DocumentationAPI Reference
Back to ConsoleLog In

Deploying an RTiC Client

The RTiC Python Client container enables your application to communicate with the RTiC server in order to send HTTP/REST requests to handle models, check their status, and execute inference requests (predict).

This page will guide you through Deci’s python client for RTiC inference.

For information about the API, see RTiC Server API Documentation.

The client code snippets below demonstrate how to use the Python client, and what you need to know about the client library.

Installing the Python Client

Install the package using pip. Skip this step if you have already installed the platform API client.

  • Python 3
python3 -m pip install deci-client

Initializing a New Client

In order to communicate with the RTiC server, you must create a new client object of the class DeciInferencerClient, as follows –

from deci_client import DeciClient

# DEFUALT RTiC SERVER CONFIG
RTIC_SERVER_PORT = 8000
RTIC_SERVER_IP_ADDRESS = 'localhost'

client = DeciClient(rtic_host=RTIC_SERVER_IP_ADDRESS,
                    rtic_port=RTIC_SERVER_PORT)

If you already registered a model while launching the RTiC server or if you registered a model with RTiC server using the RTiC client, then you can get a list of previously registered models in the RTiC server.


Did this page help you?