Analysing source code using deep learning aids compile-time decisions affecting performance in embedded devices. We propose DeepCodeGraph, a general graph-based language model, which learns patterns to identify better compilation strategies, optimal hardware configurations and software transformations. DCG includes i) A large-scale dataset containing over 100k graphs. ii) A graph neural network to implement a graph-based language model. iii) A self-supervised pre-training framework leveraging Masked Graph Autoencoders. The performance of DCG is evaluated on two downstream tasks: heterogeneous device mapping and thread block size prediction. DCG outperforms previous graph-based state-of-the-art improving previous results by 3%.
Cichetti, F., Parisi, E., Acquaviva, A., Barchi, F. (2024). DeepCodeGraph: A Language Model for Compile-Time Resource Optimization Using Masked Graph Autoencoders. GEWERBESTRASSE 11, CHAM, CH-6330, SWITZERLAND : Springer Science and Business Media Deutschland GmbH [10.1007/978-3-031-70239-6_32].
DeepCodeGraph: A Language Model for Compile-Time Resource Optimization Using Masked Graph Autoencoders
Cichetti F.;Parisi E.;Acquaviva A.;Barchi F.
2024
Abstract
Analysing source code using deep learning aids compile-time decisions affecting performance in embedded devices. We propose DeepCodeGraph, a general graph-based language model, which learns patterns to identify better compilation strategies, optimal hardware configurations and software transformations. DCG includes i) A large-scale dataset containing over 100k graphs. ii) A graph neural network to implement a graph-based language model. iii) A self-supervised pre-training framework leveraging Masked Graph Autoencoders. The performance of DCG is evaluated on two downstream tasks: heterogeneous device mapping and thread block size prediction. DCG outperforms previous graph-based state-of-the-art improving previous results by 3%.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
