Update app.py

app.py

@@ -14,18 +14,18 @@ from tqdm import tqdm
 from colpali_engine.models import ColQwen2, ColQwen2Processor
 
 
-@spaces.GPU
 def install_fa2():
     print("Install FA2")
     os.system("pip install flash-attn --no-build-isolation")
-# install_fa2()
+
+install_fa2()
 
 
 model = ColQwen2.from_pretrained(
         "vidore/colqwen2-v1.0",
         torch_dtype=torch.bfloat16,
         device_map="cuda:0",  # or "mps" if on Apple Silicon
-        # attn_implementation="flash_attention_2", # should work on A100
+        attn_implementation="flash_attention_2", # should work on A100
     ).eval()
 processor = ColQwen2Processor.from_pretrained("vidore/colqwen2-v1.0")
 
@@ -84,7 +84,6 @@ def query_gpt4o_mini(query, images, api_key):
     return "Enter your OpenAI API key to get a custom response"
 
 
-@spaces.GPU
 def search(query: str, ds, images, k, api_key):
     k = min(k, len(ds))
     device = "cuda:0" if torch.cuda.is_available() else "cpu"
@@ -124,8 +123,8 @@ def convert_files(files):
     for f in files:
         images.extend(convert_from_path(f, thread_count=4))
 
-    if len(images) >= 150:
-        raise gr.Error("The number of images in the dataset should be less than 150.")
+    if len(images) >= 500:
+        raise gr.Error("The number of images in the dataset should be less than 500.")
     return images