agent:
  classname:           algorithms.marl.networks.RecurrentAC
  n_agents:            2
  obs_emb_size:        96
  action_emb_size:     16
  hidden_size_actor:   64
  hidden_size_critic:  64
  use_agent_embedding: False
env:
  classname:          environments.factory.make
  env_name:           "DirtyFactory-v0"
  n_agents:           2
  max_steps:          250
  pomdp_r:            2
  stack_n_frames:     0
  individual_rewards: True
method:               algorithms.marl.LoopSEAC
algorithm:
  gamma:              0.99
  entropy_coef:       0.01
  vf_coef:            0.5
  n_steps:            5
  max_steps:          1000000