version présentable, routes et données dynamiques

src/data/clusters_statistics_mayotte.json.py

@@ -0,0 +1,9 @@
+from utils.fonctions import creer_donnees_comparaison
+
+file_paths = {
+    "2022": "projet-slums-detection/data-prediction/PLEIADES/MAYOTTE/2022/Segmentation/1/statistics_clusters.parquet",
+    "2023": "projet-slums-detection/data-prediction/PLEIADES/MAYOTTE/2023/Segmentation/1/statistics_clusters.parquet",
+}
+
+data = creer_donnees_comparaison(file_paths)
+print(data.to_json())

src/data/clusters_statistics_reunion.json.py

@@ -0,0 +1,11 @@
+from utils.fonctions import creer_donnees_comparaison
+
+file_paths = {
+    "2022": "projet-slums-detection/data-prediction/PLEIADES/REUNION/2022/Segmentation/1/statistics_clusters.parquet",
+    "2023": "projet-slums-detection/data-prediction/PLEIADES/REUNION/2023/Segmentation/1/statistics_clusters.parquet",
+}
+
+data = creer_donnees_comparaison(file_paths)
+print(data.to_json())
+
+data

src/data/utils/fonctions.py

@@ -0,0 +1,73 @@
+import os
+import geopandas as gpd
+import s3fs
+import numpy as np
+
+
+def merge_gdfs(gdfs, id_columns, value_columns):
+    """Merge multiple GeoDataFrames on specified columns."""
+    base_gdf = None
+    for year, gdf in gdfs.items():
+        year_columns = [*id_columns, *[f"{col}_{year}" for col in value_columns]]
+        current_gdf = gdf[year_columns]
+
+        if base_gdf is None:
+            base_gdf = current_gdf
+        else:
+            base_gdf = base_gdf.merge(current_gdf, on=id_columns)
+
+    return base_gdf
+
+
+def creer_donnees_comparaison(file_paths):
+
+    id_columns = ["ident_ilot", "code", "depcom_2018", "ident_up", "dep", "geometry"]
+    value_columns = ["area_cluster", "area_building", "pct_building"]
+
+    # Set up S3 filesystem
+    fs = s3fs.S3FileSystem(
+        client_kwargs={"endpoint_url": f"https://{os.environ['AWS_S3_ENDPOINT']}"},
+        key=os.getenv("AWS_ACCESS_KEY_ID"),
+        secret=os.getenv("AWS_SECRET_ACCESS_KEY"),
+        # token=os.environ["AWS_SESSION_TOKEN"],
+    )
+
+    gdfs = {year: gpd.read_parquet(path, filesystem=fs) for year, path in file_paths.items()}
+
+    # Rename columns in each GeoDataFrame
+    gdfs = {
+        year: gdf.rename(columns={col: f"{col}_{year}" for col in value_columns})
+        for year, gdf in gdfs.items()
+    }
+
+    # Merge all GeoDataFrames
+    merged_gdf = merge_gdfs(gdfs, id_columns, value_columns)
+
+    merged_gdf.loc[:, "building_2023"] = merged_gdf.loc[
+        :, "area_building_2023"
+    ]*1e6
+
+    merged_gdf.loc[:, "area_building_change_absolute"] = (
+        merged_gdf.loc[:, "area_building_2023"] - merged_gdf.loc[:, "area_building_2022"]
+    ) * 1e6
+
+    merged_gdf.loc[:, "area_building_change_relative"] = (merged_gdf.loc[
+        :, "area_building_change_absolute"
+    ] / (merged_gdf.loc[:, "area_building_2022"] * 1e6))*100
+
+    # Order columns
+    ordered_columns = (
+        id_columns[:-1]  # All ID columns except geometry
+        + [
+            "pct_building_2023",
+            "building_2023",
+            "area_building_change_absolute",
+            "area_building_change_relative",
+        ]  # Value columns
+        + ["geometry"]  # Put geometry at the end
+    )
+
+    # Remplacer les NaN et Infinity par 0
+    final_gdf_cleaned = merged_gdf.replace([np.nan, np.inf, -np.inf], 0)
+
+    return (final_gdf_cleaned[ordered_columns].set_index("ident_ilot").to_crs("EPSG:4326"))

src/departement/[departement].md

@@ -30,7 +30,13 @@ import { quantileProbs, colorScales, departementConfig } from '../utils/config.j
 ```
 
 ```js
-const statistics = FileAttachment("../data/clusters_statistics.json").json();
+const files = [
+  {id: "mayotte", file: FileAttachment("../data/clusters_statistics_mayotte.json")},
+  {id: "reunion", file: FileAttachment("../data/clusters_statistics_reunion.json")},
+]
+// Trouve le fichier correspondant au département
+const selec = files.find(f => f.id === departement);
+const statistics = await selec.file.json();
 ```
 
 ```js