Update app.py

app.py

@@ -250,19 +250,14 @@ tree_models = ["RandomForest", "DecisionTree"]
 non_tree_models = ["KNN", "SVM", "LogisticRegression"]
 #models = ["RandomForest", "DecisionTree", "KNN", "SVM", "LogisticRegression"]
 
-print("="*80)
-print("REPLICATION: Uddin & Lu (2024) - Pairwise Model Comparisons")
-print("="*80)
+
 
 # Store all results
 all_results = []
 
 # For each metric
 for metric in ["accuracy", "precision", "recall", "f1_score"]:
-    print(f"\n({metric.upper()})")
-    print("-"*80)
-    print(f"{'#':<3} {'Tree Model':<20} {'Non-Tree Model':<20} {'Mean 1':<10} {'Mean 2':<10} {'t':<8} {'p-value':<10} {'Sig.'}")
-    print("-"*80)
+    
 
     comparison_num = 1
 
@@ -320,19 +315,12 @@ for metric in ["accuracy", "precision", "recall", "f1_score"]:
             comparison_num += 1
         models = other_models.copy()
 
-# Summary
-print("\n" + "="*80)
-print("SUMMARY")
-print("="*80)
 results_df = pd.DataFrame(all_results)
 significant_count = (results_df['p_value'] < 0.05).sum()
 total_count = len(results_df)
-print(f"\nSignificant comparisons (p < 0.05): {significant_count}/{total_count}")
-print(f"Tree models won in: {(results_df['tree_mean'] > results_df['non_tree_mean']).sum()} comparisons")
 
 # Save detailed results
 #results_df.to_csv('pairwise_comparison_results.csv', index=False)
-print("\nDetailed results saved to: pairwise_comparison_results.csv")
 import pandas as pd
 from scipy.stats import ttest_rel
 sig1 = {}
@@ -345,20 +333,12 @@ for key in list(DATASET_CATEGORIES.keys()):
   non_tree_models = ["KNN", "SVM", "LogisticRegression"]
   #models = ["RandomForest", "DecisionTree", "KNN", "SVM", "LogisticRegression"]
   models = ["RandomForest", "DecisionTree", "KNN", "SVM", "LogisticRegression"]
-  print("="*80)
-  print("REPLICATION: Uddin & Lu (2024) - Pairwise Model Comparisons")
-  print("="*80)
 
   # Store all results
   all_results = []
 
   # For each metric
   for metric in ["accuracy", "precision", "recall", "f1_score"]:
-      print(f"\n({metric.upper()})")
-      print("-"*80)
-      print(f"{'#':<3} {'Tree Model':<20} {'Non-Tree Model':<20} {'Mean 1':<10} {'Mean 2':<10} {'t':<8} {'p-value':<10} {'Sig.'}")
-      print("-"*80)
-
       comparison_num = 1
 
       models = ["RandomForest", "DecisionTree", "KNN", "SVM", "LogisticRegression"]
@@ -397,7 +377,6 @@ for key in list(DATASET_CATEGORIES.keys()):
 
               sig = "< 0.001" if p_val < 0.001 else f"{p_val:.3f}"
 
-              print(f"{comparison_num:<3} {tree_model:<20} {non_tree_model:<20} {mean1:<10.5f} {mean2:<10.5f} {t_stat:<8.2f} {sig:<10} {'True' if p_val < 0.05 else 'False'}")
 
               all_results.append({
                   'metric': metric,
@@ -527,7 +506,7 @@ def compare_groups(data_choice, model1, model2):
   return fig, "\n".join(p_values_text)
 import gradio as gr
 models = ["RandomForest", "DecisionTree", "KNN", "SVM", "LogisticRegression"]
-with gr.Blocks(theme=gr.themes.Soft()) as demo:
+with gr.Blocks() as demo:
   with gr.Tabs() as tabs:
     with gr.TabItem("Compaes groups of datasets"):
       with gr.Column():