Update app.py

app.py

@@ -5,6 +5,7 @@ import matplotlib.pyplot as plt
 from wordcloud import WordCloud
 from src.feature_extraction import FeatureExtractor
 import os
+import numpy as np
 
 # Load Model
 MODEL_PATH = 'best_svm_model.joblib'
@@ -21,42 +22,54 @@ feature_extractor = FeatureExtractor()
 def predict_summary(title, text):
     if not title or not text:
         return "Please provide both Title and Content.", pd.DataFrame()
-
+    
     # 1. Extract Features
     features_list = feature_extractor.calculate_features(title, text)
     if not features_list:
         return "Could not extract sentences.", pd.DataFrame()
-
+    
     df_features = pd.DataFrame(features_list)
-
+    
     # 2. Prepare X for prediction (ensure correct column order matching training)
-    # The columns expected by the model are likely:
     feature_cols = ['f1_lead', 'f2_judul_sim', 'f3_freq_word', 'f4_sim_sent', 'f5_len_norm', 'f6_overlap']
     X_pred = df_features[feature_cols]
-
+    
     # 3. Predict
     if model:
         predictions = model.predict(X_pred)
+        # Get probabilities for fallback
+        if hasattr(model, "predict_proba"):
+            probs = model.predict_proba(X_pred)[:, 1] # Probability of class 1
+        else:
+            # If model doesn't support probability (e.g. LinearSVC), use decision_function
+            probs = model.decision_function(X_pred)
     else:
         return "Model not loaded.", pd.DataFrame()
-
+    
     df_features['prediction'] = predictions
-
+    df_features['probability'] = probs
+    
     # 4. Generate Summary Text
     summary_sentences = df_features[df_features['prediction'] == 1]['teks_kalimat'].tolist()
+    
+    # Fallback Logic: If no sentences selected
     if not summary_sentences:
-        summary_text = "No sentences were selected as summary."
-    else:
-        summary_text = " ".join(summary_sentences)
-
+        # Select the sentence with the highest probability
+        best_idx = np.argmax(probs)
+        summary_sentences = [df_features.iloc[best_idx]['teks_kalimat']]
+        # Update prediction for display
+        df_features.at[best_idx, 'prediction'] = 1
+        
+    summary_text = " ".join(summary_sentences)
+        
     # Return Summary and DataFrame
-    return summary_text, df_features[['teks_kalimat', 'prediction'] + feature_cols]
+    return summary_text, df_features[['teks_kalimat', 'prediction', 'probability'] + feature_cols]
 
 
 def process_batch_file(file_obj):
     if file_obj is None:
         return None, None, None, None
-
+    
     # Read File
     try:
         if file_obj.name.endswith('.csv'):
@@ -68,39 +81,42 @@ def process_batch_file(file_obj):
     except Exception as e:
         return None, None, None, None
 
-    # Check columns
-    # We expect 'title' and 'teks_kalimat' columns.
-    # WAIT: The training data `korpus_mentah_new.csv` has 'teks_kalimat' as sentences.
-    # But batch input usually implies articles.
-    # Let's assume input has 'title' and 'content' (full text).
-
     required_cols = ['title', 'content']
     if not all(col in df.columns for col in required_cols):
-        # Fallback for user friendliness if they upload something else
         return None, None, None, None
 
     all_results = []
-
+    
     for idx, row in df.iterrows():
         title = str(row['title'])
         content = str(row['content'])
-
+        
         features_list = feature_extractor.calculate_features(title, content)
         if not features_list:
             continue
-
+            
         df_feat = pd.DataFrame(features_list)
         X_pred = df_feat[['f1_lead', 'f2_judul_sim', 'f3_freq_word', 'f4_sim_sent', 'f5_len_norm', 'f6_overlap']]
-
+        
         if model:
             preds = model.predict(X_pred)
+            if hasattr(model, "predict_proba"):
+                probs = model.predict_proba(X_pred)[:, 1]
+            else:
+                probs = model.decision_function(X_pred)
+                
             df_feat['prediction'] = preds
-            df_feat['article_id'] = idx
-
-            # Reconstruct Summary
+            df_feat['probability'] = probs
+            
+            # Reconstruct Summary with Fallback
             summary_sents = df_feat[df_feat['prediction'] == 1]['teks_kalimat'].tolist()
+            
+            if not summary_sents:
+                best_idx = np.argmax(probs)
+                summary_sents = [df_feat.iloc[best_idx]['teks_kalimat']]
+            
             summary_text = " ".join(summary_sents)
-
+            
             # Store result for download (Article level)
             all_results.append({
                 'article_id': idx,
@@ -111,28 +127,24 @@ def process_batch_file(file_obj):
                 'summary_length': len(summary_text.split()),
                 'compression_ratio': len(summary_text.split()) / len(content.split()) if len(content.split()) > 0 else 0
             })
-
+            
     df_results = pd.DataFrame(all_results)
     output_csv = "batch_results.csv"
     df_results.to_csv(output_csv, index=False)
-
+    
     # Analysis
-    # 1. Summary vs Non-Summary Distribution (sentence level - conceptually, but here we have article level ratios)
-    # Let's plot Average Compression Ratio
     fig1, ax1 = plt.subplots()
     ax1.hist(df_results['compression_ratio'], bins=10, color='skyblue', edgecolor='black')
     ax1.set_title('Distribution of Compression Ratios')
     ax1.set_xlabel('Ratio (Summary Words / Original Words)')
     ax1.set_ylabel('Frequency')
-
-    # 2. Length Comparison
+    
     fig2, ax2 = plt.subplots()
     ax2.scatter(df_results['original_length'], df_results['summary_length'], alpha=0.5)
     ax2.set_title('Original Length vs Summary Length')
     ax2.set_xlabel('Original Length (words)')
     ax2.set_ylabel('Summary Length (words)')
-
-    # 3. Word Cloud of all summaries
+    
     all_summary_text = " ".join(df_results['generated_summary'].tolist())
     if all_summary_text.strip():
         wordcloud = WordCloud(width=800, height=400, background_color='white').generate(all_summary_text)
@@ -153,18 +165,18 @@ def get_template():
 # Gradio Interface
 with gr.Blocks(title="Balinese News Summarizer") as demo:
     gr.Markdown("# 📰 Balinese News Summarizer (SVM)")
-
+    
     with gr.Tab("📝 Manual Input"):
         with gr.Row():
             with gr.Column():
                 input_title = gr.Textbox(label="News Title", placeholder="Enter title here...")
                 input_content = gr.Textbox(label="News Content", placeholder="Enter full news text here...", lines=10)
                 btn_predict = gr.Button("Summarize", variant="primary")
-
+            
             with gr.Column():
                 output_summary = gr.Textbox(label="Generated Summary", lines=5, interactive=False)
-                output_df = gr.Dataframe(label="Sentence Analysis", headers=['teks_kalimat', 'prediction', 'f1', 'f2', 'f3', 'f4', 'f5', 'f6'])
-
+                output_df = gr.Dataframe(label="Sentence Analysis", headers=['teks_kalimat', 'prediction', 'probability', 'f1', 'f2', 'f3', 'f4', 'f5', 'f6'])
+        
         btn_predict.click(fn=predict_summary, inputs=[input_title, input_content], outputs=[output_summary, output_df])
 
     with gr.Tab("📂 Batch Analysis"):
@@ -172,21 +184,21 @@ with gr.Blocks(title="Balinese News Summarizer") as demo:
         with gr.Row():
             file_upload = gr.File(label="Upload File", file_types=['.csv', '.xlsx'])
             btn_template = gr.Button("Download Template")
-            template_file = gr.File(label="Template", interactive=False, visible=False)
-
-            btn_template.click(fn=get_template, inputs=None, outputs=file_upload) # Reuse file upload component to serve download or separate?
-            # Better to use a separate DownloadButton if available in this version, or File output.
-            # Using File output for template download:
-
+            
+            # Using a workaround for template download if needed, or just file output
+            # Assuming file_upload can display the output file as well
+            
+            btn_template.click(fn=get_template, inputs=None, outputs=file_upload)
+            
         btn_process = gr.Button("Process Batch", variant="primary")
-
+        
         with gr.Row():
             download_result = gr.File(label="Download Results")
-
+        
         with gr.Row():
             plot1 = gr.Plot(label="Compression Ratio Distribution")
             plot2 = gr.Plot(label="Length Comparison")
-
+        
         with gr.Row():
             plot3 = gr.Plot(label="Word Cloud")