standardize throws an error because of package lmfit about d47crunch HOT 5 CLOSED

Could you test whether ed04198 works for you? It uses a 6-digit hash followed by a sanitized version of the original sample name. I expect it to work well with any arbitrary sample name.

from d47crunch.

KeyError: "'D47_AU002_(2)' is not a valid Parameters name"

That is an error from lmfit, because lmfit model parameter names cannot use a number of special characters (because it clashes with the expression of mathematical constraints such as param_a = 1/(param_b+1)). From here:

the name must match [a-z_][a-z0-9_]* and cannot be a Python reserved word.

D47crunch uses the (crude) function pf() to solve common sample name problems.

• option 1: edit pf() to handle parentheses. Hacky but easy.
• option 2: instead of parameter names, use a well-chosen hash of the sample name. Safe (no collisions), but now when an error is raised it will be annoying to look up which sample is the problem.

Any preference between these two options?

from d47crunch.

Thanks for the quick reply! Ah, I guess an informative error would suffice here, really. I think it makes total sense not to allow parentheses or other weird characters in the UIDs. I'll just get that stuff out in R and then try again.

from d47crunch.

Hmm after trying to resolve this, I guess forcing the sample id's to not start with a number might be limiting. I got this to work by replacing all (, ), [, ], and + with _, but there are also many samples that have ^[a-z0-9_]*.
Furthermore, it seems like all ETH-1, ETH-2's etc. with their - are also a problem.
EDIT: hmm that cannot be the problem, as it's now explicitly asking for an ETH-1, ETH-2, and ETH-3.

from d47crunch.

This seems to work! The standardize step is pretty slow (several minutes for a 552 samples and 1959 anchors) but I have no idea how fast it should be, and computing session ETFs with fancy maths is also very slow in my R version.
It didn't throw an immediate error about the wrong Sample names this time and seems to have worked out okay!

from d47crunch.