shivaverma / Orbit
Licence: mit
Open source collection of Reinforcement Learning Environments.
Stars: ✭ 44
Programming Languages
python
139335 projects - #7 most used programming language
Projects that are alternatives of or similar to Orbit
Bullet3
Bullet Physics SDK: real-time collision detection and multi-physics simulation for VR, games, visual effects, robotics, machine learning etc.
Stars: ✭ 8,714 (+19704.55%)
Mutual labels: game-development, reinforcement-learning
Deepqlearning.jl
Implementation of the Deep Q-learning algorithm to solve MDPs
Stars: ✭ 38 (-13.64%)
Mutual labels: reinforcement-learning
Fullstackmachinelearning
Mostly free resources for end-to-end machine learning engineering, including open courses from CalTech, Columbia, Berkeley, MIT, and Stanford (in alphabetical order).
Stars: ✭ 39 (-11.36%)
Mutual labels: reinforcement-learning
Qualia2.0
Qualia is a deep learning framework deeply integrated with automatic differentiation and dynamic graphing with CUDA acceleration. Qualia was built from scratch.
Stars: ✭ 41 (-6.82%)
Mutual labels: reinforcement-learning
Bolero
Behavior Optimization and Learning for Robots
Stars: ✭ 39 (-11.36%)
Mutual labels: reinforcement-learning
Game Dogfight
Air to air combat game, created in Python 3 using HARFANG 3D.
Stars: ✭ 41 (-6.82%)
Mutual labels: game-development
Endless Runner Entitas Ecs
Runner template for Unity
Stars: ✭ 41 (-6.82%)
Mutual labels: game-development
Minilua
Single-file port of Lua, a powerful scripting language.
Stars: ✭ 41 (-6.82%)
Mutual labels: game-development
Rengfx
lightweight, expressive, extensible 2D/3D game engine
Stars: ✭ 41 (-6.82%)
Mutual labels: game-development
Deep Q Learning
Minimal Deep Q Learning (DQN & DDQN) implementations in Keras
Stars: ✭ 1,013 (+2202.27%)
Mutual labels: reinforcement-learning
Phaser Node Kit
Rapid Game Development with PhaserJS and Node for Modern Browsers
Stars: ✭ 39 (-11.36%)
Mutual labels: game-development
Blocks
Blocks World -- Simulator, Code, and Models (Misra et al. EMNLP 2017)
Stars: ✭ 39 (-11.36%)
Mutual labels: reinforcement-learning
Awesome Mlss
List of summer schools in machine learning + related fields across the globe
Stars: ✭ 1,001 (+2175%)
Mutual labels: reinforcement-learning
Deepbootcamp
Solved lab problems, slides and notes of the Deep Reinforcement Learning bootcamp 2017 held at UCBerkeley
Stars: ✭ 39 (-11.36%)
Mutual labels: reinforcement-learning
Simpleton Engine
What a stupid name for a library
Stars: ✭ 42 (-4.55%)
Mutual labels: game-development
Tf2.0 reinforcement learning
Basic reinforcement learning implementation with tensorflow version 2.0
Stars: ✭ 43 (-2.27%)
Mutual labels: reinforcement-learning
Gym Gridworlds
Gridworld environments for OpenAI gym.
Stars: ✭ 43 (-2.27%)
Mutual labels: reinforcement-learning
Tank island
Top down 2D shooter game that involves blowing up tanks
Stars: ✭ 42 (-4.55%)
Mutual labels: game-development
Machine Learning From Scratch
Succinct Machine Learning algorithm implementations from scratch in Python, solving real-world problems (Notebooks and Book). Examples of Logistic Regression, Linear Regression, Decision Trees, K-means clustering, Sentiment Analysis, Recommender Systems, Neural Networks and Reinforcement Learning.
Stars: ✭ 42 (-4.55%)
Mutual labels: reinforcement-learning
Orbit
Orbit is a open-source project and a collection of reinforcement learning environments. If you want to contribute to this project by creating your own environment then please drop a mail to [email protected]. You can know how to create an environment through this blog:
Requirements
-
python-3.7 -
keras-2.4.3 -
tensorflow-2.2.0
Checkout random_policy.py to see how to use an environment.
Orbit contains following RL environments as of today.
Environment 1: Paddle
Author: Shiva Verma
The task is to take the ball on paddle. Inbuilt turtle library is used to create the environment.
Action space (3)
-
0- move paddle to left -
1- do nothing -
2- move paddle to right
State space (5)
- x position of paddle
- x and y position of ball
- x and y velocity of ball
Reward function
-
+3.0- when paddle hit the ball -
-3.0- when ball touchs the ground -
-0.1- when paddle moves
Episode termination
- Episode ends when ball touchs the ground.
Environment 2: Jump
Author: Vinod Kumar
The task to dodge the kangaroo from moving balls. Inbuilt turtle library is used to create the environment.
Action space (2)
-
0- jump -
1- do nothing
State space (13)
- y position of kangaroo
- x and y position of all moving ball
Reward function
-
+5.0- when kangaroo dodge a ball -
-1.0- for each jump -
+0.1- for each time step
Episode termination
- Episode ends when kangaroo touchs a ball.
Environment 3: Cannon
Author: Sathish Kumar
The task to hit the dropping balls. pygame library is used to create the environment.
Action space (4)
-
0- rotate left -
1- rotate right -
2- shoot bullet -
3- do nothing
State space (8)
- x position of cannon
- x and y position of all balls
- sin of angle of cannon
Reward function
-
+5.0- hitting the ball -
-0.5- shooting bullet -
-0.1- rotating canon -
-5.0- episode termination
Episode termination
- Episode ends when a ball touchs the line.
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].