Model is running forever without progress bar

Questions version agnostic
1 
sampler_config= {"progressbar": True}


tvp_normal_config = {'intercept': {'dist': 'Normal', 'kwargs': {'mu': 0, 'sigma': 2}},
 'beta_channel': {'dist': 'LogNormal',
  'kwargs': {'mu': np.array([2, 1]), 'sigma': np.array([2.1775326 , 1.14026088])}},
 'alpha': {'dist': 'Beta', 'kwargs': {'alpha': 1, 'beta': 3}},
 'lam': {'dist': 'Gamma', 'kwargs': {'alpha': 3, 'beta': 1}},
 'likelihood': {'dist': 'Normal',
  'kwargs': {'sigma': {'dist': 'HalfNormal', 'kwargs': {'sigma': 2}}}},
 'gamma_control': {'dist': 'Normal', 'kwargs': {'mu': 0, 'sigma': 2}},
 'gamma_fourier': {'dist': 'Laplace', 'kwargs': {'mu': 0, 'b': 1}}}

custom_beta_channel_prior = {
    'beta_channel': {
        'dist': 'InverseGamma',
        'mu': prior_mu_array,
        'sigma': 0.3,
         'kwargs': {'mu': 0, 'sigma': 2},
        'dims': ('channel',)
    }
}

custom_lam_prior = {
    'lam': {
        'dist': 'Gamma',
        'mu': scaled_lam_array,
        'sigma': 1,
         'kwargs': {'mu': 0, 'sigma': 2},
        'dims': ('channel',)
    }
}

# Merge the configurations
model6_config = {
    **tvp_normal_config,
                 **custom_beta_channel_prior,
                 **custom_lam_prior
                 }

with pm.Model() as model_6:

    mmm_6 = DelayedSaturatedMMM(
        model_config = model6_config,
        sampler_config = sampler_configuration,
        date_column="wc_sun",
        channel_columns=media_variables,
        control_columns=[
                        'm_affiliates_commiss',
                        'week_1_of_month',
            ]
            ,
        adstock_max_lag=4,
        yearly_seasonality=10
    )

with pm.Model() as model_6:
    trace_6 = mmm_6.fit(X=X_train_sorted,
                    y=y_train_sorted['kpi_volume'],
                    target_accept=0.8,
                    chains=2)
with model_6:
    trace_6 = pm.sample_posterior_predictive(trace_6, extend_inferencedata=True)
2

Welcome!

Could you make your example runnable? There are several variables that don’t exist (e.g., prior_mu_array, scaled_lam_array, sampler_configuration, media_variables, etc.). Then someone might be able to help you out!

3

prior_mu_array

eml_Cost 0.000113
PaidSearch_Cost 0.259171
PaidSocial_Cost 0.075319
Display_Cost 0.165396

scaled_lam_array

eml_Cost 2.000000
PaidSearch_Cost 2.911076
PaidSocial_Cost 2.351029
Display_Cost 3.000000

Sampler_configuration ={“progressbar”: True}

Variables

[‘eml_SENT’,
‘Display_Impressions’,
‘PaidSearch_Impressions’,
‘PaidSocial_Impressions’]

4

I’m not sure how to run these. Can I get a runnable version?

1 Like