Installation

  1. Install or make sure you have installed Python3.
  2. Install or make sure you have installed pip3.
  3. If you do not have it installed already, install virtualenv.
  4. If you wish to run Bittensor in dockerized containers, install Docker Desktop.
  5. Clone the Bittensor repository locally:
git clone git@github.com:opentensor/bittensor.git
  1. If you wish to simply run a dockerized Bittensor instance, you may do so by running:
./start_bittensor.sh

This will set up and bootstrap Bittensor in a local Docker container. If the process succeeds, you should see a “Bittensor” ASCII banner as the model is bootstrapped and run on docker. The default model that runs in this case is the “MNIST” dataset model. This is an example model that is set up under examples/mnist. You can specify other models by using the -n, --neuron flag.

  1. Create a new virtual environment and activate it:
python3 -m venv env
source env/bin/activate
  1. Install required packages:
pip3 install -r requirements.txt
  1. Install protobufs
pip3 install -e .
  1. You can verify your installation has succeeded by going to one of the examples and running its main.py folder.
cd examples/mnist
python3 main.py

Similarly to step 6, this will run the example with all the default flags. If everything installed as it should, you should see an ASCII “Bittensor” printed before the model starts training.