Discussed in #165
Originally posted by Yasin7795 September 30, 2025
Hello,
I might be missing something, but i noticed that notebook 2_03 glm_2 (and subsequent notebooks) claim that the control file being used in the notebooks is the one created in the Pest_from notebook 2_01. Where observations were supposedly weighted in accordance with the measurement noise (they were not!). However, the notebooks actually use the control file created in notebook 2_02 (obs_and_weights). Where observations weightes were set for visibility, and a standard deviation column was added in the observation dataset for use in uncertainty anlysis. Was this done on purpose?
Using these weights in the GLM notebook results in a pretty big objective function at the end of the run. I tried a seperate run with weights set as the inverse of the values set in the standard deviation column and the objective function was much smaller.
Thank you in advance for any help!
Discussed in #165
Originally posted by Yasin7795 September 30, 2025
Hello,
I might be missing something, but i noticed that notebook 2_03 glm_2 (and subsequent notebooks) claim that the control file being used in the notebooks is the one created in the Pest_from notebook 2_01. Where observations were supposedly weighted in accordance with the measurement noise (they were not!). However, the notebooks actually use the control file created in notebook 2_02 (obs_and_weights). Where observations weightes were set for visibility, and a standard deviation column was added in the observation dataset for use in uncertainty anlysis. Was this done on purpose?
Using these weights in the GLM notebook results in a pretty big objective function at the end of the run. I tried a seperate run with weights set as the inverse of the values set in the standard deviation column and the objective function was much smaller.
Thank you in advance for any help!