Updated handler.py

handler.py

@@ -34,10 +34,10 @@ class EndpointHandler:
         raw_images = inputs
 
         inputs = self.processor(inputs, return_tensors="pt").to("cuda")
-
-        out = self.model.generate(**inputs)
                                      
         processed_image = self.processor(images=raw_images, return_tensors="pt").to(device)
+
+        out = self.model.generate(**processed_image)
         # processed_image["pixel_values"] = processed_image["pixel_values"].to(device)
         # processed_image = {**processed_image, **parameters}
         
@@ -46,6 +46,6 @@ class EndpointHandler:
         #         **processed_image
         #     )
         captions = self.processor.decode(out[0], skip_special_tokens=True)
-        
+
         # postprocess the prediction
         return {"captions": captions}