Add application file

app.py

@@ -7,7 +7,9 @@ model_name_or_path = "lyttt/VLV_captioner"
 model = AutoModel.from_pretrained(model_name_or_path, revision="master", trust_remote_code=True,low_cpu_mem_usage=False)
 
 def greet(image):
-    image = image.convert("RGB") 
+    if image.dtype != np.uint8:
+        image = (np.clip(image, 0, 1) * 255).astype(np.uint8)
+    image = Image.fromarray(image, mode='RGB')
     with torch.no_grad():
         outputs = model([image]).generated_text[0]
     return output