Update app.py

app.py

@@ -7,6 +7,7 @@ from pdf2image import convert_from_path
 from PIL import Image
 from torch.utils.data import DataLoader
 from tqdm import tqdm
+import tempfile
 
 from colpali_engine.models import ColQwen2, ColQwen2Processor
 
@@ -61,9 +62,16 @@ def index(files, ds):
 
 def convert_files(files):
     images = []
-    for f in files:
-        images.extend(convert_from_path(f, thread_count=4))
-
+    with tempfile.TemporaryDirectory() as temp_dir:
+        for f in files:
+            file_path = f['filepath']
+            temp_file_path = os.path.join(temp_dir, f['name'])
+            shutil.copy(file_path, temp_file_path)
+            try:
+                images.extend(pdf2image.convert_from_path(temp_file_path, thread_count=4))
+            except Exception as e:
+                print(f"Error converting {temp_file_path}: {e}")
+                # Handle the error, maybe skip the file or raise a Gradio error
     if len(images) >= 150:
         raise gr.Error("The number of images in the dataset should be less than 150.")
     return images