The portability principle

Models should move seamlessly across environments and infrastructures, without extensive rework or environment-specific fixes.. With trac you can:

  • Use the model runtime in your IDE to build trac models (in Python)

  • Deploy those models to the trac platform, without modification

  • Deploy those models to any Python execution environment, without modification

You still need to prove data consistency but with trac you don’t need to re-code the model

Trac runtime and Model API

The model runtime is a python package that provides the context within which trac models run. It guarantees that models always receive data and parameters that meet their requirements and takes responsibility for accessing, marshalling and formatting data away from model code.

select file

Build once, run anywhere

The same code can be called in multiple environments — the model runtime manages data marshalling to ensure the model always receives the correct inputs.

select file

Using trac in an IDE

The model runtime can be used in any modern IDE and many notebooks. All the normal debugging facilities will be available to help step through code, inspect data in memory and interrogate log outputs.

The TRAC runtime package can be installed directly from PyPI:

pip install tracdap-runtime