add delta analysis

projects/altitude_spatial_model/delta_graph.py

+import numpy as np
+import matplotlib.pyplot as plt
+from extreme_data.meteo_france_data.scm_models_data.abstract_study import AbstractStudy
+from extreme_data.meteo_france_data.scm_models_data.safran.safran import SafranSnowfall1Day, SafranTemperature, \
+    SafranPrecipitation1Day
+from extreme_data.meteo_france_data.scm_models_data.visualization.main_study_visualizer import \
+    STUDY_CLASS_TO_ABBREVIATION
+from extreme_data.meteo_france_data.scm_models_data.visualization.study_visualizer import StudyVisualizer
+from projects.contrasting_trends_in_snow_loads.article2_snowfall_versus_time_and_altitude.snowfall_plot import \
+    fit_linear_regression
+
+
+def delta_graph(study_class, altitudes, maxima=True):
+    ax = plt.gca()
+    all_delta_point_for_regression = []
+    # colors = ['orange', 'red', 'blue', 'green', 'yellow']
+    for altitude in altitudes:
+        delta_points = []
+        study = study_class(altitude=altitude)  # type : AbstractStudy
+        study_temperature = SafranTemperature(altitude=altitude)
+        for massif_name in study.study_massif_names:
+            if maxima:
+                values = study.massif_name_to_annual_maxima[massif_name]
+            else:
+                values = study.massif_name_to_annual_total[massif_name]
+            values_temperature = study_temperature.massif_name_to_annual_total[massif_name]
+            delta_point = (compute_delta(values_temperature, relative=False), compute_delta(values, relative=True))
+            delta_points.append(delta_point)
+        all_delta_point_for_regression.extend(delta_points)
+        # Plot part
+        x, y = list(zip(*delta_points))
+        ax.scatter(x, y, label='{}m'.format(altitude))
+        aggregation_str = 'maxima' if maxima else 'total'
+        ylabel = 'Delta for {} {}'.format(aggregation_str, STUDY_CLASS_TO_ABBREVIATION[study_class])
+        ax.set_ylabel(ylabel)
+        ax.set_xlabel('Delta mean Temperature')
+    # Plot linear regression
+    x_all, y_all = list(zip(*all_delta_point_for_regression))
+    a, b, r2_score = fit_linear_regression(x_all, y_all)
+    a = a[0]
+    x_plot = np.linspace(start=np.min(x_all), stop=np.max(x_all), num=100)
+    y_plot = a * x_plot + b
+    rounded_number = [str(np.round(e, 2)) for e in [a, b, r2_score]]
+    ax.plot(x_plot, y_plot, label='{} x + {} (R2 = {})'.format(*rounded_number))
+    visualizer = StudyVisualizer(study=study, show=False, save_to_file=True)
+    visualizer.plot_name = ylabel
+    ax.legend()
+
+    # Show / Save plot
+    visualizer.show_or_save_to_file(no_title=True)
+    plt.close()
+
+
+def compute_delta(values, relative=True):
+    index = 30
+    before, after = values[:index], values[index:]
+    mean_before, mean_after = np.mean(before), np.mean(after)
+    delta = mean_after - mean_before
+    if relative:
+        delta *= 100 / mean_before
+    return delta
+
+
+if __name__ == '__main__':
+    fast = False
+    if fast is None:
+        altitudes = [900, 1800, 2700]
+    elif fast:
+        altitudes = [900]
+    else:
+        altitudes = [300, 600, 900, 1200, 1500, 1800, 2100, 2400, 2700, 3000, 3300, 3600][:]
+    for study_class in [SafranSnowfall1Day, SafranPrecipitation1Day]:
+        for maxima in [True, False]:
+            delta_graph(study_class, altitudes, maxima=maxima)

projects/seasonal_analysis/main_season_repartition_of_maxima.py

@@ -64,7 +64,7 @@ def get_month_to_maxima(massif_names, studies, years):
 
 if __name__ == '__main__':
     study_class = SafranSnowfall1Day
-    'Vercors'
+    # 'Vercors'
 
     norht_massif_names = ['Oisans', 'Grandes-Rousses', 'Haute-Maurienne', 'Vanoise',
                           'Maurienne', 'Belledonne', 'Chartreuse', 'Haute-Tarentaise',