Fix/use dagster

deforestation_api/dependencies/deforestationdata.py

@@ -33,6 +33,7 @@ def fetch_parquet(path) -> pd.DataFrame:
         path,
         storage_options={"anon": True},
     )
+    df = df.rename(columns={"HYBAS_ID": "id"})
     logger.info(f"Done reloading data from %s", path)
     return df
 

deforestation_api/routers/deforestation.py

@@ -55,6 +55,22 @@ def add_treecover_loss_data(
 ) -> None:
     basin_loss = treecover_loss[treecover_loss["id"] == feature["id"]]
     basin_loss = basin_loss.loc[basin_loss["year"].between(start_year, end_year)]
+
+    basin_loss = (
+        basin_loss.groupby(["id", "year"])
+        .agg(
+            tree_loss_incidents=("tree_loss_incidents", "sum"),
+            first_cell_area=("first_cell_area", "first"),
+        )
+        .reset_index()
+    )
+    basin_loss["area"] = (
+        basin_loss["tree_loss_incidents"] * basin_loss["first_cell_area"]
+    )
+    basin_loss["relative_area"] = (
+        basin_loss["area"] / feature["properties"]["basin_area"]
+    )
+
     total_loss = basin_loss["area"].sum()
     relative_loss = basin_loss["relative_area"].sum()
     loss_per_year = basin_loss.drop(columns="id").to_json(orient="records")

deforestation_api/settings.py

@@ -1,4 +1,5 @@
 from pydantic_settings import BaseSettings
+from os import environ
 
 
 class Settings(BaseSettings):
@@ -7,9 +8,10 @@ class Settings(BaseSettings):
     uvicorn_host: str = "0.0.0.0"
     uvicorn_reload: bool = True
     uvicorn_proxy_headers: bool = False
-    basin_data_path: str = "s3://databricks-data-openepi/deforestation/basins.parquet"
+    dagster_data_bucket: str = environ.get("dagster_data_bucket", "placeholder-bucket")
+    basin_data_path: str = f"s3://{dagster_data_bucket}/basin/basins.parquet"
     lossyear_data_path: str = (
-        "s3://databricks-data-openepi/deforestation/lossyear.parquet"
+        f"s3://{dagster_data_bucket}/deforestation/treeloss_per_basin/"
     )
     api_root_path: str = ""
     api_description: str = 'This is a RESTful service that provides aggregated deforestation data over the period from 2001 to 2022 based on data provided by <a href="https://glad.umd.edu/">Global Land Analysis and Discovery (GLAD)</a> laboratory at the University of Maryland, in partnership with <a href="https://www.globalforestwatch.org/">Global Forest Watch (GFW)</a>. The data are freely available for use under a <a href="https://creativecommons.org/licenses/by/4.0/">Creative Commons Attribution 4.0 International License</a>.<br/><i>Citation: Hansen, M. C., P. V. Potapov, R. Moore, M. Hancher, S. A. Turubanova, A. Tyukavina, D. Thau, S. V. Stehman, S. J. Goetz, T. R. Loveland, A. Kommareddy, A. Egorov, L. Chini, C. O. Justice, and J. R. G. Townshend. 2013. High-Resolution Global Maps of 21st-Century Forest Cover Change. Science 342 (15 November): 850-53. Data available on-line from: <a href="https://glad.earthengine.app/view/global-forest-change">https://glad.earthengine.app/view/global-forest-change</a></i>.<br/><br/>The data provided by the `basin` endpoint are aggregated over river basin polygons provided by <a href="https://www.hydrosheds.org/products/hydrobasins">HydroSHEDS</a>. The basin data are feely available for non-commercial and commercial use under a licence agreement included in the <a href="https://data.hydrosheds.org/file/technical-documentation/HydroSHEDS_TechDoc_v1_4.pdf">HydroSHEDS Technical Documentation</a>.'

deployment/kubernetes/deforestation-api.yaml

@@ -18,15 +18,15 @@ spec:
         prometheus.io/path: "/metrics"
     spec:
       containers:
-        - image: ghcr.io/openearthplatforminitiative/deforestation-api:0.3.0
+        - image: ghcr.io/openearthplatforminitiative/deforestation-api:0.3.1
           name: deforestation-api
           ports:
             - containerPort: 8080
           env:
             - name: API_ROOT_PATH
               value: "/deforestation"
             - name: VERSION
-              value: 0.3.0
+              value: 0.3.1
             - name: API_DOMAIN
               valueFrom:
                 configMapKeyRef: