This repository provides an example Allora network worker node designed to offer price predictions for Ethereum (ETH). It demonstrates the utilization of a basic inference model operating within a dedicated container and showcases its seamless integration with the Allora network infrastructure, enabling it to contribute with valuable inferences.
This code is based on the repository from the Allora Network: https://github.com/allora-network/basic-coin-prediction-node/tree/main
This Flask app exposes endpoints to generate inferences and update the model. It acts as the gateway to the model from external requests.
/inference/<string:token>: Generates inference for the given token./update: Updates the model by downloading and training with the latest data./status: Checks the status of the model and data.
This service is designed to hit the /update endpoint on the inference service, ensuring the model state is updated when needed. It can also be scheduled to run periodically.
This script combines the Allora inference base, node function, and custom logic. The Allora chain makes requests to the head node, which then directs requests to the workers.
The scripts utilize environment variables for configuration:
INFERENCE_API_ADDRESS: The address of the inference API.
docker-compose builddocker-compose up -ddocker-compose logs -fdocker-compose down