inferyx

A Python SDK for the Inferyx AI/Data Analytics platform.

inferyx lets you programmatically create, manage, and orchestrate Inferyx platform entities — datasources, data pipelines, data products, business rules, ML models, schedulers, and more — directly from Python.

Installation

pip install inferyx

Getting started

Every operation starts with an AppConfig, which holds your platform host and authentication tokens.

from inferyx.config import AppConfig
from inferyx.components.admin import Datasource

app_config = AppConfig(
    host="analytic.inferyx.com",
    appToken="<YOUR_APP_TOKEN>",
    adminToken="<YOUR_ADMIN_TOKEN>",
)

datasource = Datasource(
    app_config=app_config,
    name="my_postgres_source",
    type="POSTGRES",
)

response = datasource.create()
print(response)

You can also authenticate using a JWT token (with optional auto-refresh):

app_config = AppConfig(
    host="analytic.inferyx.com",
    jwt_token="<JWT_TOKEN>",
    token_refresher=lambda: get_new_jwt_token(),
)

What you can do

The SDK is organized into components, each mapping to a core area of the Inferyx platform:

Example: Deploying a model

from inferyx.config import AppConfig
from inferyx.components.data_science import ModelDeploy

app_config = AppConfig(
    host="analytic.inferyx.com",
    appToken="<YOUR_APP_TOKEN>",
    adminToken="<YOUR_ADMIN_TOKEN>",
)

deploy_instance = ModelDeploy(app_config)

response = deploy_instance.deploy(
    file="/path/to/model.pmml",
    resource_name="resource_local_deploy",
    datasource_name="file_upload_s3_source",
    model_name="my_model",
    model_type="PMML",
)

print(response.text)

Requirements

• Python 3.8+
• A valid Inferyx platform host and access tokens

Links

• Homepage
• Bug Tracker

License

Copyright © Inferyx. All rights reserved.