Skip to content

kitrakrev

u8-LunarLander-v2

kitrakrev/u8-LunarLander-v2

u8-LunarLander-v2 is a reinforcement learning model from kitrakrev. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product.

This is a trained model of a PPO agent playing LunarLander-v2.

Downloads · 30 days

0

Access

Public

Updated May 29, 2023

Repo size

153 KB

Likes

0

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.0110 KB · 62%

At a glance

Task
Reinforcement Learning
Access
Public
Created
May 29, 2023
Updated
May 29, 2023
SHA
3d94e7e2
Task
Reinforcement Learning
Created
May 29, 2023
Updated
May 29, 2023