Deep Program Understanding
Teaching machines to understand complex algorithms
We have open-sourced many of our work and implementations.
Libraries
- dpu-utils (opens in new tab): Useful Python utilities for projects on deep program understanding.
- gated-graph-neural-networks (opens in new tab): A set of efficient TensorFlow implementations of graph neural networks that can handle large and sparse graphs.
- tf2-gnn (opens in new tab): TensorFlow 2 library implementing Graph Neural Networks
- ptgnn (opens in new tab): A PyTorch Graph Neural Network Library
Project-Specific Utilities
- constrained-graph-variational-autoencoders (opens in new tab): code for constrained graph VAEs.
- DeepCoder-Utils (opens in new tab): Code used in the experiments of the DeepCoder paper (ICLR 2017)
- graph-partition-neural-network-samples (opens in new tab): Sample code for Graph Partition Neural Networks.
- dpu-learning-to-represent-edits (opens in new tab): C# data extraction for “Learning to Represent Edits”
- graph-based-code-modelling (opens in new tab): The code for the ICLR’18 and ICLR’19 papers