Code for collecting ezkl performance metrics on
Zero-Knowledge Proof generation over inference tasks
for multiple deep neural network models following a previously-introduced methodology1.
The entrypoint main.py provides a command-line interface powered by Hydra,
which exposes configuration options that can be listed using the --help flag:
$ python src/main.py --helpDetails on important settings are provided below.
Controls which neural network model is used for proof generation and verification.
It accepts a list of strings, corresponding to the names of the models to be used.
For a list of supported models, see model.py.
If left unspecified (default), all models will be used.
Footnotes
-
Cerioli, L.: A survey on Zero Knowledge proofs and their applications on machine learning. Master’s thesis, Politecnico di Milano, Scuola di Ingegneria Industriale e dell’Informazione (5 2023), https://hdl.handle.net/10589/209893 ↩