← Back to all projects
04 / Back end & machine learning

DRL for Split Model Allocation.

A FastAPI back end and React portal for managing split deep learning models and edge-cloud allocation experiments.

Research project
Back end & machine learning

Context

The project explored allocation of split deep learning models between edge and cloud environments.

My contribution

Built the management API and portal, and implemented DQN and PPO approaches for allocation experiments.

Outcome

A working platform for managing models and testing allocation approaches.