cvanoli

thanks !! @mxndrwgrdnr

I tested the branch with a local environment in my computer and two Future Warnings keep coming up: 1 -synthpop\ipu\ipu.py: Line191 > RuntimeWarning: divide by zero encountered in double_scalars `adj...

Hi @hanase! We could create an optional feature with warnings printed when the effect of relaxing the filtering conditions acts. Building on that, I imagine that a callback function could...

Sorry for the late reply. The attribute that corresponds to the census_helpers script is called "tract_to_puma". Try changing that. Thanks

I don't think the problem with the format in the yaml file is caused by the `print(model.summary_table)` before saving it. It does relate to the multiple sources to fit the...

Per potential questions: 1. I agree with you that it is better to store the results as a dictionary. In that way, it will be flexible among different models and...

Regarding the `choicemodels` variant, I think would be better. As you mentioned, the `results.results` would be a tidier dict defined here https://github.com/UDST/choicemodels/blob/master/choicemodels/mnl.py#L262

Hi @LorenzoBottaccioli thanks for helping us improve. Yes, please do send the data and the code for testing. You are using the master branch right? We are working on some...