metadata
tags:
- Pixelcopter-PLE-v0
- reinforce
- reinforcement-learning
- custom-implementation
- deep-rl-class
model-index:
- name: REINFORCE-PixelCopter
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: Pixelcopter-PLE-v0
type: Pixelcopter-PLE-v0
metrics:
- type: mean_reward
value: 6.50 +/- 5.99
name: mean_reward
verified: false
Reinforce Agent playing Pixelcopter-PLE-v0
This is a trained model of a Reinforce agent playing Pixelcopter-PLE-v0 .
Portable weights: model_state_dict.pt. Define the matching Unit 4 Policy
(CartPole: two layers; PixelCopter: three layers), instantiate it with
state_space, action_space, and h_size from hyperparameters.json, then
use policy.load_state_dict(torch.load(weights_path, map_location="cpu", weights_only=True)).
The legacy model.pt contains a full Python model and requires its matching class.
To learn to use this model and train yours check Unit 4 of the Deep Reinforcement Learning Course: https://huggingface.co/deep-rl-course/unit4/introduction