Added North Uganda 2020 Data

data/raw.dvc

@@ -1,6 +1,6 @@
 outs:
-- md5: 0916e33f6eef6c80a87e319427005f5e.dir
-  size: 446720790
-  nfiles: 408
+- md5: f08af5bba486092e8c1ce82375a2e247.dir
+  size: 447063881
+  nfiles: 410
   path: raw
   hash: md5

data/report.txt

@@ -521,3 +521,12 @@ eo_data_complete    1000
 ✔ training amount: 387, positive class: 1.3%
 ✔ validation amount: 294, positive class: 1.0%
 ✔ testing amount: 319, positive class: 1.3%
+
+
+
+Uganda_NorthCEO2020 (Timesteps: 24)
+----------------------------------------------------------------------------
+eo_data_complete    1000
+✔ training amount: 387, positive class: 21.4%
+✔ validation amount: 294, positive class: 15.3%
+✔ testing amount: 319, positive class: 14.1%

datasets.py

@@ -540,6 +540,38 @@ def load_labels(self) -> pd.DataFrame:
         return df
 
 
+class Uganda_NorthCEO2020(LabeledDataset):
+    def load_labels(self) -> pd.DataFrame:
+        raw_folder = raw_dir / "Uganda_North_2020"
+        df1 = pd.read_csv(
+            raw_folder
+            / "ceo-UNHCR-North-Uganda-Feb-2019---Feb-2020-(Set-1)-sample-data-2024-10-01.csv"
+        )
+        df2 = pd.read_csv(
+            raw_folder
+            / "ceo-UNHCR-North-Uganda-Feb-2019---Feb-2020-(Set-2)-sample-data-2024-10-01.csv"
+        )
+        df = pd.concat([df1, df2])
+
+        # Discard rows with no label
+        df = df[~df["Does this pixel contain active cropland?"].isna()].copy()
+        df[CLASS_PROB] = df["Does this pixel contain active cropland?"] == "Crop"
+        df[CLASS_PROB] = df[CLASS_PROB].astype(int)
+        df["num_labelers"] = 1
+        df = df.groupby([LON, LAT], as_index=False, sort=False).agg(
+            {
+                CLASS_PROB: "mean",
+                "num_labelers": "sum",
+                "plotid": join_unique,
+                "sampleid": join_unique,
+                "email": join_unique,
+            }
+        )
+        df[START], df[END] = date(2019, 1, 1), date(2020, 12, 31)
+        df[SUBSET] = train_val_test_split(df.index, 0.3, 0.3)
+        return df
+
+
 class Uganda_NorthCEO2021(LabeledDataset):
     def load_labels(self) -> pd.DataFrame:
         raw_folder = raw_dir / "Uganda_North_2021"
@@ -1571,6 +1603,7 @@ def load_labels(self) -> pd.DataFrame:
     FranceCropArea2020(),
     Uganda_NorthCEO2016(),
     Uganda_NorthCEO2017(),
+    Uganda_NorthCEO2020(),
 ]
 
 if __name__ == "__main__":