DISCO needs to be able to advertise its available models and parameters through API calls, as well as creating a new DISCOllaborative. It also needs to be able to upload a trained model to an external endpoint.
Exposing parameters and DISCOllaborative creation is relatively easy. To expose models, they first need to be decoupled with tasks, then made available by discojs to the server.
DISCO needs to be able to advertise its available models and parameters through API calls, as well as creating a new DISCOllaborative. It also needs to be able to upload a trained model to an external endpoint.
Exposing parameters and DISCOllaborative creation is relatively easy. To expose models, they first need to be decoupled with tasks, then made available by discojs to the server.