add bar graph plot showing vac/unvac percent indicator grouped by age

drees.py

@@ -8,8 +8,8 @@ from functools import partial
 from multiprocessing import Pool
 from typing import Any, Dict, List, Optional, OrderedDict, Tuple
 
-import matplotlib
 import numpy as np
+import pandas as pd
 import requests
 from matplotlib import dates as md
 from matplotlib import pyplot as plt
@@ -55,7 +55,7 @@ def get_data(
     refresh=False,
 ) -> Dict[str, Any]:
     """
-    Collect covid data by age from DREES
+    collect covid data by age from DREES
     """
     os.makedirs(DATA_REPOSITORY, exist_ok=True)
     data_url = DATA_URL.format(extension=extension)
@@ -78,7 +78,7 @@ def get_data(
 
 def group_by_age_date(data: Dict[str, Any], fields: List[str]) -> Dict[dt, Any]:
     """
-    Group the original dictionnary into a more readable one
+    group the original dictionnary into a more readable one
     'date': {
         'age' : {
             'vac_status' : {
@@ -221,11 +221,11 @@ def extract_field_values(fields: List[Dict[str, Any]], field: Field) -> np.ndarr
     return np.asarray(field_values)
 
 
-def plot_data_by_field(
+def plot_data_by_age_vac(
     dic_data_grouped: Dict[dt, Any], age: AgeGroup, vac_status: VacStatus
 ) -> None:
     """
-    Plot data by vaccine status, age and field
+    plot data by vaccine status, age and field
     """
     fig = get_plot_fig()
 
@@ -238,8 +238,64 @@ def plot_data_by_field(
         plt.ylabel("nombre")
         plt.title(f"{age}ans - {vac_status}")
 
-    save_and_close_fig(
+    save_and_close_fig(fig, os.path.join(OUTPUT_REPOSITORY, f"{age}_{vac_status}.pdf"))
-        fig, os.path.join(OUTPUT_REPOSITORY, f"{age}_{vac_status}_{field}.pdf")
+
+
+def group_by_date_age_vac(
+    dic_data_grouped: Dict[dt, Any],
+    field: Field,
+    is_vac: Optional[bool] = True,
+    limit_days: Optional[int] = 30,
+) -> Dict[str, Any]:
+    dic_data: Dict[str, Any] = OrderedDict()
+    for date, dic_age in dic_data_grouped.items():
+        if abs((date - dt.now())).days >= limit_days:
+            continue
+        date_format = date.strftime(DATE_FORMAT)
+        dic_data[date_format] = OrderedDict()
+        for age, dic_vac in dic_age.items():
+            nb_vac, nb_unvac = 0, 0
+            for vac_status, dic_field in dic_vac.items():
+                if vac_status == VacStatus.NC.value:
+                    nb_unvac += dic_field.get(field.value, 0)
+                    continue
+                nb_vac += dic_field.get(field.value, 0)
+            sum_vac = nb_vac + nb_unvac
+            try:
+                percent_vac = (nb_vac / sum_vac) * 100
+            except ZeroDivisionError:
+                percent_vac = 0
+            try:
+                percent_unvac = (nb_unvac / sum_vac) * 100
+            except ZeroDivisionError:
+                percent_unvac = 0
+            dic_data[date_format][age] = percent_vac if is_vac else percent_unvac
+    return dic_data
+
+
+def plot_bar_data_by_field(
+    dic_data_grouped: Dict[dt, Any], field: Field, is_vac: Optional[bool] = True
+) -> None:
+    """
+    display a bar graph by field grouped by age over the data period
+    bars displays vaccine status percent
+    """
+    plt.rcParams["font.size"] = "24"
+    dic_data = group_by_date_age_vac(dic_data_grouped, field, is_vac=is_vac)
+    df = pd.DataFrame(dic_data).T
+
+    ax = df.plot.bar(figsize=(26, 15))
+    ax.set_title(f"{field.value} vaccinate percent grouped by age")
+    ax.set_xlabel("date")
+    fig = ax.get_figure()
+
+    plt.xticks(rotation=45)
+    plt.legend(loc="upper right")
+    plt.tight_layout()
+
+    filename = "vac" if is_vac else "unvac"
+    fig.savefig(
+        os.path.join(OUTPUT_REPOSITORY, f"{filename}_age_grouped_{field.value}.pdf")
     )
 
 
@@ -260,6 +316,7 @@ if __name__ == "__main__":
     Plots availables :
         - cumulative deaths by age
         - indicators by vaccine status and age
+        - indicators vaccine/unvaccine percent grouped by age 
     Main indicators are :
         - hospitalisations
         - criticals
@@ -287,9 +344,13 @@ if __name__ == "__main__":
     )
 
     plot_data_pool_args = build_data_pool_args()
-    f = partial(plot_data_by_field, dic_data_grouped)
+    f = partial(plot_data_by_age_vac, dic_data_grouped)
     with Pool() as pool:
         pool.starmap(f, plot_data_pool_args)
 
     for field in Field:
         plot_cumulative_field(dic_data_grouped, field)
+
+    for field in Field:
+        plot_bar_data_by_field(dic_data_grouped, field)
+        plot_bar_data_by_field(dic_data_grouped, field, is_vac=False)