Cuong Duong

When there's a constant value series, does the model fit actually succeed? I'm wondering if the values in `.params` are just the prior values. And if the model fit doesn't...

Could you please provide some example plots from your branch to validate that the functionality is still working as intended, thanks!

Nice one! I guess we can allow the user to pass a command line flag to use GPU / OpenCL for compilation. There's really no downside to this if it's...

Yes, sure if anyone wants to have a go at creating API docs that'd be great. I'm not very experienced in this space but I'm guessing as a minimum we'd...

We had a similar issue with plotting that was resolved by upgrading to the latest versions of numpy and pandas, could you try that first?

This issue (and the other linked ones) might be related: https://github.com/facebook/prophet/issues/2357

It seems like PyPI has increased their security for releases. Github Actions now has to be added as a "trusted publisher" for the PyPI upload to work, which only the...

Feel free to, although as the original post says, the __init__ method is already quite bloated. What do you think about letting `changepoint_range` be `float | tuple[datetime.datetime, datetime.datetime]` ? Seems...

Hey there, we recently added utility functions to return the coefficients of the extra regressors on the original scale of the data. See https://facebook.github.io/prophet/docs/seasonality,_holiday_effects,_and_regressors.html#coefficients-of-additional-regressors Let me know if you have...

Yes the output of the utility functions is on the scale of the original data. Here is the derivation from the source code: https://github.com/facebook/prophet/blob/b75844e07c3b09bf3bc383c9d43241c554dd251b/python/prophet/utilities.py#L28-L77