Setting Up Environment

Running Locally

We've provided support for 2 environment tools: venv and pyenv-virtualenv.

Linux/MacOS

venv

If you use python alias, run: source scripts/env.sh venv

If you use python3 alias, run: source scripts/env.sh venv --python3

pyenv

If you use python alias, run: source scripts/env.sh pyenv

If you use python3 alias, run: source scripts/env.sh pyenv --python3

Windows

venv

Run the following commands inside an admin powershell.

If you use python alias, run: . scripts/env.sh venv

If you use python3 alias, run: . scripts/env.sh venv --python3

Without Script

If for whatever reason the script doesn't work, you can manually set up the environment by creating a virtual environment and installing the dependencies inside scripts/requirements.in.

Running in Docker

To start the docker container: docker-compose up --build Once inside /app in the container: source scripts/env.sh venv Note the container mounts your local file system of the repo to /app in the container.

Adding Dependencies

If you need to add dependencies, add them to scripts/requirements.in and run the same commmand you used to setup the environment.

Exchange Client Program

Overview

main.py is the entry point for the strategy. It is responsible for:

1. Initializing the TradingClient
2. Initializing the SharedState and Prioritizer using the exchange client
3. Initializing the Strategy
4. Passing the Strategy to the TradingClient (will run trigger based strategy)
5. Starting the Strategy (subscribe to exchange websocket streams and run strategy)

Strategy

There are 2 triggers:

1. Orderbook Update: This trigger is called when the orderbook is updated. The SharedState will contain all updated info. Call the quoter to place trades.
2. Portfolio Update: This trigger is called when a portfolio snapshot (positions, balance, open orders, etc.) is sent. The SharedState will contain all updated info. Call the quoter to place trades.

Quoter

We have provided a simple quoter that performs client-side rate limiting in Prioritizer. You can modify the place order function to implement prioritization logic and decide whether or not to submit trades.

Filtered Orderbook

The orderbook is stored in the SharedState. The orderbook is, by default, filtered to exclude your own orders. You can access both the raw orderbook if needed, though.

To access the depth of a bid for a particular ticker and price level: depth = shared_state.orderbooks[ticker]["bids"][price_level]

Likewise, to access the depth of an ask: depth = shared_state.orderbooks[ticker]["asks"][price_level]

User Portfolio

The user portfolio is stored in the SharedState. You can access the your positions, balance, open orders, etc. The portfolio is stored as a dictionary.

To access the balance of a particular ticker: balance = shared_state.portfolio.positions[ticker]

To access the open orders of a particular ticker: open_orders = shared_state.portfolio.orders[ticker]

To access the balance: balance = shared_state.portfolio.balance

To access the PnL: pnl = shared_state.portfolio.pnl