metadata
tags:
- Pixelcopter-PLE-v0
- reinforce
- reinforcement-learning
- custom-implementation
- deep-rl-class
model-index:
- name: Reinforce-PixelCopter-v2
results:
- task:
type: reinforcement-learning
name: reinforcement-learning
dataset:
name: Pixelcopter-PLE-v0
type: Pixelcopter-PLE-v0
metrics:
- type: mean_reward
value: 18.50 +/- 21.02
name: mean_reward
verified: false
pixelcopter_hyperparameters = {
"h_size": 64,
"n_training_episodes": 50000,
"n_evaluation_episodes": 10,
"max_t": 10000,
"gamma": 0.99,
"lr": 1e-4,
"env_id": env_id,
"state_space": s_size,
"action_space": a_size,
}