Commit | Date | |
---|---|---|
[r25560]
by
bugman
Loosened a value check to allow the Relax_disp.test_r1rho_kjaergaard_missing_r1 system test to pass. This test fails on MS Windows systems. |
2014-09-02 16:00:11 | Tree |
[r25559]
by
tlinnet
Fix for test data extracted from wrong path. task #7824(https://gna.org/task/index.php?7824): Model parameter ERROR estimation from Jacobian and Co-variance matrix of dispersion models. |
2014-09-02 15:52:10 | Tree |
[r25558]
by
tlinnet
Added initial simulated dataset for CR72 with two. The problem is though, that I canÃ't write up the Jacobian. task #7824(https://gna.org/task/index.php?7824): Model parameter ERROR estimation from Jacobian and Co-variance matrix of dispersion models. |
2014-09-02 15:52:04 | Tree |
[r25557]
by
tlinnet
Implemented verify_estimate_par_err_compare_mc, as a general method to compare model parameter estimation, Monte-Carlo and boot strapping. task #7824(https://gna.org/task/index.php?7824): Model parameter ERROR estimation from Jacobian and Co-variance matrix of dispersion models. |
2014-09-02 14:30:48 | Tree |
[r25556]
by
bugman
Improvements for the spectrum.error_analysis button in the spectrum list GUI element. The subset argument is set to None if no spectra are selected. |
2014-09-02 14:27:34 | Tree |
2014-09-02 13:59:09 | Tree | |
[r25554]
by
bugman
Improvements for the spectrum.error_analysis button in the spectrum list GUI element. The user function is now launched as being modal so that the rest of the GUI freezes, and after the |
2014-09-02 13:51:13 | Tree |
[r25553]
by
bugman
Modified the behaviour of the spectrum.error_analysis button in the spectrum list GUI element. Now the subset argument of this user function will be pre-set to any spectra selected in the list. |
2014-09-02 13:40:29 | Tree |
[r25552]
by
tlinnet
Added path to other dataset with BFGS. task #7824(https://gna.org/task/index.php?7824): Model parameter ERROR estimation from Jacobian and Co-variance matrix of dispersion models. |
2014-09-02 13:29:10 | Tree |
[r25551]
by
tlinnet
Added dataset which was produced with BFGS. This shows, that gradient calculation is either wrong in relax, or in minfx. It is also easily to see on the graphs. param: r2a_err, with err: 0.70912495, compared to MC: 0.10395753, with 500 boot 0.66165472 task #7824(https://gna.org/task/index.php?7824): Model parameter ERROR estimation from Jacobian and Co-variance matrix of dispersion models. |
2014-09-02 13:29:08 | Tree |