8c17debc9dc6552da8c317c0c852b55f03c80c61,slm_lab/agent/algorithm/dqn.py,DQNBase,update,#DQNBase#,146

Before Change

        """Update epsilon or boltzmann for policy after net training"""
        epi = util.s_get(self, "aeb_space.clock").get("e")
        rise = self.explore_var_end - self.explore_var_start
        slope = rise / float(self.explore_anneal_epi)
        self.explore_var = max(
            slope * (epi - 1) + self.explore_var_start, self.explore_var_end)
        // print(f"Explore var: {self.explore_var}")

After Change

                logger.debug("Updating target_net by replacing")
                self.target_net = deepcopy(self.net)
        elif self.update_type == "polyak":
            logger.debug("Updating net by averaging")
            avg_params = self.polyak_weight * net_util.flatten_params(self.target_net) + \
                (1 - self.polyak_weight) * net_util.flatten_params(self.net)
            self.target_net = net_util.load_params(self.target_net, avg_params)
        else:

Instances