Update app.py

app.py

@@ -203,12 +203,12 @@ def find_last_dense_layer(model):
 
 # ---- GRAD-CAM ----
 def make_gradcam(image_pil, model, last_conv_layer_name, class_index, progress=None):
-    if model is None: return np.array(image_pil)
-    start_time = time.time()
+    if model is None: 
+        return np.array(image_pil)
     try:
         _update_progress(progress, 0, desc="Préparation de l'image...")
-        input_size = model.input_shape[1:3]
 
+        input_size = model.input_shape[1:3]
         if 'xception' in model.name.lower(): 
             preprocessor = preprocess_xception
         elif 'resnet50' in model.name.lower(): 
@@ -222,7 +222,7 @@ def make_gradcam(image_pil, model, last_conv_layer_name, class_index, progress=N
         img_resized = cv2.resize(img_np, input_size)
         img_array_preprocessed = preprocessor(np.expand_dims(img_resized, axis=0))
 
-        _update_progress(progress, 25, desc="Calcul des gradients...", sleep=0.005)
+        _update_progress(progress, 20, desc="Calcul des gradients...")
 
         try:
             conv_layer = model.get_layer(last_conv_layer_name)
@@ -230,7 +230,6 @@ def make_gradcam(image_pil, model, last_conv_layer_name, class_index, progress=N
             return img_resized
 
         dense_layer = find_last_dense_layer(model)
-        
         grad_model = Model(model.inputs, [conv_layer.output, model.output])
         input_name = get_primary_input_name(model)
         input_for_model = {input_name: img_array_preprocessed}
@@ -242,15 +241,16 @@ def make_gradcam(image_pil, model, last_conv_layer_name, class_index, progress=N
             class_channel = preds[:, int(class_index)]
 
         grads = tape.gradient(class_channel, last_conv_layer_output)
-
         if grads is None:
             return img_resized
 
-        _update_progress(progress, 50, desc="Génération de la heatmap...", sleep=0.005)
+        _update_progress(progress, 40, desc="Pooling des gradients...")
 
         pooled_grads = tf.reduce_mean(grads, axis=(0, 1, 2))
         last_conv_layer_output = last_conv_layer_output[0]
-        
+
+        _update_progress(progress, 55, desc="Construction de la heatmap...")
+
         heatmap = last_conv_layer_output @ pooled_grads[..., tf.newaxis]
         heatmap = tf.squeeze(heatmap)
         heatmap = tf.maximum(heatmap, 0)
@@ -260,26 +260,30 @@ def make_gradcam(image_pil, model, last_conv_layer_name, class_index, progress=N
             heatmap = tf.ones_like(heatmap) * 0.5
         else:
             heatmap = heatmap / max_val
-        
-        heatmap_np = heatmap.numpy()
 
-        _update_progress(progress, 75, desc="Application de la superposition...", sleep=0.005)
+        _update_progress(progress, 70, desc="Conversion NumPy...")
 
+        heatmap_np = heatmap.numpy()
         heatmap_np = np.clip(heatmap_np.astype(np.float32), 0, 1)
+
+        _update_progress(progress, 80, desc="Application du colormap...")
+
         heatmap_resized = cv2.resize(heatmap_np, (img_resized.shape[1], img_resized.shape[0]))
         heatmap_uint8 = np.uint8(255 * heatmap_resized)
         heatmap_colored = cv2.applyColorMap(heatmap_uint8, cv2.COLORMAP_JET)
-        
+
         img_bgr = cv2.cvtColor(img_resized, cv2.COLOR_RGB2BGR)
         superimposed_img = cv2.addWeighted(img_bgr, 0.6, heatmap_colored, 0.4, 0)
 
-        _update_progress(progress, 100, desc="✅ Grad-CAM terminé !", animate=True)
+        _update_progress(progress, 100, desc="✅ Grad-CAM terminé !")
 
         return cv2.cvtColor(superimposed_img, cv2.COLOR_BGR2RGB)
+
     except Exception as e:
         import traceback; traceback.print_exc()
         return np.array(image_pil)
 
+
 # ---- GESTION ASYNCHRONE / ÉTAT ----
 current_image = None
 current_predictions = None