Giter Site home page Giter Site logo

halite's Introduction

Two Sigma Halite Challenge Code Repo

I developed a Gym environment for the Halite environment in order to train various reinforcement learning algorithms. For fast experimentation and testing these algorithms, I am using Ray a distributed computing library with the RLLib library which contains reinforcement learning components. Each of the algorithms in Ray can be scaled to a cluster. Therefore, I can perform distributed environment data collection and iterate faster to a solution.

Some of the algorithms I am testing are:

  • DQN
  • APEX DQN
  • PPO

Additionally, I am testing out using Population Based Training in order to tune hyperparameters in one training cycle. Also, I worked on creating an autoencoder for different board distributions of Halite, which can be found in the analysis directory.

Installation

  1. Fill in necessary information in the ray-cluster.yaml file, this contains information such as what cloud provider you are using, what images to launch and how many to launch
  2. Go to GCP and create the images for the Head and Worker, make sure to clone this repo and install the requirements.txt
  3. ray up ray-cluster.yaml
  4. Use the ssh command from the result of the last command
  5. Launch one of the various algorithms and it will run on the distributed cluster you set up

For more information on setup and the Ray environment checkout the ray project.

Authors

halite's People

Contributors

anthonytec2 avatar brianwilcox1 avatar dependabot[bot] avatar

Stargazers

 avatar  avatar

Watchers

 avatar  avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.