https://github.com/hrshl212/td3-libtorch

TD3 reinforcement learning algorithm using libtorch in simple environment
https://github.com/hrshl212/td3-libtorch

deep-reinforcement-learning libtorch pytorch reinforcement-learning

Last synced: about 1 month ago
JSON representation

TD3 reinforcement learning algorithm using libtorch in simple environment

• Host: GitHub
• URL: https://github.com/hrshl212/td3-libtorch
• Owner: hrshl212
• Created: 2022-02-09T10:34:00.000Z (over 4 years ago)
• Default Branch: main
• Last Pushed: 2024-06-11T19:58:24.000Z (about 2 years ago)
• Last Synced: 2024-06-12T04:42:04.400Z (about 2 years ago)
• Topics: deep-reinforcement-learning, libtorch, pytorch, reinforcement-learning
• Language: Makefile
• Homepage:
• Size: 19.6 MB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# TD3-Pytorch-C++-

TD3 deep reinforcement learning algorithm using C++/libtorch in simple environment of finding shortest distance between two points. The environment is taken from https://github.com/mhubii/ppo_libtorch

This is an implementation of the twin-delayed deep deterministic (TD3) policy gradient algorithm for the C++ API of Pytorch. It uses a simple TestEnvironment to test the algorithm. Below is a small visualization of the environment, the algorithm is tested in.

![This is an image](/img/epoch_30.gif)

Fig. 1: The agent in training during epoch 30.

You first need to install LibTorch. I used the version 1.4.0+cpu.

Do

```

mkdir build

cd build

cmake -DCMAKE_PREFIX_PATH=/absolut/path/to/libtorch ..

make

```

# Run

Run the executable with

```

cd build

./train_ppo

```

It should produce something like shown below.

![This is an image](/img/epoch_4.gif)

![This is an image](/img/epoch_11.gif)

Fig. 2: From left to right, the agent for successive epochs in training mode as it takes actions in the environment to reach the goal.

# Visualization

The results are saved to data/data.csv and can be visualized by running python plot.py.