def env_creator(env_config):
# env = MultiCarlaEnv(env_config) # (env_actor_configs)
env = StopSignUrbanIntersection3Car() # Urban2Car()
# Apply wrappers to: convert to Grayscale, resize to 84 x 84,
# stack frames & some more op
env = wrap_deepmind(env, dim=84, num_framestack=num_framestack)
return env
def env_creator(env_config):
import macad_gym
env = gym.make("HomoNcomIndePOIntrxMASS3CTWN3-v0")
# Apply wrappers to: convert to Grayscale, resize to 84 x 84,
# stack frames & some more op
env = wrap_deepmind(env, dim=84, num_framestack=num_framestack)
return env
def env_creator(env_config):
# NOTES: env_config.worker_index & vector_index are useful for
# curriculum learning or joint training experiments
# env = MultiCarlaEnv(env_config) # (env_actor_configs)
env = StopSignUrbanIntersection3Car() # Urban2Car()
# Apply wrappers to: convert to Grayscale, resize to 84 x 84,
# stack frames & some more op
env = wrap_deepmind(env, dim=84, num_framestack=num_framestack)
return env
def env_creator(env_config):
# NOTES: env_config.worker_index & vector_index are useful for
# curriculum learning or joint training experiments
configs = DEFAULT_MULTIENV_CONFIG
configs["render"] = False
env = MultiCarlaEnv(configs)
# Apply wrappers to: convert to Grayscale, resize to 84 x 84,
# stack frames & some more op
env = wrap_deepmind(env, dim=84, num_framestack=num_framestack)
return env
def env_creator(env_config):
# NOTES: env_config.worker_index & vector_index are useful for
# curriculum learning or joint training experiments
import macad_gym
env = gym.make("HomoNcomIndePOIntrxMASS3CTWN3-v0")
# Apply wrappers to: convert to Grayscale, resize to 84 x 84,
# stack frames & some more op
env = wrap_deepmind(env, dim=84, num_framestack=num_framestack)
return env
