Skip to content

Agent

The Agent component is the core of the system, which will be in charge of:

  • Running the ML or AI models distributed in the system
    • Running the models which require intensive computation
    • LLM models
    • Text2Speech models
    • Emotion Recognition models
    • etc.

It is writen in Python, and it is a pretty standard Python project.

Each different task will have a subfolder within the modules folder

Latency Logger

Key thing to notice is that we create two classes to log the time point and duration to profile the latency performance of the models.

  • Agent/utils/time_logger.py: log time point
  • Agent/utils/time_tracker.py: track duration

Docker setup

We also setup the docker for the Agent component, which is in the Dockerfile and docker-compose.yml file.

Storage solution

How we handle the different storage solution is inside the storage.py file.

Data

As we mentioned in the introduction, models will be need to be downloaded to the data/models folder, it is normally automatically.

Unless you want to run our emotion detection model, if you want to do that, refer to our introduction page.