visualize_forecast.py

def plot_results(mean_predicted_values, confidence_interval_predicted_values, time_series):
    """
    This function plots actual time series data against SARIMA model-predicted values. 
    We include the confidence interval for the predictions. 
    Args:
        mean_predicted_values: Series of float values. The model-predicted values.
        confidence_interval_predicted_values: Pandas dataframe, containing the lower and
        upper confidence intervals.
        time_series: Series of float values. Actual time series values that we want to graph
    Outputs:
        None. Plot of the time series values, as well as the predicted values and associated 
        confidence interval.
    """
    ax = time_series.plot(label='Observed')
    mean_predicted_values.plot(ax=ax, label = 'Forecast', alpha=.7, figsize=(14, 4))
    ax.fill_between(confidence_interval_predicted_values.index,
                    confidence_interval_predicted_values.iloc[:, 0],
                    confidence_interval_predicted_values.iloc[:, 1], color='k', alpha=.2)
    ax.set_xlabel('Date Index')
    ax.set_ylabel('Value')
    plt.legend()
    plt.show()

### EXECUTE IN MAIN FUNCTION ###
#Plot the predictions against the real data
plot_results(mean_predicted_values, 
             confidence_interval_predicted_values, 
             df['Geothermal_net_generation'][400:])