A Grammar-Based Structural CNN Decoder for Code Generation

Understand

Code generation maps a program description to executable source code in a programming language.

• Existing approaches mainly rely on a recurrent neural network (RNN) as the decoder.
• However, we find that a program contains significantly more tokens than a natural language sentence, and thus it may be inappropriate for RNN to capture such a long sequence.
• In this paper, we propose a grammar-based structural convolutional neural network (CNN) for code generation.