MMM.fit takes long time run

[IG2804]

So, I was just trying to follow along with the example notebook in a local Jupyter notebook in my Anaconda environment.
When I run this part of the code : mmm.fit(X=X, y=y, target_accept=0.85, chains=4, random_seed=rng), it continues to run forever. When I interrupt the kernel finally after 2 hours or so, I get a ValueError message:
ValueError: Not enough samples to build a trace.
Anyone encountered this issue before? I am very new to PyMC package, and was just trying to replicate what has been done in this example notebook given here: https://www.pymc-marketing.io/en/stable/notebooks/mmm/mmm_example.html

[williambdean]

Hi. Are you just running that notebook as is? Or are you changing the data and fit parameters?

[AnantChandra]

Try using the "numpyro" sampler:

mmm.fit(X=X, y=y, target_accept=0.85, chains=4, random_seed=rng, nuts_sampler="numpyro")

[williambdean]

Good suggestion. Rule of thumb is that the default sampler (pymc / C backend) is not used often