/relax?rev=25519&view=rev
Log:
Further extended the script for analysing errors with Jacobian.

Now loops over the dimension.

task #7824(https://gna.org/task/index.php?7824): Model parameter ERROR 
estimation from Jacobian and Co-variance matrix of dispersion models.

Modified:
    branches/est_par_error/specific_analyses/relax_disp/estimate_r2eff.py

Modified: 
branches/est_par_error/specific_analyses/relax_disp/estimate_r2eff.py
URL: 
http://svn.gna.org/viewcvs/relax/branches/est_par_error/specific_analyses/relax_disp/estimate_r2eff.py?rev=25519&r1=25518&r2=25519&view=diff
==============================================================================
--- branches/est_par_error/specific_analyses/relax_disp/estimate_r2eff.py   
    (original)
+++ branches/est_par_error/specific_analyses/relax_disp/estimate_r2eff.py   
    Mon Sep  1 23:25:36 2014
@@ -38,9 +38,9 @@
 from pipe_control.minimise import assemble_scaling_matrix
 from pipe_control.mol_res_spin import generate_spin_string, spin_loop
 from specific_analyses.relax_disp.checks import check_model_type
-from specific_analyses.relax_disp.data import average_intensity, 
is_r1_optimised, loop_exp_frq_offset_point, loop_time, return_cpmg_frqs, 
return_offset_data, return_param_key_from_data, return_r1_data, 
return_r1_err_data, return_r2eff_arrays, return_spin_lock_nu1
+from specific_analyses.relax_disp.data import average_intensity, 
generate_r20_key, is_r1_optimised, loop_exp_frq_offset, 
loop_exp_frq_offset_point, loop_time, return_cpmg_frqs, return_offset_data, 
return_param_key_from_data, return_r1_data, return_r1_err_data, 
return_r2eff_arrays, return_spin_lock_nu1
 from specific_analyses.relax_disp.parameters import assemble_param_vector, 
disassemble_param_vector, param_num
-from specific_analyses.relax_disp.variables import MODEL_CR72, 
MODEL_R2EFF, MODEL_TSMFK01
+from specific_analyses.relax_disp.variables import MODEL_CR72, 
MODEL_R2EFF, MODEL_TSMFK01, PARAMS_R20
 from target_functions.chi2 import chi2_rankN, dchi2
 from target_functions.relax_disp import Dispersion

@@ -243,8 +243,53 @@
         jacobian = tfunc.jacobian(param_vector)
         weights = 1. / tfunc.errors**2

-        # Get the co-variance
-        pcov = multifit_covar(J=jacobian, weights=weights)
+        # Get the shape of the data.
+        NJ, NE, NS, NM, NO, ND = jacobian.shape
+        if NS != 1:
+            raise RelaxError("The number of spins does not fit.")
+
+        # Get the parameters fitted in the model.
+        params = cur_spin.params
+
+        # Set the spin index to 0.
+        si = 0
+        # Loop over the data.
+        for exp_type, frq, offset, ei, mi, oi in 
loop_exp_frq_offset(return_indices=True):
+            param_key = generate_r20_key(exp_type=exp_type, frq=frq)
+
+            # Extract weights.
+            cur_weights = weights[ei, si, mi, oi]
+
+            # Extract every column/row from the first to last columns. Is 
this correct?
+            cur_jacobian = jacobian[0:NJ:1, ei, si, mi, oi]
+
+            # Get the co-variance
+            pcov = multifit_covar(J=cur_jacobian, weights=cur_weights)
+
+            # To compute one standard deviation errors on the parameters, 
take the square root of the diagonal covariance.
+            param_vector_error = sqrt(diag(pcov))
+
+            # Loop over params.
+            for i, param in enumerate(params):
+                # Set the param error name
+                param_err = param + '_err'
+
+                # If param in PARAMS_R20, values are stored in with 
parameter key.
+                if param in PARAMS_R20:
+                    # Copy parameter attribute to error attribute, if not 
in spin Class.
+                    if not hasattr(cur_spin, param_err):
+                        setattr(cur_spin, param_err, 
deepcopy(getattr(cur_spin, param)))
+
+                    # Set error.
+                    getattr(cur_spin, param_err)[param_key] = 
deepcopy(param_vector_error[i])
+
+                else:
+                    # Copy parameter attribute to error attribute, if not 
in spin Class.
+                    if not hasattr(cur_spin, param_err):
+                        setattr(cur_spin, param_err, 
deepcopy(getattr(cur_spin, param)))
+
+                    # Set error.
+                    setattr(cur_spin, param_err, param_vector_error[i])


 #### This class is only for testing.


_______________________________________________
relax (http://www.nmr-relax.com)

This is the relax-commits mailing list
relax-commits@xxxxxxx

To unsubscribe from this list, get a password
reminder, or change your subscription options,
visit the list information page at
https://mail.gna.org/listinfo/relax-commits



Related Messages


Powered by MHonArc, Updated Tue Sep 02 10:00:11 2014