Hierachi index issue

Questions version agnostic
1

hi

with pm.Model(coords=coords) as varying_intercept:
    floor_idx = pm.MutableData("floor_idx", floor_measure, dims="obs_id")
    county_idx = pm.MutableData("county_idx", county, dims="obs_id")

    # Priors
    mu_a = pm.Normal("mu_a", mu=0.0, sigma=10.0)
    sigma_a = pm.Exponential("sigma_a", 1)

    # Random intercepts
    alpha = pm.Normal("alpha", mu=mu_a, sigma=sigma_a, dims="county")
    # Common slope
    beta = pm.Normal("beta", mu=0.0, sigma=10.0)

    # Model error
    sd_y = pm.Exponential("sd_y", 1)

    # Expected value
    y_hat = alpha[county_idx] + beta * floor_idx

my question is: what the alpha[county_idx] means? any document with the syntax?alpha shape is shape(30,1),and the county_idx shape is (1000,1)

2

county_idx is just indexing into alpha. The result will be (1000, ). For example try:

import numpy as np
alpha = np.array([1.5, 2.5])
ix = [0,0,1,1,0,0,1,0,1]
alpha[ix]
>> array([1.5, 1.5, 2.5, 2.5, 1.5, 1.5, 2.5, 1.5, 2.5])
3

got it,thanks!