TwitterMediumGithub

Flock - Training Node

Training Node Guide - Start Participating Early to Gain an Edge When the Competition Starts

1. Prerequisites and Setup

  • Wallet: Use MetaMask.

  • Network: Base Sepolia.

  • Recommendation: Use a new wallet or a testnet wallet to avoid risks with mainnet assets.

2. Verify GitHub and Claim $FML Tokens

  • Go to GitHub and verify your account.

  • After verification, claim $FML tokens.

3. Stake Your Tokens

  1. Go to Stake-to-Earn.

  2. Search for cryptonode3 or address 0xe0a9ac439135dfadd0c1502d6d12c38f272d8102.

  3. Stake your $FML tokens to participate.

4. Get Your API Key

  • In the Stake-to-Earn dashboard, click the dropdown at the top left corner to get your API Key.

    This API key will be used in later steps, so keep it safe.

5. Create a Token on Hugging Face

  • Register on Hugging Face using your GitHub or Google account: Hugging Face Tokens.

  • In Settings, create a token named "Flock".

  • Save both your Hugging Face token and username.

6. Set Up Your Environment

Windows Users: Install WSL

Linux Users: Install Miniconda

  1. Download the Miniconda installer:

    wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh

  2. Run the installer:

    bash Miniconda3-latest-Linux-x86_64.sh

  3. Add Conda to Shell (Current Session):

    eval "$(/root/miniconda3/bin/conda shell.bash hook)"

  4. Permanently Initialize Conda:

    conda init

    • Restart your terminal or source it:

      source ~/.bashrc

  5. Verify Conda Installation:

    conda --version

7. Clone the Training Node Quickstart Repository

  • Clone the repository that contains the setup script:

    git clone https://github.com/FLock-io/testnet-training-node-quickstart.git

  • Enter the project directory and install packages:

    conda create -n training-node python==3.10
conda activate training-node
pip install -r requirements.txt

8. Select Your Task ID

9. Run the Full Automation Script

Use your API key, task ID, Hugging Face token, and username to start training your node.

Linux Command:

TASK_ID=<task-id> FLOCK_API_KEY="<your-flock-api-key>" HF_TOKEN="<your-hf-token>" CUDA_VISIBLE_DEVICES=0 HF_USERNAME="<your-hf-username>" python full_automation.py

Windows Command (PowerShell):

$env:TASK_ID = "<task-id>"
$env:FLOCK_API_KEY = "<your-flock-api-key>"
$env:HF_TOKEN = "<your-hf-token>"
$env:CUDA_VISIBLE_DEVICES = "0"
$env:HF_USERNAME = "<your-hf-username>"
python full_automation.py

10. Check Node Status

Once your training node participates in AI model validation, it will appear on the Stake-to-Earn Dashboard.

Additional Resources

Last updated