Quentin Gallouédec
Initial commit
db5318d
!!python/object/apply:collections.OrderedDict
- - - batch_size
- 256
- - gradient_steps
- 1
- - learning_rate
- 0.0003
- - learning_starts
- 10000
- - n_timesteps
- 1000000.0
- - noise_std
- 0.1
- - noise_type
- normal
- - policy
- MlpPolicy
- - train_freq
- 1