e4b6611cb73ef7658f028831be1aa6bd85ecbed0,src/garage/tf/policies/gaussian_mlp_policy.py,GaussianMLPPolicy,get_action,#GaussianMLPPolicy#Any#,168

Before Change

        
        sample, mean, log_std = self._f_dist(np.expand_dims([observation], 1))
        sample = self.action_space.unflatten(np.squeeze(sample, 1)[0])
        mean = self.action_space.unflatten(np.squeeze(mean, 1)[0])
        log_std = self.action_space.unflatten(np.squeeze(log_std, 1)[0])
        return sample, dict(mean=mean, log_std=log_std)

    def get_actions(self, observations):
        Get multiple actions from this policy for the input observations.

After Change

                distribution.

        
        actions, agent_infos = self.get_actions([observation])
        return actions[0], {k: v[0] for k, v in agent_infos.items()}

    def get_actions(self, observations):
        Get multiple actions from this policy for the input observations.

Instances