policy:
  input_dim: 4
  output_dim: 2
  hidden_sizes: [64, 64]

ppo:
  learning_rate: 3e-4
  n_steps: 2048
  batch_size: 64
  n_epochs: 10
  gamma: 0.99
  gae_lambda: 0.95
  clip_range: 0.2
  total_timesteps: 1000000
  eval_interval: 2048
  eval_deterministic: true
  eval_episodes: 10
  seed: 42

loggers:
  - type: terminal
  - type: wandb
    project: "PPO_project"
    entity: "your_entity"
    push_interval: 10