aplr 10.13.0__cp310-cp310-win_amd64.whl → 10.15.0__cp310-cp310-win_amd64.whl

aplr/aplr.py

@@ -75,6 +75,7 @@ class APLRRegressor:
         penalty_for_interactions: float = 0.0,
         max_terms: int = 0,
         ridge_penalty: float = 0.0001,
+        mean_bias_correction: bool = False,
     ):
         self.m = m
         self.v = v
@@ -122,6 +123,7 @@ class APLRRegressor:
         self.penalty_for_interactions = penalty_for_interactions
         self.max_terms = max_terms
         self.ridge_penalty = ridge_penalty
+        self.mean_bias_correction = mean_bias_correction
 
         # Creating aplr_cpp and setting parameters
         self.APLRRegressor = aplr_cpp.APLRRegressor()
@@ -183,6 +185,7 @@ class APLRRegressor:
         self.APLRRegressor.penalty_for_interactions = self.penalty_for_interactions
         self.APLRRegressor.max_terms = self.max_terms
         self.APLRRegressor.ridge_penalty = self.ridge_penalty
+        self.APLRRegressor.mean_bias_correction = self.mean_bias_correction
 
     def fit(
         self,
@@ -334,12 +337,9 @@ class APLRRegressor:
         :param path: The file path to save the plot. If empty and save is True, a default path will be used.
         """
         try:
-            import pandas as pd
             import matplotlib.pyplot as plt
         except ImportError:
-            raise ImportError(
-                "pandas and matplotlib are required for plotting. Please install them."
-            )
+            raise ImportError("matplotlib is required for plotting. Please install it.")
 
         all_affiliations = self.get_unique_term_affiliations()
         if affiliation not in all_affiliations:
@@ -362,43 +362,55 @@ class APLRRegressor:
 
         predictor_names = affiliation.split(" & ")
 
-        shape_df = pd.DataFrame(shape, columns=predictor_names + ["contribution"])
-
         is_main_effect: bool = len(predictor_indexes_used) == 1
         is_two_way_interaction: bool = len(predictor_indexes_used) == 2
 
         if is_main_effect:
             fig = plt.figure()
-            plt.plot(shape_df.iloc[:, 0], shape_df.iloc[:, 1])
-            plt.xlabel(shape_df.columns[0])
+            # Sort by predictor value for a clean line plot
+            sorted_indices = np.argsort(shape[:, 0])
+            plt.plot(shape[sorted_indices, 0], shape[sorted_indices, 1])
+            plt.xlabel(predictor_names[0])
             plt.ylabel("Contribution to linear predictor")
-            plt.title(f"Main effect of {shape_df.columns[0]}")
+            plt.title(f"Main effect of {predictor_names[0]}")
             plt.grid(True)
         elif is_two_way_interaction:
             fig = plt.figure(figsize=(8, 6))
-            pivot_table = shape_df.pivot_table(
-                index=shape_df.columns[0],
-                columns=shape_df.columns[1],
-                values=shape_df.columns[2],
-                aggfunc="mean",
-            )
+
+            # Get unique coordinates and their inverse mapping
+            y_unique, y_inv = np.unique(shape[:, 0], return_inverse=True)
+            x_unique, x_inv = np.unique(shape[:, 1], return_inverse=True)
+
+            # Create grid for sums and counts
+            grid_sums = np.zeros((len(y_unique), len(x_unique)))
+            grid_counts = np.zeros((len(y_unique), len(x_unique)))
+
+            # Populate sums and counts to later calculate the mean
+            np.add.at(grid_sums, (y_inv, x_inv), shape[:, 2])
+            np.add.at(grid_counts, (y_inv, x_inv), 1)
+
+            # Calculate mean, avoiding division by zero
+            with np.errstate(divide="ignore", invalid="ignore"):
+                pivot_table_values = np.true_divide(grid_sums, grid_counts)
+                # Where there's no data, pivot_table_values will be nan, which is fine for imshow.
+
             plt.imshow(
-                pivot_table.values,
+                pivot_table_values,
                 aspect="auto",
                 origin="lower",
                 extent=[
-                    pivot_table.columns.min(),
-                    pivot_table.columns.max(),
-                    pivot_table.index.min(),
-                    pivot_table.index.max(),
+                    x_unique.min(),
+                    x_unique.max(),
+                    y_unique.min(),
+                    y_unique.max(),
                 ],
                 cmap="Blues_r",
             )
             plt.colorbar(label="Contribution to the linear predictor")
-            plt.xlabel(shape_df.columns[1])
-            plt.ylabel(shape_df.columns[0])
+            plt.xlabel(predictor_names[1])
+            plt.ylabel(predictor_names[0])
             plt.title(
-                f"Interaction between {shape_df.columns[0]} and {shape_df.columns[1]}"
+                f"Interaction between {predictor_names[0]} and {predictor_names[1]}"
             )
         else:
             print(
@@ -407,9 +419,7 @@ class APLRRegressor:
             return
 
         if save:
-            save_path = (
-                path if path else f"shape_of_{affiliation.replace(' & ', '_')}.png"
-            )
+            save_path = path or f"shape_of_{affiliation.replace(' & ', '_')}.png"
             plt.savefig(save_path)
 
         if plot:
@@ -458,6 +468,7 @@ class APLRRegressor:
             "penalty_for_interactions": self.penalty_for_interactions,
             "max_terms": self.max_terms,
             "ridge_penalty": self.ridge_penalty,
+            "mean_bias_correction": self.mean_bias_correction,
         }
 
     # For sklearn

{aplr-10.13.0.dist-info → aplr-10.15.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: aplr
-Version: 10.13.0
+Version: 10.15.0
 Summary: Automatic Piecewise Linear Regression
 Home-page: https://github.com/ottenbreit-data-science/aplr
 Author: Mathias von Ottenbreit
@@ -15,8 +15,7 @@ Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: numpy>=1.11
 Provides-Extra: plots
-Requires-Dist: pandas; extra == "plots"
-Requires-Dist: matplotlib; extra == "plots"
+Requires-Dist: matplotlib>=3.0; extra == "plots"
 Dynamic: author
 Dynamic: author-email
 Dynamic: classifier

aplr-10.15.0.dist-info/RECORD

@@ -0,0 +1,8 @@
+aplr_cpp.cp310-win_amd64.pyd,sha256=gGn3FH67Yzpn1StyJzhWQCOEAaEpIMMWVLb4roR5yhU,636928
+aplr/__init__.py,sha256=oDFSgVytP_qQ8ilun6oHxKr-DYEeqjEQp5FciX45lls,21
+aplr/aplr.py,sha256=hGxqgMt6sUBi3P18W0p_Q88qqHsKX7wxpm1iQFw5Rig,33072
+aplr-10.15.0.dist-info/licenses/LICENSE,sha256=YOMo-RaL4P7edMZGD96-NskKpxyMZdP3-WiiMMmihNk,1134
+aplr-10.15.0.dist-info/METADATA,sha256=2JIO-RGn0S7aMiPlatABLlWNAUB0g6Lr0PHuXCjdj6M,2627
+aplr-10.15.0.dist-info/WHEEL,sha256=KUuBC6lxAbHCKilKua8R9W_TM71_-9Sg5uEP3uDWcoU,101
+aplr-10.15.0.dist-info/top_level.txt,sha256=DXVC0RIFGpzVnPeKWAZTXQdJheOEZL51Wip6Fx7zbR4,14
+aplr-10.15.0.dist-info/RECORD,,

aplr-10.13.0.dist-info/RECORD

@@ -1,8 +0,0 @@
-aplr_cpp.cp310-win_amd64.pyd,sha256=Ppxk0ZYVz2GVHsxoNCvnAvzDrRt5pS7a7XIEkvZtP7Q,632832
-aplr/__init__.py,sha256=oDFSgVytP_qQ8ilun6oHxKr-DYEeqjEQp5FciX45lls,21
-aplr/aplr.py,sha256=m1Lv4I_tWoWSUqOgXzrM4vFiPcvmGOBZ1XIw_8bkKiA,32299
-aplr-10.13.0.dist-info/licenses/LICENSE,sha256=YOMo-RaL4P7edMZGD96-NskKpxyMZdP3-WiiMMmihNk,1134
-aplr-10.13.0.dist-info/METADATA,sha256=0xM0tJ0OrzMJrarG-eN8nKFAFFrGKLbRL63kOBMNo8M,2663
-aplr-10.13.0.dist-info/WHEEL,sha256=KUuBC6lxAbHCKilKua8R9W_TM71_-9Sg5uEP3uDWcoU,101
-aplr-10.13.0.dist-info/top_level.txt,sha256=DXVC0RIFGpzVnPeKWAZTXQdJheOEZL51Wip6Fx7zbR4,14
-aplr-10.13.0.dist-info/RECORD,,