MarianCG: a code generation transformer model inspired by machine translation
Open paper
Benchmark
Model
Rank
Results
code-generation-on-django
MarianCG
#1
Accuracy: 81.83
BLEU Score: 90.41
code-generation-on-conala
MarianCG
#2
BLEU: 34.43
Exact Match Accuracy: 10.2
Rank counts only results with a code link.