Medical-zkML-base

Description

As the base library of Medical-zkML. Using leo_translate tool generate machine learning model prediction process, likes decision tree，K-Means and XGBoost. Additional libraries will be added in the future.

Directory structure

• data: The UCI nine medical datasets, "new_data.tsv" file is generated after data preprocessing.
• leo_translate: Leo transpiler, include Leo language different module
  • context: Process the transformation context to build the complete Noir file
  • core_module: Include some core components such as statements, struct and function
  • submodule: Include some base components such as base type sign and keywords
  • utils: tools such as code format
• model_generate: Include three ML algorithm train and generate model obj
• src: Include three ML algorithm and preprocess code, Used python generated Leo code and data preprocess
• tests: Testcase for leo transpiler and translation model from python to Leo

Build guide

• Python 3.7+
• Anaconda
• Leo for Visual Studio Code 0.16.0