Update README.md

README.md

@@ -57,9 +57,9 @@ inputs  = processor(images, texts[::-1], padding=True, return_tensors="pt")
 outputs_swapped = model(**inputs)
 
 print('Loss', outputs.loss.item())
+# Loss 0.00191505195107311
 print('Loss with swapped images', outputs_swapped.loss.item())
-# Loss 0.0027269450947642326
-# Loss with swapped images 2.987490177154541
+# Loss with swapped images 2.1259872913360596 
 ```
 
 Here is how to use this model to perform image and text matching