geocif 0.1.58__tar.gz → 0.1.60__tar.gz

{geocif-0.1.58/geocif.egg-info → geocif-0.1.60}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: geocif
-Version: 0.1.58
+Version: 0.1.60
 Summary: Models to visualize and forecast crop conditions and yields
 Home-page: https://ritviksahajpal.github.io/yield_forecasting/
 Author: Ritvik Sahajpal

{geocif-0.1.58 → geocif-0.1.60}/geocif/geocif.py

@@ -730,6 +730,8 @@ class Geocif:
 
         if self.median_yield_as_feature:
             self.feature_names.append(f"Median {self.target}")
+            self.feature_names.append(f"Median {self.target} (2014-2018)")
+            self.feature_names.append(f"Median {self.target} (2013-2017)")
 
         if self.lag_yield_as_feature:
             # For the number of years specified in self.number_lag_years
@@ -988,6 +990,14 @@ class Geocif:
             df, self.all_seasons_with_yield, self.number_median_years, self.target
         )
 
+        df = fe.compute_user_median_statistics(
+            df, [2014, 2015, 2016, 2017, 2018]
+        )
+
+        df = fe.compute_user_median_statistics(
+            df, [2013, 2014, 2015, 2016, 2017]
+        )
+
         if self.median_area_as_feature:
             df = fe.compute_median_statistics(
                 df, self.all_seasons_with_yield, self.number_median_years, "Area (ha)"

{geocif-0.1.58 → geocif-0.1.60}/geocif/ml/embedding.py

@@ -32,7 +32,7 @@ def _compute_correlations(X, y):
         f_series = X[feature]
 
         # Ignore NaN values in either y and f_series
-        mask = ~np.isnan(y) | ~np.isnan(f_series)
+        mask = ~ (np.isnan(y) | np.isnan(f_series))
         y = y[mask]
         f_series = f_series[mask]
 
@@ -137,7 +137,7 @@ def get_all_features_correlation(inputs, targets, method):
         df = pd.DataFrame(split_keys, columns=["Metric", method])
         df["Value"] = values
 
-        # If you want to pivot the DataFrame so each metric becomes a column name and include the year as a separate column
+        # Pivot the DataFrame so each metric becomes a column name and include the year as a separate column
         df_pivoted = df.pivot_table(
             index=method, columns="Metric", values="Value", aggfunc="first"
         ).reset_index()

{geocif-0.1.58 → geocif-0.1.60}/geocif/ml/feature_engineering.py

@@ -105,6 +105,49 @@ def compute_median_statistics(
     return df
 
 
+def compute_user_median_statistics(df, user_years, target_col="Yield (tn per ha)"):
+    """
+    Enhances the DataFrame with a new column that contains the median yield computed
+    using only the yields from the user-specified list of years.
+
+    Args:
+        df (DataFrame): The original DataFrame containing yield data.
+        user_years (array-like): List of years to consider for computing the median yield.
+        target_col (str): The column name from which to compute the median yield.
+
+    Returns:
+        DataFrame: The original DataFrame enhanced with a new column for median yield.
+    """
+    # Ensure 'Harvest Year' is treated as integer for accurate comparisons.
+    df["Harvest Year"] = df["Harvest Year"].astype(int)
+
+    # Sort the user_years list to reliably extract the earliest and latest years.
+    user_years_sorted = sorted(user_years)
+    first_year = user_years_sorted[0]
+    last_year = user_years_sorted[-1]
+
+    # Define the new column name to include the range of years.
+    new_col_name = f"Median {target_col} ({first_year}-{last_year})"
+
+    # Initialize the new column with NaN values.
+    df[new_col_name] = np.nan
+
+    # Group by region and compute the median yield for the specified years.
+    for region, group in tqdm(df.groupby("Region"), desc="Median yield", leave=False):
+        # Skip if the target column is completely null for this region.
+        if group[target_col].isnull().all():
+            continue
+
+        # Filter the rows to only include harvest years that are in the user provided list.
+        mask = group["Harvest Year"].isin(user_years)
+        median_yield = group.loc[mask, target_col].median()
+
+        # Assign the computed median yield to all rows in the current region.
+        df.loc[df["Region"] == region, new_col_name] = median_yield
+
+    return df
+
+
 def compute_lag_yield(
     df, all_seasons_with_yield, number_lag_years, target_col="Yield (tn per ha)"
 ):

{geocif-0.1.58 → geocif-0.1.60}/geocif/ml/stats.py

@@ -209,7 +209,7 @@ def add_statistics(
     # Check if country and crop exist in the fewsnet database
     mask = (df_fewsnet["country"] == country) & (df_fewsnet["product"] == crop)
 
-    if mask.sum() == 0 or country == "Zambia":
+    if mask.sum() == 0:
         df = add_GEOGLAM_statistics(dir_stats, df, stats, method, admin_zone)
     else:
         group_by = ["Region", "Harvest Year"]

{geocif-0.1.58 → geocif-0.1.60/geocif.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: geocif
-Version: 0.1.58
+Version: 0.1.60
 Summary: Models to visualize and forecast crop conditions and yields
 Home-page: https://ritviksahajpal.github.io/yield_forecasting/
 Author: Ritvik Sahajpal

{geocif-0.1.58 → geocif-0.1.60}/setup.py

@@ -50,6 +50,6 @@ setup(
     test_suite="tests",
     tests_require=test_requirements,
     url="https://ritviksahajpal.github.io/yield_forecasting/",
-    version="0.1.58",
+    version="0.1.60",
     zip_safe=False,
 )