!!python/object/apply:collections.OrderedDict
- - - batch_size
    - 32
  - - buffer_size
    - 100000
  - - exploration_final_eps
    - 0.01
  - - exploration_fraction
    - 0.1
  - - gradient_steps
    - 1
  - - learning_rate
    - 0.0001
  - - learning_starts
    - 1000
  - - n_timesteps
    - 100000.0
  - - optimize_memory_usage
    - false
  - - policy
    - MlpPolicy
  - - target_update_interval
    - 1000
  - - train_freq
    - 4