geocif 0.1.38__tar.gz → 0.1.39__tar.gz

{geocif-0.1.38/geocif.egg-info → geocif-0.1.39}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: geocif
-Version: 0.1.38
+Version: 0.1.39
 Summary: Models to visualize and forecast crop conditions and yields
 Home-page: https://ritviksahajpal.github.io/yield_forecasting/
 Author: Ritvik Sahajpal

{geocif-0.1.38 → geocif-0.1.39}/geocif/experiments.py

@@ -14,7 +14,7 @@ import warnings
 warnings.simplefilter(action="ignore", category=FutureWarning)
 
 
-def run(inputs, logger, parser, section, item, type, values):
+def main(inputs, logger, parser, section, item, type, values):
     if type == "str":
         original_value = parser.get(section, item)
     elif type == "bool":
@@ -41,7 +41,7 @@ def run(inputs, logger, parser, section, item, type, values):
     return parser
 
 
-def main(path_config_files=[Path("../config/geocif.txt")]):
+def run(path_config_files=[Path("../config/geocif.txt")]):
     logger, parser = log.setup_logger_parser(path_config_files)
     inputs = gc.gather_inputs(parser)
 
@@ -51,13 +51,13 @@ def main(path_config_files=[Path("../config/geocif.txt")]):
 
     # Experiment 1: include_lat_lon
     logger.info("Experiment 1: include_lat_lon")
-    parser = run(
+    parser = main(
         inputs, logger, parser, "ML", "include_lat_lon", "bool", [True, False]
     )
 
     # Experiment 2: feature_selection
     logger.info("Experiment 2: feature_selection")
-    parser = run(
+    parser = main(
         inputs,
         logger,
         parser,
@@ -69,13 +69,13 @@ def main(path_config_files=[Path("../config/geocif.txt")]):
 
     # Experiment 3: lag_years
     logger.info("Experiment 3: lag_years")
-    parser = run(
+    parser = main(
         inputs, logger, parser, "ML", "lag_years", "int", [1, 2, 3, 4, 5]
     )
 
     # Experiment 4: lag_yield_as_feature
     logger.info("Experiment 4: lag_yield_as_feature")
-    parser = run(
+    parser = main(
         inputs,
         logger,
         parser,
@@ -87,13 +87,13 @@ def main(path_config_files=[Path("../config/geocif.txt")]):
 
     # Experiment 5: median_years
     logger.info("Experiment 5: median_years")
-    parser = run(
+    parser = main(
         inputs, logger, parser, "ML", "median_years", "int", [2, 3, 4, 5]
     )
 
     # Experiment 6: median_yield_as_feature
     logger.info("Experiment 6: median_yield_as_feature")
-    parser = run(
+    parser = main(
         inputs,
         logger,
         parser,
@@ -105,7 +105,7 @@ def main(path_config_files=[Path("../config/geocif.txt")]):
 
     # Experiment 7: analogous_year_yield_as_feature
     logger.info("Experiment 7: analogous_year_yield_as_feature")
-    parser = run(
+    parser = main(
         inputs,
         logger,
         parser,
@@ -117,10 +117,10 @@ def main(path_config_files=[Path("../config/geocif.txt")]):
 
     # Experiment 8: optimize
     logger.info("Experiment 8: optimize")
-    parser = run(
+    parser = main(
         inputs, logger, parser, "DEFAULT", "optimize", "bool", [True, False]
     )
 
 
 if __name__ == "__main__":
-    main()
+    run()

{geocif-0.1.38 → geocif-0.1.39}/geocif/geocif.py

@@ -535,7 +535,7 @@ class Geocif:
                 cei = parts[0] if parts[1].isdigit() else "_".join(parts[:2])
 
                 # Check if any element of dict_selected_features is in _t
-                for x in selected_features:
+                for x in selected_features["CEI"].values:
                     if x not in cei:
                         continue
 

{geocif-0.1.38 → geocif-0.1.39}/geocif/ml/feature_selection.py

@@ -3,6 +3,31 @@ from tqdm import tqdm
 from sklearn.ensemble import RandomForestRegressor
 
 
+def are_all_features_non_eo(features):
+    """
+    Check if all the features non eo features
+
+    Args:
+        feature:
+
+    Returns:
+
+    """
+    non_eo_features = ['Median Yield (tn per ha)',
+                       'Analogous Year',
+                       'Analogous Year Yield',
+                       'lon',
+                       'lat',
+                       't -1 Yield (tn per ha)',
+                       't -2 Yield (tn per ha)',
+                       't -3 Yield (tn per ha)',
+                       't -4 Yield (tn per ha)',
+                       't -5 Yield (tn per ha)']
+
+    # Check if all features are non-eo features, return True if they are
+    return all(feature in non_eo_features for feature in features)
+
+
 def select_features(X, y, method="RFE", min_features_to_select=3):
     """
 
@@ -29,6 +54,7 @@ def select_features(X, y, method="RFE", min_features_to_select=3):
     # selector = VarianceThreshold(threshold=scaled_data.var().mean())
     # X = selector.fit_transform(scaled_data)
     selector = None
+    X_original = X.copy()
 
     # Fill in columns with median of that column
     X = X.fillna(X.median())
@@ -46,24 +72,7 @@ def select_features(X, y, method="RFE", min_features_to_select=3):
     np.float = np.float64
     np.bool = np.bool_
 
-    if method == "SelectKBest":
-        from sklearn.feature_selection import SelectKBest, f_regression
-
-        k = 15  # Number of features to select
-        selector = SelectKBest(score_func=f_regression, k=k)
-
-        # Fit the selector to the data and transform the data to select the best features
-        try:
-            X_new = selector.fit_transform(X, y)
-        except:
-            breakpoint()
-
-        # Get the selected feature indices
-        selected_features = selector.get_support(indices=True)
-
-        # Get the selected feature names
-        selected_features = X.columns[selected_features].tolist()
-    elif method == "SHAP":
+    if method == "SHAP":
         import pandas as pd
         from catboost import CatBoostRegressor
         from fasttreeshap import TreeExplainer as FastTreeExplainer
@@ -257,6 +266,26 @@ def select_features(X, y, method="RFE", min_features_to_select=3):
         raise ValueError("Method not recognized. Use BorutaPy, Genetic, or RFE")
     # tentative_features = X.columns[selector.support_weak_].tolist()
 
+    non_eo = are_all_features_non_eo(selected_features)
+    if non_eo or method == "SelectKBest":
+        from sklearn.feature_selection import SelectKBest, f_regression
+
+        k = 15  # Number of features to select
+        selector = SelectKBest(score_func=f_regression, k=k)
+
+        # Fit the selector to the data and transform the data to select the best features
+        try:
+            X_new = selector.fit_transform(X, y)
+        except:
+            breakpoint()
+
+        # Get the selected feature indices
+        selected_features = selector.get_support(indices=True)
+
+        # Get the selected feature names
+        selected_features = X.columns[selected_features].tolist()
+
+    print(selected_features)
     # Filter the dataset for selected features
     X_filtered = X.loc[:, selected_features]
 

{geocif-0.1.38 → geocif-0.1.39/geocif.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: geocif
-Version: 0.1.38
+Version: 0.1.39
 Summary: Models to visualize and forecast crop conditions and yields
 Home-page: https://ritviksahajpal.github.io/yield_forecasting/
 Author: Ritvik Sahajpal

{geocif-0.1.38 → geocif-0.1.39}/setup.py

@@ -50,6 +50,6 @@ setup(
     test_suite="tests",
     tests_require=test_requirements,
     url="https://ritviksahajpal.github.io/yield_forecasting/",
-    version="0.1.38",
+    version="0.1.39",
     zip_safe=False,
 )