Need a simplified interface to simultaneous fitting for "combining" data (Trac #28)

[butlerpd]

a lot of SANS data is collected over more than one instrument setting in order to extend the q range probed. This is normally not an issue for data analysis as it is the job of the reduction to combine those sets into one set with appropriate resolution information attached to each point. However, it may happen that this is not the case from time to time, probably because they came from different instruments (i.e. SANS and USANS). Even then it would be straightforward to put them in one file of course, unless one is not on an absolute scale for example.

It would be nice to provide a simple interface for dealing with this. It can of course be done using the general simultaneous fit, but something that either allows one to identify two data sets as being treated as one for model fitting purpose, or setting up a simultaneous fit with only one model (not one model per data set) and one set of parameters. It might be appropriate to also allow the choice of independent scale factors or not (i.e. in case they are not on the same absolute scale)

Migrated from http://trac.sasview.org/ticket/28

{
    "status": "new",
    "changetime": "2017-10-26T09:43:32",
    "_ts": "2017-10-26 09:43:32.012499+00:00",
    "description": "a lot of SANS data is collected over more than one instrument setting in order to extend the q range probed.  This is normally not an issue for data analysis as it is the job of the reduction to combine those sets into one set with appropriate resolution information attached to each point.  However, it may happen that this is not the case from time to time, probably because they came from different instruments (i.e. SANS and USANS).  Even then it would be straightforward to put them in one file of course, unless one is not on an absolute scale for example.\n\nIt would be nice to provide a simple interface for dealing with this.  It can of course be done using the general simultaneous fit, but something that either allows one to identify two data sets as being treated as one for model fitting purpose, or setting up a simultaneous fit with only one model (not one model per data set) and one set of parameters.  It might be appropriate to also allow the choice of independent scale factors or not (i.e. in case they are not on the same absolute scale)",
    "reporter": "butlerpd",
    "cc": "",
    "resolution": "",
    "workpackage": "SasView GUI Redesign",
    "time": "2012-03-29T17:18:57",
    "component": "SasView",
    "summary": "Need a simplified interface to simultaneous fitting for \"combining\" data",
    "priority": "minor",
    "keywords": "",
    "milestone": "SasView 5.1.0",
    "owner": "",
    "type": "enhancement"
}

[butlerpd]

Trac update at 2012/05/06 15:58:05: butlerpd changed milestone from "" to "WishList"

[butlerpd]

Trac update at 2015/01/30 02:32:45: butler changed workpackage from "" to "SasView GUI Redesign"

[ajj]

Trac update at 2017/01/17 13:05:47: ajj changed component from "SansView" to "SasView"

[butlerpd]

Trac update at 2017/10/26 09:43:32:

• butler changed description from:

a lot of SANS data is collected over more than one instrument setting in order to extend the q range probed. This is normally not an issue for data analysis as it is the job of the reduction to combine those sets into one set with appropriate resolution information attached to each point. However, it may happen that this is not the case from time to time, probably because they came from different instruments (i.e. SANS and USANS). Even then it would be straightforward to put them in one file of course, unless one is not on an absolute scale for example.

It would be nice to provide a simple interface for dealing with this. It can of course be done using the general simultaneous fit, but something that either allows one to identify to data sets as being treated as one for model fitting purpose, or setting up a simultaneous fit with only one model (not one model per data set) and one set of parameters. It might be appropriate to also allow the choice of independent scale factors or not (i.e. in case they are not on the same absolute scale)

to:

a lot of SANS data is collected over more than one instrument setting in order to extend the q range probed. This is normally not an issue for data analysis as it is the job of the reduction to combine those sets into one set with appropriate resolution information attached to each point. However, it may happen that this is not the case from time to time, probably because they came from different instruments (i.e. SANS and USANS). Even then it would be straightforward to put them in one file of course, unless one is not on an absolute scale for example.

It would be nice to provide a simple interface for dealing with this. It can of course be done using the general simultaneous fit, but something that either allows one to identify two data sets as being treated as one for model fitting purpose, or setting up a simultaneous fit with only one model (not one model per data set) and one set of parameters. It might be appropriate to also allow the choice of independent scale factors or not (i.e. in case they are not on the same absolute scale)

• butler changed milestone from "SasView WishList" to "SasView 5.1.0"