Add another model internals to the streamlit app

[wd60622]

How about adding GPs and Fourier transformation pages to the streamlit app?

Would just have to be small wrappers around the components that exist

from pymc_marketing.prior import Prior
from pymc_marketing.mmm import YearlyFourier

# User input
n_order = 6
b = 0.1

# Setup
prior = Prior("Laplace", mu=0, b=b, dims="fourier")
fourier = YearlyFourier(n_order=n_order, prior=prior)

# Displayed in the APP
parameters = fourier.sample_prior()
curve = fourier.sample_curve(parameters)
fig, axes = fourier.plot_curve(curve)

Same thing in the GP case:

• Limit the user input for display purposes but provide some
  Maybe the parameters here:

  pymc-marketing/pymc_marketing/mmm/hsgp.py

  Lines 631 to 638 in bb5f1e5

  	eta_mass: float = 0.05,
  	eta_upper: float = 1.0,
  	ls_lower: float = 1.0,
  	ls_upper: float | None = None,
  	ls_mass: float = 0.9,
  	cov_func: CovFunc = CovFunc.ExpQuad,
  	centered: bool = False,
  	drop_first: bool = True,

• Sample the curve and plot it in the UI

This would be on a separate tab and would just have some surrounding context between the page. Doesn't have to be too complicated

Thoughts? @juanitorduz @cetagostini @louismagowan

[louismagowan]

Yeah great idea! I think there's so many things that the app could be expanded with - and this would be cool to see in there 😁