try: # Set seaborn style for better visualization sns.set_style("whitegrid") # Initialize a new HyperStudy S = bl.HyperStudy() # Load built-in coal mining disaster dataset S.loadExampleData() # Define Poisson observation model for disaster counts L = bl.om.Poisson('rate') # Define Gaussian random walk transition model for rate changes # sigma controls step size, using 20 steps between 0 and 1.0 T = bl.tm.GaussianRandomWalk('sigma', bl.cint(0, 1.0, 20), target='rate') # Configure and fit the model S.set(L, T) S.fit() # Create figure with specified size plt.figure(figsize=(8, 3)) # Left plot: Disaster counts and inferred rate plt.subplot2grid((1, 3), (0, 0), colspan=2) plt.xlim([1852, 1961]) # Time range matches typical coal mining data plt.bar(S.rawTimestamps, S.rawData, align='center', facecolor='red', alpha=0.5, label='Disasters') S.plot('rate', color='blue', label='Inferred Rate') plt.xlabel('Year') plt.ylabel('Disaster Count') plt.legend() # Right plot: Distribution of sigma hyperparameter plt.subplot2grid((1, 3), (0, 2)) plt.xlim([0, 1]) S.plot('sigma', facecolor='green', alpha=0.7, lw=1, edgecolor='black') plt.xlabel('Sigma') plt.ylabel('Density') # Adjust layout and display plt.tight_layout() plt.show()except ImportError as e: print(f"Error: Required library not found - {e}") print("Please install required packages: pip install bayesloop matplotlib seaborn numpy") except Exception as e: print(f"An error occurred: {e}")

Bayesloop: The Game-Changing Probabilistic Tool for Time-Varying Time Series

Subscribe to our premium content to read the rest.

Keep Reading

AlgoEdge Insights