Hmc Checker — !!top!!

ess_ratio = ess / total_samples if np.any(ess_ratio < ess_ratio_threshold): results["warnings"].append(f"Low ESS/total_samples (< {ess_ratio_threshold})")

# 6. Energy plot check (text summary) if hasattr(inference_data, "sample_stats") and hasattr(inference_data.sample_stats, "energy"): energy = inference_data.sample_stats.energy.values # simple check: coefficient of variation across chains chain_means = energy.mean(axis=1) cv = np.std(chain_means) / np.mean(chain_means) if cv > 0.1: results["warnings"].append(f"Energy means vary across chains (CV={cv:.3f})") hmc checker

# 2. ESS / total samples ess = az.ess(inference_data).to_array().values n_samples = inference_data.posterior.sizes["draw"] n_chains = inference_data.posterior.sizes["chain"] total_samples = n_samples * n_chains ess_ratio = ess / total_samples if np

# For demo, create dummy data import pymc as pm with pm.Model(): x = pm.Normal("x") trace = pm.sample(1000, chains=2, return_inferencedata=True) "sample_stats") and hasattr(inference_data.sample_stats