Update app.py

app.py

@@ -9,7 +9,6 @@ from transformers import DetrImageProcessor, TableTransformerForObjectDetection
 # ===============================
 # Chargement des modèles
 # ===============================
-
 DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
 
 det_processor = DetrImageProcessor.from_pretrained(
@@ -29,7 +28,6 @@ struct_model = TableTransformerForObjectDetection.from_pretrained(
 # ===============================
 # OCR cellule
 # ===============================
-
 def ocr_cell(image):
     gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
     text = pytesseract.image_to_string(gray, config="--psm 6")
@@ -38,14 +36,14 @@ def ocr_cell(image):
 # ===============================
 # Fonction principale
 # ===============================
-
 def extract_description(image_pil):
-
+    # Convertir PIL -> np.array
     image = np.array(image_pil)
     h, w, _ = image.shape
 
     # ---- Détection du tableau ----
-    inputs = det_processor(images=image_pil, return_tensors="pt").to(DEVICE)
+    inputs = det_processor(images=image, return_tensors="pt")
+    inputs = {k: v.to(DEVICE) for k, v in inputs.items()}
     outputs = det_model(**inputs)
 
     results = det_processor.post_process_object_detection(
@@ -67,7 +65,8 @@ def extract_description(image_pil):
     table_img = image[y0:y1, x0:x1]
 
     # ---- Structure du tableau ----
-    inputs = struct_processor(images=Image.fromarray(table_img), return_tensors="pt").to(DEVICE)
+    inputs = struct_processor(images=table_img, return_tensors="pt")
+    inputs = {k: v.to(DEVICE) for k, v in inputs.items()}
     outputs = struct_model(**inputs)
 
     results = struct_processor.post_process_object_detection(
@@ -87,7 +86,6 @@ def extract_description(image_pil):
 
     # ---- Grouper par colonne (X) ----
     cells_sorted = sorted(cells, key=lambda b: (b[0] + b[2]) / 2)
-
     columns = {}
     for cell in cells_sorted:
         cx = (cell[0] + cell[2]) // 2
@@ -116,7 +114,7 @@ def extract_description(image_pil):
 
     if desc_col is None:
         return "❌ Colonne 'Description' non trouvée", "\n\n".join(
-            [" | ".join(col) for col in column_texts]
+            [f"Col {i}: " + " | ".join(col) for i, col in enumerate(column_texts)]
         )
 
     return "\n".join(desc_col[1:]), "\n\n".join(desc_col)
@@ -124,7 +122,6 @@ def extract_description(image_pil):
 # ===============================
 # Interface Gradio
 # ===============================
-
 demo = gr.Interface(
     fn=extract_description,
     inputs=gr.Image(type="pil", label="Image de facture"),