Skip to content

This is the code for "OpenAI Five vs DOTA 2 Explained" By Siraj Raval on Youtube

License

Notifications You must be signed in to change notification settings

guerraepaz/OpenAI_Five_vs_Dota2_Explained

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Overview

This is the code for this video on Youtube by Siraj Raval on OpenAI Five vs DOTA 2. The author of this code is alexis-jacq. The real code is not yet publically available, but this is a basic version of the algorithm.

Dependencies

  • PyTorch
  • OpenAI Gym

Usage

Run 'python main.py (gym_environment_name)' in terminal

Pytorch-DPPO

Pytorch implementation of Distributed Proximal Policy Optimization: https://arxiv.org/abs/1707.02286 Using PPO with clip loss (from https://arxiv.org/pdf/1707.06347.pdf).

I finally fixed what was wrong with the gradient descent step, using previous log-prob from rollout batches. At least ppo.py is fixed, the rest is going to be corrected as well very soon.

In the following example I was not patient enough to wait for million iterations, I just wanted to check if the model is properly learning:

Progress of single PPO:

InvertedPendulum

InvertedPendulum

InvertedDoublePendulum

InvertedDoublePendulum

HalfCheetah

HalfCheetah

hopper (PyBullet)

hopper (PyBullet)

halfcheetah (PyBullet)

halfcheetah (PyBullet)

Progress of DPPO (4 agents) [TODO]

Acknowledgments

The structure of this code is based on https://github.com/ikostrikov/pytorch-a3c.

Hyperparameters and loss computation has been taken from https://github.com/openai/baselines

About

This is the code for "OpenAI Five vs DOTA 2 Explained" By Siraj Raval on Youtube

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%