(self)
| 104 | self.test_dict_log[key].append(val) |
| 105 | |
| 106 | def get_worker_logs(self): |
| 107 | if len(self.reward_log) > 0: |
| 108 | ori_reward = sum(self.ori_reward_log) / len(self.ori_reward_log) |
| 109 | reward = sum(self.reward_log) / len(self.reward_log) |
| 110 | reward_max = sum(self.reward_max_log) / len(self.reward_max_log) |
| 111 | eps_lengths = sum(self.eps_lengths) / len(self.eps_lengths) |
| 112 | eps_lengths_max = sum(self.eps_lengths_max) / len(self.eps_lengths_max) |
| 113 | temperature = sum(self.temperature_log) / len(self.temperature_log) |
| 114 | visit_entropy = sum(self.visit_entropies_log) / len(self.visit_entropies_log) |
| 115 | priority_self_play = sum(self.priority_self_play_log) / len(self.priority_self_play_log) |
| 116 | distributions = self.distributions_log |
| 117 | |
| 118 | self.ori_reward_log = [] |
| 119 | self.reward_log = [] |
| 120 | self.reward_max_log = [] |
| 121 | self.eps_lengths = [] |
| 122 | self.eps_lengths_max = [] |
| 123 | self.temperature_log = [] |
| 124 | self.visit_entropies_log = [] |
| 125 | self.priority_self_play_log = [] |
| 126 | self.distributions_log = {} |
| 127 | |
| 128 | else: |
| 129 | ori_reward = None |
| 130 | reward = None |
| 131 | reward_max = None |
| 132 | eps_lengths = None |
| 133 | eps_lengths_max = None |
| 134 | temperature = None |
| 135 | visit_entropy = None |
| 136 | priority_self_play = None |
| 137 | distributions = None |
| 138 | |
| 139 | if len(self.test_dict_log) > 0: |
| 140 | test_dict = self.test_dict_log |
| 141 | |
| 142 | self.test_dict_log = {} |
| 143 | test_counter = self.test_counter |
| 144 | else: |
| 145 | test_dict = None |
| 146 | test_counter = None |
| 147 | |
| 148 | return ori_reward, reward, reward_max, eps_lengths, eps_lengths_max, test_counter, test_dict, temperature, visit_entropy, priority_self_play, distributions |
nothing calls this directly
no outgoing calls
no test coverage detected