catanatron 2.1.0

Fast Settlers of Catan Python Implementation

Catanatron

Fast Settlers of Catan Python implementation and strong AI player.

The goal of this project is to find the strongest Settlers of Catan bot possible.

Usage

Install with pip:

pip install catanatron

Make your own bot by implementing the following API (see examples in catanatron/players and experimental/machine_learning/players):

from catanatron.game import Game
from catanatron.models.actions import Action
from catanatron.models.player import Player

class MyPlayer(Player):
    def decide(self, game: Game, playable_actions: Iterable[Action]):
        """Should return one of the playable_actions.

        Args:
            game (Game): complete game state. read-only.
            playable_actions (Iterable[Action]): options to choose from
        Return:
            action (Action): Chosen element of playable_actions
        """
        raise NotImplementedError

Then run a game (or many) like:

from catanatron.game import Game
from catanatron.models.player import RandomPlayer, Color

players = [
    MyPlayer(Color.RED),
    RandomPlayer(Color.BLUE),
    RandomPlayer(Color.WHITE),
    RandomPlayer(Color.ORANGE),
]
game = Game(players)
game.play()  # returns winning color

You can then inspect the game state any way you want (e.g. game.state.player_state, game.state.actions, game.state.board.buildings, etc...). See documentation for more.

For watching these games in a UI see watching games.

Advanced Usage

Cloning the repo and using directly will allow you to access additional tools not included in the core package. In particular, a web UI for watching games and a experimental/play.py script that provides a blueprint to run many games, collect summary statistics (avg vps, avg game length, etc...), save game for viewing in browser, and/or generate machine learning datasets.

Create a virtualenv with Python 3.8 and install requirements:

python3.8 -m venv venv
source ./venv/bin/activate
pip install -r dev-requirements.txt

Run games with the play.py script. It provides extra options you can explore with --help:

python experimental/play.py --num=100

Currently, we can execute one game in ~76 milliseconds.

Watching Games

We provide a docker-compose.yml with everything needed to watch games (useful for debugging). It contains all the web-server infrastructure needed to render a game in a browser.

Ensure you have Docker Compose installed, and run:

docker-compose up

To open a game from another command line process in the browser, set the following environment variable:

export DATABASE_URL=postgresql://catanatron:victorypoint@localhost:5432/catanatron_db

and use the open_link helper function:

from catanatron_server.utils import open_link
open_link(game)  # opens game in browser

Documentation

See https://catanatron.readthedocs.io for more details on how we represent the state and actions.

In summary, Actions are tuples of enums like: (ActionType.PLAY_MONOPOLY, Resource.WHEAT) or (ActionType.BUILD_SETTLEMENT, 3) (i.e. build settlement on node 3).

State is currently represented by a simple data container class and is mutated by the functions in the state_functions module. This functional style allows us to create state copies (for bots that search through state space) faster. The closer we make this State class to an array of immutable primitives, the faster it will be to copy.

Catanatron OpenAI's Gym API

See catanatron_gym.

Architecture

For debugging and entertainment purposes, we wanted to provide a UI with which to inspect games.

We decided to use the browser as a rendering engine (as opposed to the terminal or a desktop GUI) because of HTML/CSS's ubiquitousness and the ability to use modern animation libraries in the future (https://www.framer.com/motion/ or https://www.react-spring.io/).

To achieve this, we separated the code into three components:

• catanatron: A pure python implementation of the game logic. Uses networkx for fast graph operations. Is pip-installable (see setup.py) and can be used as a Python package.

• catanatron_server: Contains a Flask web server in order to serve game states from a database to a Web UI. The idea of using a database, is to ease watching games from different processes (you can play a game in a standalone Python script and save it for viewing). It defaults to using an ephemeral in-memory sqlite database.

• React Web UI: A web UI to render games. The ui folder.

Experimental Folder

The experimental folder contains unorganized code with many failed attempts at finding the best possible bot.

AI Bots Leaderboard

The best bot is AlphaBetaPlayer with n = 2. Here a list of bots strength. Experiments done by running 1000 (when possible) 1v1 games against previous in list.

Player	% of wins in 1v1 games	num games used for result
AlphaBeta(n=2)	80% vs ValueFunction	25
ValueFunction	90% vs GreedyPlayouts(n=25)	25
GreedyPlayouts(n=25)	100% vs MCTS(n=100)	25
MCTS(n=100)	60% vs WeightedRandom	15
WeightedRandom	53% vs WeightedRandom	1000
VictoryPoint	60% vs Random	1000
Random	-	-

Developing for Catanatron

To develop for Catanatron core logic you can use the following test suite:

coverage run --source=catanatron -m pytest tests/ && coverage report

Or you can run the suite in watch-mode with:

ptw --ignore=tests/integration_tests/ --nobeep

Machine Learning

Generate data (GZIP CSVs of features and PGSQL rows) by running:

python experimental/play.py --num=100 --outpath=my-data-path/

You can then use this data to build a machine learning model, and then implement a Player subclass that implements the corresponding "predict" step of your model. There are some examples of these type of players in experimental/machine_learning/players/reinforcement.py.

Appendix

Running Components Individually

As an alternative to running the project with Docker, you can run the following 3 components: a React UI, a Flask Web Server, and a PostgreSQL database in three separate Terminal tabs.

React UI

Make sure you have yarn installed (https://classic.yarnpkg.com/en/docs/install/).

cd ui/
yarn install
yarn start

This can also be run via Docker independetly like (after building):

docker build -t bcollazo/catanatron-react-ui:latest ui/
docker run -it -p 3000:3000 bcollazo/catanatron-react-ui

Flask Web Server

Ensure you are inside a virtual environment with all dependencies installed and use flask run.

python3.8 -m venv venv
source ./venv/bin/activate
pip install -r requirements.txt
flask run

This can also be run via Docker independetly like (after building):

docker build -t bcollazo/catanatron-server:latest .
docker run -it -p 5000:5000 bcollazo/catanatron-server

PostgreSQL Database

Make sure you have docker-compose installed (https://docs.docker.com/compose/install/).

docker-compose up

Or run any other database deployment (locally or in the cloud).

Other Useful Commands

TensorBoard

For watching training progress, use keras.callbacks.TensorBoard and open TensorBoard:

tensorboard --logdir logs

Docker GPU TensorFlow

docker run -it tensorflow/tensorflow:latest-gpu-jupyter bash
docker run -it --rm -v $(realpath ./notebooks):/tf/notebooks -p 8888:8888 tensorflow/tensorflow:latest-gpu-jupyter

Testing Performance

python -m cProfile -o profile.pstats experimental/play.py --num=5
snakeviz profile.pstats

pytest --benchmark-compare=0001 --benchmark-compare-fail=mean:10% --benchmark-columns=min,max,mean,stddev

Head Large Datasets with Pandas

In [1]: import pandas as pd
In [2]: x = pd.read_csv("data/mcts-playouts-labeling-2/labels.csv.gzip", compression="gzip", iterator=True)
In [3]: x.get_chunk(10)

Publishing to PyPi

catanatron Package

pip install twine
rm -rf build
rm -rf dist
python setup.py sdist bdist_wheel
twine check dist/*
twine upload --repository-url https://test.pypi.org/legacy/ dist/*
twine upload dist/*

Building Docs

sphinx-quickstart docs
sphinx-apidoc -o docs/source catanatron
sphinx-build -b html docs/source/ docs/build/html

Contributing

I am new to Open Source Development, so open to suggestions on this section. The best contributions would be to make the core bot stronger by tinkering with the weights of each of the hand-crafted features in experimental/machine_learning/players/minimax.py, or coming up with new hand-crafted features! In particular, you can edit the CONTENDER_WEIGHTS and/or contender_fn function and run a command like: python experimental/play.py --players=AB:2:False:C,AB:2 --num=100 to see if your changes improve the main bot.

Here is also a list of ideas:

• Improve catanatron package running time performance.

  • Continue refactoring the State to be more and more like a primitive dict or array. (Copies are much faster if State is just a native python object).
  • Move RESOURCE to be ints. Python enums turned out to be slow for hashing and using.
  • Move .actions to a Game concept. (to avoid copying when copying State)
  • Remove .current_prompt. It seems its redundant with (is_moving_knight, etc...) and not needed.

• Improve AlphaBetaPlayer:

  • Explore and improve prunning
  • Use Bayesian Methods or SPSA to tune weights and find better ones.

• Experiment ideas:

  • DQN Render Method. Use models/mbs=64__1619973412.model. Try to understand it.
  • DQN Two Layer Algo. With Simple Action Space.
  • Simple Alpha Go
  • Try Tensorforce with simple action space.
  • Try simple flat CSV approach but with AlphaBeta-generated games.
  • Visualize tree with graphviz. With colors per maximizing/minimizing.

• Bugs:

  • Shouldn't be able to use dev card just bought.

• Features:

  • Continue implementing actions from the UI (not all implemented).
  • Chess.com-like UI for watching game replays (with Play/Pause and Back/Forward).
  • A terminal UI? (for ease of debugging)