📄️ Concepts and Entities
📄️ Configuring ModelBox Server
ModelBox server reads a toml configuration file to initialize the various service dependencies. A sample configuration is here -
📄️ Logging Experiment Metadata
Log experiment metadata like hyperparameters, discrete operational events from the trainer, model metrics after every epoch, and even hardware metrics where training is being run.
📄️ Creating Model and Model Versions
📄️ Logging Experiment and Model Metrics
ModelBox integrates with metrics storage services to store training hardware, experiment and model metrics.
📄️ Distributing Models to Inference Services
📄️ Exporting metadata to TensorBoard
📄️ Using the Python SDK
The Python SDK can be used to interact with the ModelBox API from model trainers and other MLOps services.
Generate server configuration
📄️ Troubleshooting of the ModelBox Control Plane Service
📄️ Developing ModelBox
We cover how to contribute to ModelBox using GitPod and some miscellaneous topics related to the development of the service and SDK.