861657d2c9b321961994c8cdd0e58b6c4fe0645f,slm_lab/agent/algorithm/reinforce.py,Reinforce,update,#Reinforce#,166

Before Change

    @lab_api
    def update(self):
        space_clock = util.s_get(self, "aeb_space.clock")
        for net in [self.net]:
            net.update_lr(space_clock)
        explore_vars = [self.action_policy_update(self, body) for body in self.agent.nanflat_body_a]
        explore_var_a = self.nanflat_to_data_a("explore_var", explore_vars)
        return explore_var_a

After Change

        return policy_loss

    @lab_api
    def update(self):
        for net_name in self.net_names:
            net = getattr(self, net_name)
            net.update_lr(self.body.env.clock)
        explore_var = self.action_policy_update(self, self.body)
        return explore_var

Instances