A framework for code-agnostic, interactive prototyping of DNNs.
- Transparent and elastic scheduling of DNN training jobs on modern HPC systems.
- Monitoring and visualizing model parameters and computational performance statistics.
- Perform semi-automatic hyperparameter tuning/optimization and architecture search using evolutionary algorithms.
- A user-defined interactive interface to drive the framework/ design process, not bound to any particular framework.
- Scaling the functionality and performance of the model as the resources increase.
How do I get set up?
pip3 install protonnfor latest stable release
pip3 install git+https://github.com/undertherain/protoNN.gitfor recent development version
- Python 3.4 or later is required
For licensing information, please see LICENSE