Update app.py

app.py

@@ -84,50 +84,25 @@ def sequence_to_kmer_vector(sequence: str, k: int = 4) -> np.ndarray:
 ###############################################################################
 # 3. SHAP-VALUE (ABLATION) CALCULATION
 ###############################################################################
+
+
 def calculate_shap_values(model, x_tensor):
     model.eval()
-    device = next(model.parameters()).device
-
-    try:
-        # Create background as a torch tensor (using zeros may be acceptable for DeepExplainer)
-        background = torch.zeros((300, x_tensor.shape[1]), device=device)
-        explainer = shap.DeepExplainer(model, background)
-        shap_values_all = explainer.shap_values(x_tensor)
-        # For binary classification, get SHAP for class 1 and 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 proper input shape and conversion to tensor
-        def model_predict(x):
-            if not isinstance(x, np.ndarray):
-                x = np.array(x)
-            if x.ndim == 1:
-                x = np.expand_dims(x, axis=0)
-            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
-                return probs.cpu().numpy()
-
-        # Instead of using zeros as background, use the input sample repeated 300 times.
-        x_numpy = x_tensor.cpu().numpy()
-        background = np.repeat(x_numpy, 300, axis=0)
-
-        explainer = shap.KernelExplainer(model_predict, background)
-        # Increase nsamples for a more robust estimate.
-        shap_values = explainer.shap_values(x_numpy, nsamples=1000)
-        # If a list is returned, select the first element.
-        if isinstance(shap_values, list):
-            shap_values = shap_values[0]
-
-    # Get the human probability from the model output.
     with torch.no_grad():
-        output = model(x_tensor)
-        probs = torch.softmax(output, dim=1)
-        prob_human = probs[0, 1].item()
-
-    return np.array(shap_values), prob_human
+        baseline_output = model(x_tensor)
+        baseline_probs = torch.softmax(baseline_output, dim=1)
+        baseline_prob = baseline_probs[0, 1].item()  # Prob of 'human'
+        shap_values = []
+        x_zeroed = x_tensor.clone()
+        for i in range(x_tensor.shape[1]):
+            original_val = x_zeroed[0, i].item()
+            x_zeroed[0, i] = 0.0
+            output = model(x_zeroed)
+            probs = torch.softmax(output, dim=1)
+            prob = probs[0, 1].item()
+            shap_values.append(baseline_prob - prob)
+            x_zeroed[0, i] = original_val
+    return np.array(shap_values), baseline_prob
 
 
 ###############################################################################