Hi Matt,
I was wondering if you'd consider a small change to how lmfit
loads results from JSON. Currently, load_modelresult()
requires a filename or file-like object. If it could also accept a raw JSON string, that would make it easier to store model results alongside other data in a single JSON file, and then re-read just the relevant portion as needed.
I assume one could work around this by writing the string to a temporary file and reading it back, but it seems like a cleaner solution would be to support strings directly — perhaps via an additional function like ModelResult.from_json_string()
or an overload in load_modelresult()
.
Thanks for considering it!
Larry Lurio
--
You received this message because you are subscribed to the Google Groups "lmfit-py" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lmfit-py+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/lmfit-py/CAK4HuLE1sBHjj3NAR3EvrErmiKuJMyump0YFke-WLgPPCXf_%2Bg%40mail.gmail.com.
--
You received this message because you are subscribed to the Google Groups "lmfit-py" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lmfit-py+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/lmfit-py/CAK4HuLE1sBHjj3NAR3EvrErmiKuJMyump0YFke-WLgPPCXf_%2Bg%40mail.gmail.com.