Code Completion using Neural Attention and Byte Pair Encoding
This is an incremental improvement for code completion tasks, potentially simplifying neural network architectures.
The paper tackles code completion by applying Byte Pair Encoding (BPE) to source code as natural text, without using abstract syntax trees, and compares an attention-enhanced LSTM with a pointer network to see if BPE can eliminate the need for the pointer network.
In this paper, we aim to do code completion based on implementing a Neural Network from Li et. al.. Our contribution is that we use an encoding that is in-between character and word encoding called Byte Pair Encoding (BPE). We use this on the source code files treating them as natural text without first going through the abstract syntax tree (AST). We have implemented two models: an attention-enhanced LSTM and a pointer network, where the pointer network was originally introduced to solve out of vocabulary problems. We are interested to see if BPE can replace the need for the pointer network for code completion.