bootstrapping with spins=3

[vanq2024]

When spins=3 is added to the dipolar modeling, can bootstrap be used?

expinfo=dl.ex_4pdeer(tau1,tau2, pulselength=0.08, pathways=[1])
Vmodel = dl.dipolarmodel(t,r, experiment=expinfo, spins=3)
compactness = dl.dipolarpenalty(Pmodel=None, r=r, type='compactness')
results=dl.fit(Vmodel,Vexp, penalties=compactness)

If I remove the "spins=3", everything runs with no problem.

[HKaras]

Hi,

Can you provide more information on the error?

I have never used multi-spin (spins>2) personally and all the people who were using it have now left 🤣.

In theory everything that works normally should work through bootstrapping but there can always be bugs.

[vanq2024]

Here's the error output
DeerLab_error_spins3_pdf.pdf

[vanq2024]

Sorry, I meant "compactness" in my question, not bootstrap.

[HKaras]

Hi,
I have just checked and in the spins>2, the dipolar model creates a multi-gaussian based model not a non-paramagentic. This can be seen by printing Vmodel and is mentioned in the docs: https://jeschkelab.github.io/DeerLab/_autosummary/deerlab.dipolarmodel.html#deerlab.dipolarmodel

The compactness penalty makes no sense in such a situation as there is P vector to compact.

I hope this helps