|
--- |
|
tags: |
|
- CartPole-v1 |
|
- reinforce |
|
- reinforcement-learning |
|
- custom-implementation |
|
- deep-rl-class |
|
model-index: |
|
- name: reinforce-CartPole-v2 |
|
results: |
|
- task: |
|
type: reinforcement-learning |
|
name: reinforcement-learning |
|
dataset: |
|
name: CartPole-v1 |
|
type: CartPole-v1 |
|
metrics: |
|
- type: mean_reward |
|
value: 221.40 +/- 16.12 |
|
name: mean_reward |
|
verified: false |
|
--- |
|
|
|
## parameters |
|
cartpole_hyperparameters = { <br> |
|
"h_size": 16,<br> |
|
"n_training_episodes": 1000, <br> |
|
"n_evaluation_episodes": 10, <br> |
|
"max_t": 1000, <br> |
|
"gamma": 1.0, <br> |
|
"lr": 1e-2, <br> |
|
"env_id": env_id, <br> |
|
"state_space": s_size, <br> |
|
"action_space": a_size, <br> |
|
} |
|
|
