huff 1.5.1__tar.gz → 1.5.2__tar.gz

{huff-1.5.1 → huff-1.5.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: huff
-Version: 1.5.1
+Version: 1.5.2
 Summary: huff: Huff Model Market Area Analysis
 Author: Thomas Wieland
 Author-email: geowieland@googlemail.com
@@ -18,14 +18,10 @@ Thomas Wieland [ORCID](https://orcid.org/0000-0001-5168-9846) [EMail](mailto:geo
 See the /tests directory for usage examples of most of the included functions.
 
 
-## Updates v1.5.1
-- Extensions:
-  - HuffModel.ml_fit(): Fit Huff model parameters by empirical total market areas
-  - HuffModel.summary(): Goodnes-of-fit depends on fit_by
-  - load_market_areas(): Loading table with totals and including into InteractionMatrix, HuffModel and MCIModel objects
+## Updates v1.5.2
 - Bugfixes:
-  - InteractionMatrix.summary(): NoneType parameter
-  - InteractionMatrix.huff_ml_fit(): Update estimates possible even if fit algorithm did not converge
+  - HuffModel.ml_fit(): Correct values of expected T_j, corrected calculation of model fit metrices when fit_by="totals"
+  - HuffModel.ml_fit(): Check if sum of E_ij != sum of T_j
   
 
 ## Features
@@ -34,7 +30,7 @@ See the /tests directory for usage examples of most of the included functions.
   - Defining origins and destinations with weightings
   - Creating interaction matrix from origins and destinations
   - Different function types: power, exponential, logistic
-  - Huff model parameter estimation via Maximum Likelihood (ML) by probalities and customer flows
+  - Huff model parameter estimation via Maximum Likelihood (ML) by probalities, customer flows, and total market areas
   - Huff model market simulation
 - **Multiplicative Competitive Interaction Model**: 
   - Log-centering transformation of interaction matrix

{huff-1.5.1 → huff-1.5.2}/README.md

@@ -10,14 +10,10 @@ Thomas Wieland [ORCID](https://orcid.org/0000-0001-5168-9846) [EMail](mailto:geo
 See the /tests directory for usage examples of most of the included functions.
 
 
-## Updates v1.5.1
-- Extensions:
-  - HuffModel.ml_fit(): Fit Huff model parameters by empirical total market areas
-  - HuffModel.summary(): Goodnes-of-fit depends on fit_by
-  - load_market_areas(): Loading table with totals and including into InteractionMatrix, HuffModel and MCIModel objects
+## Updates v1.5.2
 - Bugfixes:
-  - InteractionMatrix.summary(): NoneType parameter
-  - InteractionMatrix.huff_ml_fit(): Update estimates possible even if fit algorithm did not converge
+  - HuffModel.ml_fit(): Correct values of expected T_j, corrected calculation of model fit metrices when fit_by="totals"
+  - HuffModel.ml_fit(): Check if sum of E_ij != sum of T_j
   
 
 ## Features
@@ -26,7 +22,7 @@ See the /tests directory for usage examples of most of the included functions.
   - Defining origins and destinations with weightings
   - Creating interaction matrix from origins and destinations
   - Different function types: power, exponential, logistic
-  - Huff model parameter estimation via Maximum Likelihood (ML) by probalities and customer flows
+  - Huff model parameter estimation via Maximum Likelihood (ML) by probalities, customer flows, and total market areas
   - Huff model market simulation
 - **Multiplicative Competitive Interaction Model**: 
   - Log-centering transformation of interaction matrix

{huff-1.5.1 → huff-1.5.2}/huff/models.py

@@ -4,8 +4,8 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     1.5.1
-# Last update: 2025-07-01 17:10
+# Version:     1.5.2
+# Last update: 2025-07-02 21:09
 # Copyright (c) 2025 Thomas Wieland
 #-----------------------------------------------------------------------
 
@@ -1292,21 +1292,6 @@ class InteractionMatrix:
             print(f"Optimization via {method} algorithm succeeded with parameters: {', '.join(str(round(par, 3)) for par in param_results)}.")
    
         else:
-
-            # param_gamma = None
-            # param_lambda = None
-            
-            # supply_locations_metadata["weighting"][0]["param"] = param_gamma            
-            
-            # if customer_origins_metadata["weighting"][0]["func"] == "logistic":
-                               
-            #     param_lambda2 = None                
-            #     customer_origins_metadata["weighting"][0]["param"][0] = param_lambda
-            #     customer_origins_metadata["weighting"][0]["param"][1] = param_lambda2
-            
-            # else:
-                
-            #     customer_origins_metadata["weighting"][0]["param"] = param_lambda
                 
             print(f"Optimiziation via {method} algorithm failed with error message: '{ml_result.message}'. See https://docs.scipy.org/doc/scipy/reference/generated/scipy.optimize.minimize.html for all available algorithms.")
 
@@ -1316,20 +1301,28 @@ class InteractionMatrix:
         if update_estimates:
 
             if "p_ij_emp" not in self.interaction_matrix_df.columns:
+                
                 self.interaction_matrix_df["p_ij_emp"] = self.interaction_matrix_df["p_ij"]
-                print("Probabilties in interaction matrix are treated as empirical probabilties")
+                
+                print("NOTE: Probabilities in interaction matrix are treated as empirical probabilities")
+                
             else:
-                print("Interaction matrix contains empirical probabilties")
+                
+                print("NOTE: Interaction matrix contains empirical probabilities")
 
             if "E_ij_emp" not in self.interaction_matrix_df.columns:
+                
                 self.interaction_matrix_df["E_ij_emp"] = self.interaction_matrix_df["E_ij"]
-                print("Customer interactions in interaction matrix are treated as empirical interactions")
+                
+                print("NOTE: Customer interactions in interaction matrix are treated as empirical interactions")
+                
             else:
-                print("Interaction matrix contains empirical customer interactions")
+                
+                print("NOTE: Interaction matrix contains empirical customer interactions")
 
             if np.isnan(ml_result.x).any():
 
-                print("No update of estimates because fit parameters contain NaN")
+                print("WARNING: No update of estimates because fit parameters contain NaN")
 
                 update_estimates = False
 
@@ -1391,7 +1384,7 @@ class InteractionMatrix:
      
         if "transport_costs" not in interaction_matrix_metadata:
             
-            print("New destination(s) included. No transport costs calculation because not defined in original interaction matrix.")
+            print("WARNING: New destination(s) included. No transport costs calculation because not defined in original interaction matrix.")
             
             interaction_matrix_df = pd.concat(
                 [
@@ -1625,7 +1618,6 @@ class HuffModel:
                 print ("Goodness-of-fit for " + interaction_matrix_metadata["fit"]["fit_by"])
                         
                 print("Sum of squared residuals       ", round(huff_modelfit[1]["SQR"], 2))
-                print("Sum of squares                 ", round(huff_modelfit[1]["SQT"], 2))
                 print("R-squared                      ", round(huff_modelfit[1]["Rsq"], 2))
                 print("Mean squared error             ", round(huff_modelfit[1]["MSE"], 2))
                 print("Root mean squared error        ", round(huff_modelfit[1]["RMSE"], 2))
@@ -1853,7 +1845,8 @@ class HuffModel:
         bounds: list = [(0.5, 1), (-3, -1)],        
         constraints: list = [],
         fit_by = "probabilities",
-        update_estimates: bool = True
+        update_estimates: bool = True,
+        check_numbers: bool = True
         ):
         
         if fit_by in ["probabilities", "flows"]:
@@ -1869,6 +1862,16 @@ class HuffModel:
         
         elif fit_by == "totals":
             
+            if check_numbers:
+                
+                market_areas_df = self.market_areas_df
+                interaction_matrix_df = self.get_interaction_matrix_df()
+                T_j_market_areas_df = sum(market_areas_df["T_j"])
+                T_j_interaction_matrix_df = sum(interaction_matrix_df["E_ij"])
+                
+                if T_j_market_areas_df != T_j_interaction_matrix_df:
+                    print("WARNING: Sum of total market areas (" + str(int(T_j_market_areas_df)) + ") is not equal to sum of customer flows (" + str(int(T_j_interaction_matrix_df)) + ")")
+            
             supply_locations = self.interaction_matrix.supply_locations
             supply_locations_metadata = supply_locations.get_metadata()
         
@@ -1986,12 +1989,13 @@ class HuffModel:
                     update_estimates = False
 
                 else:
-
+                                       
                     self.interaction_matrix.utility()
                     self.interaction_matrix.probabilities()
-                    self.interaction_matrix.flows()
+                    self.interaction_matrix.flows()                    
                     
-                    self.interaction_matrix.marketareas()
+                    huff_model_new_marketareas = self.interaction_matrix.marketareas()
+                    self.market_areas_df["T_j"] = huff_model_new_marketareas.get_market_areas_df()["T_j"]
             
             self.interaction_matrix.metadata["fit"] = {
                 "function": "huff_ml_fit",
@@ -2042,12 +2046,12 @@ class HuffModel:
                     
                 except:
                     
-                    print("Goodness-of-fit metrics could not be calculated due to NaN values.")
+                    print("WARNING: Goodness-of-fit metrics could not be calculated due to NaN values.")
                     return None
             
             else:
                 
-                print("Goodness-of-fit metrics could not be calculated. No empirical values of probabilities in interaction matrix.")
+                print("WARNING: Goodness-of-fit metrics could not be calculated. No empirical values of probabilities in interaction matrix.")
 
                 return None
             
@@ -2069,12 +2073,12 @@ class HuffModel:
                     
                 except:
                     
-                    print("Goodness-of-fit metrics could not be calculated due to NaN values.")
+                    print("WARNING: Goodness-of-fit metrics could not be calculated due to NaN values.")
                     return None
             
             else:
                 
-                print("Goodness-of-fit metrics could not be calculated. No empirical values of customer flows in interaction matrix.")
+                print("WARNING: Goodness-of-fit metrics could not be calculated. No empirical values of customer flows in interaction matrix.")
 
                 return None
             
@@ -2095,12 +2099,12 @@ class HuffModel:
                     
                 except:
                     
-                    print("Goodness-of-fit metrics could not be calculated due to NaN values.")
+                    print("WARNING: Goodness-of-fit metrics could not be calculated due to NaN values.")
                     return None
                 
             else:
                 
-                print("Goodness-of-fit metrics could not be calculated. No empirical values of T_j in market areas data.")
+                print("WARNING: Goodness-of-fit metrics could not be calculated. No empirical values of T_j in market areas data.")
 
                 return None
 
@@ -2174,7 +2178,7 @@ class MCIModel:
                 
             except:
                 
-                print("Goodness-of-fit metrics could not be calculated due to NaN values.")
+                print("WARNING: Goodness-of-fit metrics could not be calculated due to NaN values.")
                 return None
         
         else:
@@ -2224,7 +2228,6 @@ class MCIModel:
             print ("Goodness-of-fit for probabilities")
                     
             print("Sum of squared residuals       ", round(mci_modelfit[1]["SQR"], 2))
-            print("Sum of squares                 ", round(mci_modelfit[1]["SQT"], 2))
             print("R-squared                      ", round(mci_modelfit[1]["Rsq"], 2))
             print("Mean squared error             ", round(mci_modelfit[1]["MSE"], 2))
             print("Root mean squared error        ", round(mci_modelfit[1]["RMSE"], 2))
@@ -3129,7 +3132,7 @@ def modelfit(
 
         if len(obs_exp_clean) < len(observed) or len(obs_exp_clean) < len(expected):
             if verbose:
-                print("Vectors 'observed' and/or 'expected' contain zeros which are dropped.")
+                print("NOTE: Vectors 'observed' and/or 'expected' contain zeros which are dropped.")
         
         observed = obs_exp_clean["observed"].to_numpy()
         expected = obs_exp_clean["expected"].to_numpy()

{huff-1.5.1 → huff-1.5.2}/huff/tests/tests_huff.py

@@ -4,8 +4,8 @@
 # Author:      Thomas Wieland 
 #              ORCID: 0000-0001-5168-9846
 #              mail: geowieland@googlemail.com              
-# Version:     1.5.1
-# Last update: 2025-07-01 17:10
+# Version:     1.5.2
+# Last update: 2025-07-02 21:10
 # Copyright (c) 2025 Thomas Wieland
 #-----------------------------------------------------------------------
 
@@ -354,6 +354,9 @@ huff_model_fit3 = huff_model_fit2.ml_fit(
 huff_model_fit3.summary()
 # Show summary
 
+print(huff_model_fit3.get_market_areas_df())
+# Show market areas df
+
 
 # Buffer analysis:
 

{huff-1.5.1 → huff-1.5.2}/huff.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: huff
-Version: 1.5.1
+Version: 1.5.2
 Summary: huff: Huff Model Market Area Analysis
 Author: Thomas Wieland
 Author-email: geowieland@googlemail.com
@@ -18,14 +18,10 @@ Thomas Wieland [ORCID](https://orcid.org/0000-0001-5168-9846) [EMail](mailto:geo
 See the /tests directory for usage examples of most of the included functions.
 
 
-## Updates v1.5.1
-- Extensions:
-  - HuffModel.ml_fit(): Fit Huff model parameters by empirical total market areas
-  - HuffModel.summary(): Goodnes-of-fit depends on fit_by
-  - load_market_areas(): Loading table with totals and including into InteractionMatrix, HuffModel and MCIModel objects
+## Updates v1.5.2
 - Bugfixes:
-  - InteractionMatrix.summary(): NoneType parameter
-  - InteractionMatrix.huff_ml_fit(): Update estimates possible even if fit algorithm did not converge
+  - HuffModel.ml_fit(): Correct values of expected T_j, corrected calculation of model fit metrices when fit_by="totals"
+  - HuffModel.ml_fit(): Check if sum of E_ij != sum of T_j
   
 
 ## Features
@@ -34,7 +30,7 @@ See the /tests directory for usage examples of most of the included functions.
   - Defining origins and destinations with weightings
   - Creating interaction matrix from origins and destinations
   - Different function types: power, exponential, logistic
-  - Huff model parameter estimation via Maximum Likelihood (ML) by probalities and customer flows
+  - Huff model parameter estimation via Maximum Likelihood (ML) by probalities, customer flows, and total market areas
   - Huff model market simulation
 - **Multiplicative Competitive Interaction Model**: 
   - Log-centering transformation of interaction matrix

{huff-1.5.1 → huff-1.5.2}/setup.py

@@ -7,7 +7,7 @@ def read_README():
     
 setup(
     name='huff',
-    version='1.5.1',
+    version='1.5.2',
     description='huff: Huff Model Market Area Analysis',
     packages=find_packages(include=["huff", "huff.tests"]),
     include_package_data=True,