示例#1
0
def load_policy(policy_cfg, action_spec):
    """Loads policy based on config."""
    if policy_cfg.ptype not in PTYPES:
        raise ValueError('Unknown policy type %s.' % policy_cfg.ptype)
    if policy_cfg.ptype == 'randwalk':
        policy = policies.ContinuousRandomPolicy(action_spec)
    elif policy_cfg.ptype in ['randinit', 'load']:
        a_net = networks.ActorNetwork(action_spec,
                                      fc_layer_params=policy_cfg.model_params)
        if policy_cfg.ptype == 'load':
            logging.info('Loading policy from %s...', policy_cfg.ckpt)
            policy_ckpt = tf.train.Checkpoint(policy=a_net)
            policy_ckpt.restore(policy_cfg.ckpt).expect_partial()
        policy = wrap_policy(a_net, policy_cfg.wrapper)
    return policy
 def p_net_factory():
   return networks.ActorNetwork(
       action_spec,
       fc_layer_params=model_params[0])