Update app.py

app.py

@@ -195,21 +195,37 @@ def predict(file_obj):
         )
         explanation.expected_value = 0.5  # Start from neutral prediction
 
-        # Create waterfall plot
-        plt.figure(figsize=(10, 6))
-        fig = shap.plots._waterfall.waterfall_legacy(
-            explanation,
-            show=False,
-            max_display=11  # Show all features including "Others"
-        )
-        plt.title(f"Feature contributions to human probability (final prob: {human_prob:.3f})")
-        
-        # Save plot
-        buf = io.BytesIO()
-        plt.savefig(buf, format='png', bbox_inches='tight', dpi=300)
-        buf.seek(0)
-        plot_image = Image.open(buf)
-        plt.close()
+        # Calculate step-by-step probabilities
+        step_probs = []
+        current_prob = 0.5  # Start at neutral
+        step_probs.append({"step": "Start", "probability": current_prob, "kmer": "Initial", "change": 0})
+        
+        # Process each k-mer contribution
+        for i, kmer in enumerate(important_kmers, 1):
+            change = kmer['importance']
+            current_prob += change
+            step_probs.append({
+                "step": str(i),
+                "probability": current_prob,
+                "kmer": kmer['kmer'],
+                "change": change
+            })
+        
+        # Add final "Others" contribution
+        current_prob += others_sum
+        step_probs.append({
+            "step": "Others",
+            "probability": current_prob,
+            "kmer": "Others",
+            "change": others_sum
+        })
+        
+        # Convert to JSON for React
+        steps_json = json.dumps(step_probs, indent=2)
+        print(f"Steps data: {steps_json}")  # For debugging
+        
+        # Create visualization component
+        plot_image = None  # We'll use the React component instead
 
         # Calculate final probabilities
         with torch.no_grad():