Added video sections

.gitignore

@@ -127,3 +127,8 @@ dmypy.json
 
 # Pyre type checker
 .pyre/
+Project02/.idea/inspectionProfiles/profiles_settings.xml
+Project02/.idea/inspectionProfiles/Project_Default.xml
+Project02/.idea/misc.xml
+*.iml
+*.xml

Project02/source/Evaluation/CrossValidation.py

@@ -178,6 +178,7 @@ def validate(
         stratify: bool = False,
         model_params=[],
         predict_params=[],
+        train_params=None,
     ) -> float:
         """Perform cross-validation using k=num_folds folds
 
@@ -209,6 +210,9 @@ def validate(
             # Instantiate the model
             self.algorithm = model(training_data, self.target_feature, *model_params)
 
+            # Train the model
+            if train_params is not None:
+                self.algorithm.train(*train_params)
             fold_results = self.calculate_results_for_fold(self.algorithm, test_data, predict_params=predict_params)
 
             overall_results.append(fold_results)

Project02/source/Utilities/Preprocess.py

@@ -1,11 +1,12 @@
 import math
+
 from typing import List, Callable
 import warnings
 warnings.simplefilter(action='ignore', category=FutureWarning)
 import pandas as pd
 
-Converter = dict[str, Callable]
-Bins = dict[str, int]
+Converter = Dict[str, Callable]
+Bins = Dict[str, int]
 
 
 class Preprocessor:

Project02/video.py

@@ -270,7 +270,7 @@ def demonstrate_clustering(k):
 
     # Perform one iteration of clustering
     kmeans._initialize_clusters()
-    kmeans._update_samples(False)
+    kmeans._update_samples()
 
     # Print clusters
     for index, cluster in enumerate(kmeans.clusters):
@@ -313,13 +313,14 @@ def demonstrate_average_classification_performance():
 
     # # Perform 10-fold CV for E-NN
     print("\nRunning on E-NN")
-    # Define hyper-parameters
+    # # Define hyper-parameters
     num_neighbors = 5
 
-    eknn_cv = CrossValidation(training_data, target_feature="class")
-    eknn_results = eknn_cv.validate(EditedKNN, 10, True, predict_params=[num_neighbors, False, 0.0])
 
-    # Convert the results to performance metrics
+    eknn_cv = CrossValidation(training_data.copy(), target_feature="class")
+    eknn_results = eknn_cv.validate(EditedKNN, 10, True, predict_params=[num_neighbors, False, 0.0], train_params=[num_neighbors])
+
+    # # Convert the results to performance metrics
     eknn_measures = ExperimentHelper.convert_classification_results_to_measures(
         eknn_results, classification_levels
     )
@@ -335,7 +336,7 @@ def demonstrate_average_classification_performance():
     num_neighbors = 5
     num_clusters = 5
 
-    kmeans_cv = CrossValidation(training_data, target_feature="class")
+    kmeans_cv = CrossValidation(training_data.copy(), target_feature="class")
     kmeans_results = kmeans_cv.validate(
         KNN, 10, True, model_params=[False, None, num_clusters], predict_params=[num_neighbors]
     )
@@ -457,6 +458,7 @@ def project_demonstration():
     demonstrate_average_classification_performance()
     input("")
 
+
     # Display average performance across ten folds for k-NN, k-Means, and ENN on a regression data set
     demonstrate_average_regression_performance()