Alfredo Delgado

Thanks @AlexAndorra this worked perfectly! Is this the only workaround at the moment? If I understand correctly, X and y are only needed to get the dates and the preprocessed...

Hey @wd60622 , @ferrine This R2D2M2 prior will enable extra dims like `geo` for geo-hierarchical modelling? https://github.com/pymc-labs/pymc-marketing/discussions/648#discussioncomment-10394912

Hi @wd60622 This is one of the examples I found in the PyMC docs https://www.pymc.io/projects/examples/en/2022.12.0/howto/sampling_callback.html ```python def my_callback(trace, draw): if len(trace) >= 100: raise KeyboardInterrupt() with model: trace = pm.sample(tune=0,...

@wd60622 Using the first option only works for `pymc` sampler. I was using `numpyro` but seems it doesn't work properly > ```python > mmm = MMM( > ..., > #...

Hi @wd60622 Here is the code that worked for me using pymc sampler. ```python class DivergentEarlyStopping: # only for pymc sampler (for now) def __init__(self, max_divergences): self.count = 0 self.max_divergences...

It seems `callback` for `nuts_sampler="numpyro"` is not supported at the moment. See [here](https://forum.pyro.ai/t/is-callback-supported-for-numpyro-nuts-sampler/6308/2?u=alfredo_delgado)

It's been a while but do we need `beta` back into TanhSaturation? Some other saturation classes are multiplied by `beta`: - LogisticSaturation - InverseScaledLogisticSaturation - TanhSaturationBaselined - HillSaturation - RootSaturation...

@wd60622 yep but it does not exist in TanhSaturation atm. My question is if we need it back or not. > In the case where there is beta in TanhSaturation,...

@wd60622 never mind I think I already understand what you were saying, because `TanhSaturation`, `MichaelisMentenSaturation` and `HillSaturationSigmoid` have their own scaling param, we do not really need to pass any...

Hey @louismagowan, You could use this alternative to get those channel contributions over time if you pass more model variables to `var_names`. I'll let the dev team answer why it's...