Researchers:

• Zhangcheng Qiangy
• Yuxin Zhang
• Pari Delir Haghighi
• Abdur Forkan
• Prem Prakash Jayaraman (Swinburne University)

Summary:

This study introduces an ontology-based platform (MobileDLSearch) that offers end-users greater flexibility to store, query, share and reuse pre-trained DL models for various mobile applications. The ontology represents various DL models with different backends (e.g. TensorFlow, Keras and PyTorch) and is used to semantically search and retrieve DL models using an intuitive and interactive user interface.

The implemented system also provides an automatic model converter to optimise desktop/laboratory-oriented pre-trained DL models for mobile platforms and has an on-device real-time model integration module to benchmark the model's performance on mobile devices.