Skip to Main Content
Table 4: 
Results on the official test sets for both code generation and image captioning tasks.
ModelDjangoMS-COCO
BLEUAccuracyBLEUCIDEr-D
L2R 36.74 13.6% 22.12 68.88 
SAO 42.33 16.3% 22.58 69.42 
ModelDjangoMS-COCO
BLEUAccuracyBLEUCIDEr-D
L2R 36.74 13.6% 22.12 68.88 
SAO 42.33 16.3% 22.58 69.42 
Close Modal

or Create an Account

Close Modal
Close Modal