Disclaimer...Reading!!!

This campaign rewards users who run worker nodes providing inferences for the US presidential election party winner once a day. Every inference should be the likelihood of the republican party winning the election. source run-inference-political

1. Components

• Worker: The node that publishes inferences to the Allora chain.
• Inference: A container that conducts inferences, maintains the model state, and responds to internal inference requests via a Flask application. This node operates with a basic linear regression model for price predictions.
• Updater: A cron-like container designed to update the inference node's data by daily fetching the latest market information from the data provider, ensuring the model stays current with new market trends.
• Topic ID: Running this worker on TopicId 11
• TOKEN= D For have inference D: Democrat
• TOKEN= R For have inference R: Republic
• MODEL: Own model or modify or check folder /models model.py
• Probability: Predict of % total 0 - 100%
• Dataset: polymarket.com
• An expected result: Every 24 hours

Setup Worker

The structure of topic 11 - Allora Worker Node

./root/allora-usa-election
├── config.json
├── docker-compose.yml
├── Dockerfile
├── app.py
├── model.py
├── requirements.txt
├── worker-data
       └── environment
├── inference-data
       └── dataset.csv (R/D)
       └── model
             └── model.pkl (R/D)

1. Clone this repository

   git clone https://github.com/arcxteam/allora-usa-election.git
   cd allora-usa-election

2. Provided and config environment file modify model-tunning

   Copy and read the example .env.example for your variables

   nano .env.example .env

   Here are the currently accepted configuration

   • TOKEN= (D or R)
   • MODEL= the name as model (defaults SVR or modify your own model)
   • Save ctrl X + Y and Enter

   Modify model-tunning or check /models folder

    nano model.py

3. Edit your config & initialize worker

   Edit for key wallet - name wallet - RPC endpoint & interval etc

   nano config.json

   Run the following commands root directory to initialize the worker

   chmod +x init.config
   ./init.config

4. Start the Services

   Run the following command to start the worker node, inference, and updater nodes:

   docker compose up --build -d

   Check running

   docker compose logs -f --tail=100

   To confirm that the worker successfully sends the inferences to the chain, look for the following logs:

   {"level":"debug","msg":"Send Worker Data to chain","txHash":<tx-hash>,"time":<timestamp>,"message":"Success"}
   

   

2. Testing Inference Only

Send requests to the inference model. For example, request probability of Democrat(D) or Republic(R)

curl http://127.0.0.1:8000/inference/D

Expected response of numbering: "value":"xx.xxxx"

3. Note: Checking the logs that worker are registered successfully

docker logs -f (your container id) 2>&1 | head -n 70

docker logs -f (your container id) 2>&1 | grep -i "Success"

Result every 24H: This will only get 1 request per day for topic 11 and points will depends on how your worker prediction vs the ground truth from polymarket