Introduction

Introduction

The BOSS Modeling Framework (BMF) enables developers to build and interface custom AI models with the BOSS Unity Client platform for streamlined management, experimentation, and training using data and parameters established in the BOSS Unity client (or simply, Unity client). The framework supports python-based AI models built with TensorFlow, PyTorch, Scikit-learn, XGBoost (dask module), and federated models. BMF’s python libraries support the following tasks: - accessing BOSS virtual datasets (VDSes) for model training and evaluation, - analyzing and reporting model performance metrics (e.g., with confusion matrices, ROC curves), - storing structures representing trained models and training checkpoints.

Additionally, BMF supports distributed model training using the Horovod framework (https://horovod.ai/). Please note that in version 7.0.0, BMF support for Horovod has only been tested using PyTorch models.

Model development within the BOSS Modeling Framework is best understood in conjunction with the Modeling section of the Unity Client guide .