Update app.py

app.py

@@ -93,12 +93,11 @@ def calculate_shap_values(model, x_tensor):
         background = torch.zeros((300, x_tensor.shape[1]), device=device)
         explainer = shap.DeepExplainer(model, background)
         shap_values_all = explainer.shap_values(x_tensor)
-        # Get SHAP values for human class (index 1)
+        # For binary classification, use the second output and then the first sample
         shap_values = shap_values_all[1][0]
     except Exception as e:
         print(f"DeepExplainer failed, falling back to KernelExplainer: {str(e)}")
         
-        # Define a wrapper that ensures input is a 2D tensor
         def model_predict(x):
             if not isinstance(x, np.ndarray):
                 x = np.array(x)
@@ -107,17 +106,19 @@ def calculate_shap_values(model, x_tensor):
             with torch.no_grad():
                 tensor_x = torch.tensor(x, dtype=torch.float, device=device)
                 output = model(tensor_x)
-                probs = torch.softmax(output, dim=1)[:, 1]  # Human probability
+                probs = torch.softmax(output, dim=1)[:, 1]
                 return probs.cpu().numpy()
         
-        # Use a numpy background with 300 samples
+        # Use a numpy background for KernelExplainer
         background = np.zeros((300, x_tensor.shape[1]))
         explainer = shap.KernelExplainer(model_predict, background)
         x_numpy = x_tensor.cpu().numpy()
-        # Increase nsamples to 1000 to provide enough data for regression
         shap_values = explainer.shap_values(x_numpy, nsamples=1000)
+        # If KernelExplainer returns a list, take its first element.
+        if isinstance(shap_values, list):
+            shap_values = shap_values[0]
     
-    # Get human probability from the model prediction
+    # Get human probability from model prediction
     with torch.no_grad():
         output = model(x_tensor)
         probs = torch.softmax(output, dim=1)
@@ -125,6 +126,7 @@ def calculate_shap_values(model, x_tensor):
     
     return np.array(shap_values), prob_human
 
+
 ###############################################################################
 # 4. PER-BASE SHAP AGGREGATION
 ###############################################################################