[GSAS-II] "copy results to next histogram" option: Unexpected behavior

[Ivo Alxneit]

Dear all

problem solved. It was all just my own stupidity! I was un-selecting the
HStrain parameters in one phase from a script as below

for h in phase.histograms():
    HAPdict = phase.getHAPvalues(h)
    HAPdict['HStrain'][1] = [False for i in HAPdict['HStrain'][1]]
    phase.setHAPvalues(HAPdict)

Not specifying targetlist=h in phase.setHAPvalues() uses the default
`all` thus overwriting the HAPdict in all histograms with the data of
the first one.

So, all is fine, sorry for bothering the list.

Ivo


On 24.09.19 08:22, Ivo Alxneit via GSAS-II wrote:
> Dear all
> 
> I performed a sequential refinement (7 batches with 1250 histograms
> each, results are spliced together). During the nth refinement I refined
> cell parameters of one of the phases (phase A). The result, see blue
> curve of attached pdf. During the next refinement I removed these cell
> parameters from the list of the parameters to be refined but refined
> other parameters. I also selected the "copy results to next histogram"
> option. Unexpectedly, I find that also the cell parameters of phase A
> have been copied forward during the refinement (red curve in pdf): For
> each batch the result of the first histogram is used throughout the
> whole batch leading to the horizontal segments in the red curve. Is this
> really the intended behavior? I would expect that only refined
> parameters are copied forward to be used as starting values. Non-refined
> parameters should not be touched.
> 
> Any clarification appreciated.
> 
> Ivo
> 
> 
> _______________________________________________
> GSAS-II mailing list
> GSAS-II at aps.anl.gov
> https://mailman.aps.anl.gov/mailman/listinfo/gsas-ii
> 

-- 
Dr. Ivo Alxneit
Catalysis for Energy Group
Bioenergy and Catalysis Laboratory        phone: +41 56 310 4092
Paul Scherrer Institute                     fax: +41 56 310 2688
CH-5232 Villigen                      gnupg key: 0x515E30C7
Switzerland
https://www.psi.ch/ceg/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 181 bytes
Desc: OpenPGP digital signature
URL: <http://mailman.aps.anl.gov/pipermail/gsas-ii/attachments/20190925/53259f98/attachment.sig>