| {"update": 1, "steps_done": 2048, "mean_step_reward": -0.348, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 379, "mean_ep_return": -1.892, "mean_ep_length": 5.4, "mean_collisions": 0.21, "mean_intent_rate": 0.14, "policy_loss": -0.011414, "value_loss": 14.800392, "entropy": 1.604407, "total_loss": 7.372738, "clip_frac": 0.084717, "sps": 370.1, "elapsed_s": 5.5} |
| {"update": 2, "steps_done": 4096, "mean_step_reward": -0.2652, "min_step_reward": -10.5887, "max_step_reward": 2.5, "n_episodes_finished": 468, "mean_ep_return": -1.139, "mean_ep_length": 4.4, "mean_collisions": 0.13, "mean_intent_rate": 0.157, "policy_loss": -0.021349, "value_loss": 6.539419, "entropy": 1.560679, "total_loss": 3.232754, "clip_frac": 0.194092, "sps": 414.9, "elapsed_s": 10.5} |
| {"update": 3, "steps_done": 6144, "mean_step_reward": -0.2379, "min_step_reward": -10.5873, "max_step_reward": 2.5, "n_episodes_finished": 605, "mean_ep_return": -0.815, "mean_ep_length": 3.4, "mean_collisions": 0.09, "mean_intent_rate": 0.129, "policy_loss": -0.016879, "value_loss": 8.450987, "entropy": 1.518354, "total_loss": 4.193431, "clip_frac": 0.184814, "sps": 401.0, "elapsed_s": 15.6} |
| {"update": 4, "steps_done": 8192, "mean_step_reward": -0.1902, "min_step_reward": -10.5887, "max_step_reward": 2.5, "n_episodes_finished": 639, "mean_ep_return": -0.61, "mean_ep_length": 3.2, "mean_collisions": 0.06, "mean_intent_rate": 0.131, "policy_loss": -0.016117, "value_loss": 4.027927, "entropy": 1.502632, "total_loss": 1.98282, "clip_frac": 0.094482, "sps": 393.4, "elapsed_s": 20.8} |
| {"update": 5, "steps_done": 10240, "mean_step_reward": -0.0553, "min_step_reward": -10.5869, "max_step_reward": 2.5, "n_episodes_finished": 683, "mean_ep_return": -0.165, "mean_ep_length": 3.0, "mean_collisions": 0.04, "mean_intent_rate": 0.155, "policy_loss": -0.025628, "value_loss": 3.718626, "entropy": 1.476338, "total_loss": 1.818921, "clip_frac": 0.185425, "sps": 395.7, "elapsed_s": 26.0} |
| {"update": 6, "steps_done": 12288, "mean_step_reward": -0.0235, "min_step_reward": -10.5872, "max_step_reward": 2.5, "n_episodes_finished": 558, "mean_ep_return": -0.088, "mean_ep_length": 3.7, "mean_collisions": 0.04, "mean_intent_rate": 0.161, "policy_loss": -0.019751, "value_loss": 3.175081, "entropy": 1.424359, "total_loss": 1.553546, "clip_frac": 0.191528, "sps": 400.8, "elapsed_s": 31.1} |
| {"update": 7, "steps_done": 14336, "mean_step_reward": 0.0978, "min_step_reward": -10.5872, "max_step_reward": 2.5, "n_episodes_finished": 498, "mean_ep_return": 0.407, "mean_ep_length": 4.1, "mean_collisions": 0.03, "mean_intent_rate": 0.198, "policy_loss": -0.025995, "value_loss": 3.138277, "entropy": 1.383313, "total_loss": 1.52931, "clip_frac": 0.213257, "sps": 406.0, "elapsed_s": 36.1} |
| {"update": 8, "steps_done": 16384, "mean_step_reward": 0.1186, "min_step_reward": -10.5872, "max_step_reward": 2.5, "n_episodes_finished": 432, "mean_ep_return": 0.558, "mean_ep_length": 4.8, "mean_collisions": 0.03, "mean_intent_rate": 0.205, "policy_loss": -0.014142, "value_loss": 2.915338, "entropy": 1.325493, "total_loss": 1.430272, "clip_frac": 0.234619, "sps": 408.4, "elapsed_s": 41.1} |
| {"update": 9, "steps_done": 18432, "mean_step_reward": 0.2548, "min_step_reward": -10.5872, "max_step_reward": 2.5, "n_episodes_finished": 440, "mean_ep_return": 1.19, "mean_ep_length": 4.6, "mean_collisions": 0.02, "mean_intent_rate": 0.223, "policy_loss": -0.020199, "value_loss": 3.416153, "entropy": 1.212075, "total_loss": 1.675757, "clip_frac": 0.369995, "sps": 409.2, "elapsed_s": 46.1} |
| {"update": 10, "steps_done": 20480, "mean_step_reward": 0.3284, "min_step_reward": -10.5856, "max_step_reward": 2.5, "n_episodes_finished": 347, "mean_ep_return": 1.916, "mean_ep_length": 5.9, "mean_collisions": 0.01, "mean_intent_rate": 0.257, "policy_loss": -0.017617, "value_loss": 4.098948, "entropy": 1.181559, "total_loss": 2.020041, "clip_frac": 0.215576, "sps": 414.2, "elapsed_s": 51.1} |
| {"update": 11, "steps_done": 22528, "mean_step_reward": 0.3089, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 293, "mean_ep_return": 2.173, "mean_ep_length": 6.9, "mean_collisions": 0.03, "mean_intent_rate": 0.272, "policy_loss": -0.013786, "value_loss": 4.842874, "entropy": 1.119149, "total_loss": 2.39646, "clip_frac": 0.171753, "sps": 417.0, "elapsed_s": 56.0} |
| {"update": 12, "steps_done": 24576, "mean_step_reward": 0.3173, "min_step_reward": -10.587, "max_step_reward": 2.5, "n_episodes_finished": 207, "mean_ep_return": 3.124, "mean_ep_length": 10.0, "mean_collisions": 0.07, "mean_intent_rate": 0.292, "policy_loss": -0.009881, "value_loss": 6.725228, "entropy": 1.146327, "total_loss": 3.34127, "clip_frac": 0.09021, "sps": 415.8, "elapsed_s": 60.9} |
| {"update": 13, "steps_done": 26624, "mean_step_reward": 0.338, "min_step_reward": -10.5872, "max_step_reward": 2.5, "n_episodes_finished": 186, "mean_ep_return": 3.715, "mean_ep_length": 10.9, "mean_collisions": 0.06, "mean_intent_rate": 0.294, "policy_loss": -0.005616, "value_loss": 8.194549, "entropy": 1.162722, "total_loss": 4.080031, "clip_frac": 0.165283, "sps": 424.0, "elapsed_s": 65.8} |
| {"update": 14, "steps_done": 28672, "mean_step_reward": 0.2945, "min_step_reward": -10.5798, "max_step_reward": 2.5, "n_episodes_finished": 117, "mean_ep_return": 5.192, "mean_ep_length": 17.6, "mean_collisions": 0.17, "mean_intent_rate": 0.309, "policy_loss": -0.003457, "value_loss": 12.368031, "entropy": 1.184173, "total_loss": 6.168717, "clip_frac": 0.06665, "sps": 424.4, "elapsed_s": 70.6} |
| {"update": 15, "steps_done": 30720, "mean_step_reward": 0.4353, "min_step_reward": -10.5845, "max_step_reward": 2.5, "n_episodes_finished": 80, "mean_ep_return": 11.17, "mean_ep_length": 25.6, "mean_collisions": 0.11, "mean_intent_rate": 0.329, "policy_loss": -0.004427, "value_loss": 18.76897, "entropy": 1.124451, "total_loss": 9.368814, "clip_frac": 0.071289, "sps": 427.3, "elapsed_s": 75.4} |
| {"update": 16, "steps_done": 32768, "mean_step_reward": 0.3929, "min_step_reward": -10.5869, "max_step_reward": 2.5, "n_episodes_finished": 90, "mean_ep_return": 8.871, "mean_ep_length": 22.7, "mean_collisions": 0.16, "mean_intent_rate": 0.319, "policy_loss": -0.002429, "value_loss": 13.148047, "entropy": 1.095741, "total_loss": 6.560637, "clip_frac": 0.116333, "sps": 427.8, "elapsed_s": 80.2} |
| {"update": 17, "steps_done": 34816, "mean_step_reward": 0.3346, "min_step_reward": -10.5315, "max_step_reward": 2.5, "n_episodes_finished": 96, "mean_ep_return": 7.211, "mean_ep_length": 21.3, "mean_collisions": 0.26, "mean_intent_rate": 0.314, "policy_loss": -0.002881, "value_loss": 18.177666, "entropy": 1.08489, "total_loss": 9.075103, "clip_frac": 0.074097, "sps": 430.0, "elapsed_s": 84.9} |
| {"update": 18, "steps_done": 36864, "mean_step_reward": 0.433, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 35, "mean_ep_return": 21.503, "mean_ep_length": 54.3, "mean_collisions": 0.29, "mean_intent_rate": 0.319, "policy_loss": -0.001728, "value_loss": 27.83451, "entropy": 1.032119, "total_loss": 13.905206, "clip_frac": 0.049561, "sps": 435.0, "elapsed_s": 89.6} |
| {"update": 19, "steps_done": 38912, "mean_step_reward": 0.2587, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 29, "mean_ep_return": 22.715, "mean_ep_length": 70.5, "mean_collisions": 1.69, "mean_intent_rate": 0.35, "policy_loss": 0.015874, "value_loss": 39.265742, "entropy": 1.00197, "total_loss": 19.638725, "clip_frac": 0.285156, "sps": 436.5, "elapsed_s": 94.3} |
| {"update": 20, "steps_done": 40960, "mean_step_reward": 0.3424, "min_step_reward": -10.5869, "max_step_reward": 2.5, "n_episodes_finished": 31, "mean_ep_return": 19.879, "mean_ep_length": 68.4, "mean_collisions": 1.71, "mean_intent_rate": 0.356, "policy_loss": 0.004615, "value_loss": 28.486185, "entropy": 1.0439, "total_loss": 14.237268, "clip_frac": 0.171265, "sps": 428.2, "elapsed_s": 99.1} |
| {"update": 21, "steps_done": 43008, "mean_step_reward": 0.3262, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 33, "mean_ep_return": 22.426, "mean_ep_length": 60.5, "mean_collisions": 1.15, "mean_intent_rate": 0.353, "policy_loss": 0.014591, "value_loss": 34.854514, "entropy": 1.072688, "total_loss": 17.431121, "clip_frac": 0.293091, "sps": 435.3, "elapsed_s": 103.8} |
| {"update": 22, "steps_done": 45056, "mean_step_reward": 0.3785, "min_step_reward": -10.5852, "max_step_reward": 2.5, "n_episodes_finished": 25, "mean_ep_return": 30.567, "mean_ep_length": 85.0, "mean_collisions": 1.28, "mean_intent_rate": 0.334, "policy_loss": 0.008364, "value_loss": 29.914371, "entropy": 1.017498, "total_loss": 14.955375, "clip_frac": 0.293579, "sps": 439.0, "elapsed_s": 108.5} |
| {"update": 23, "steps_done": 47104, "mean_step_reward": 0.3003, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 27, "mean_ep_return": 22.466, "mean_ep_length": 76.3, "mean_collisions": 2.22, "mean_intent_rate": 0.367, "policy_loss": 0.001325, "value_loss": 41.77098, "entropy": 1.091941, "total_loss": 20.875896, "clip_frac": 0.070801, "sps": 436.7, "elapsed_s": 113.2} |
| {"update": 24, "steps_done": 49152, "mean_step_reward": 0.3458, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 22, "mean_ep_return": 32.611, "mean_ep_length": 93.1, "mean_collisions": 2.45, "mean_intent_rate": 0.38, "policy_loss": 0.003598, "value_loss": 47.700774, "entropy": 0.966522, "total_loss": 23.844319, "clip_frac": 0.209106, "sps": 433.4, "elapsed_s": 117.9} |
| {"update": 25, "steps_done": 51200, "mean_step_reward": 0.1774, "min_step_reward": -10.5862, "max_step_reward": 2.5, "n_episodes_finished": 23, "mean_ep_return": 17.58, "mean_ep_length": 85.5, "mean_collisions": 2.09, "mean_intent_rate": 0.327, "policy_loss": 0.006176, "value_loss": 32.264449, "entropy": 1.048376, "total_loss": 16.127916, "clip_frac": 0.210083, "sps": 443.0, "elapsed_s": 122.5} |
| {"update": 26, "steps_done": 53248, "mean_step_reward": 0.4061, "min_step_reward": -10.5352, "max_step_reward": 2.5, "n_episodes_finished": 24, "mean_ep_return": 32.503, "mean_ep_length": 88.0, "mean_collisions": 1.42, "mean_intent_rate": 0.337, "policy_loss": 0.000212, "value_loss": 27.56736, "entropy": 0.990011, "total_loss": 13.773992, "clip_frac": 0.174683, "sps": 435.8, "elapsed_s": 127.2} |
| {"update": 27, "steps_done": 55296, "mean_step_reward": 0.4383, "min_step_reward": -10.5608, "max_step_reward": 2.5, "n_episodes_finished": 18, "mean_ep_return": 51.011, "mean_ep_length": 111.2, "mean_collisions": 1.61, "mean_intent_rate": 0.365, "policy_loss": -0.000628, "value_loss": 30.977938, "entropy": 1.005548, "total_loss": 15.478285, "clip_frac": 0.113647, "sps": 435.4, "elapsed_s": 131.9} |
| {"update": 28, "steps_done": 57344, "mean_step_reward": 0.481, "min_step_reward": -10.5602, "max_step_reward": 2.5, "n_episodes_finished": 19, "mean_ep_return": 50.986, "mean_ep_length": 110.1, "mean_collisions": 0.63, "mean_intent_rate": 0.338, "policy_loss": -0.003708, "value_loss": 26.403724, "entropy": 0.978819, "total_loss": 13.188366, "clip_frac": 0.059814, "sps": 437.1, "elapsed_s": 136.6} |
| {"update": 29, "steps_done": 59392, "mean_step_reward": 0.4447, "min_step_reward": -10.5864, "max_step_reward": 2.5, "n_episodes_finished": 14, "mean_ep_return": 65.739, "mean_ep_length": 143.7, "mean_collisions": 2.36, "mean_intent_rate": 0.361, "policy_loss": 4e-06, "value_loss": 44.139854, "entropy": 0.902909, "total_loss": 22.060902, "clip_frac": 0.077148, "sps": 434.8, "elapsed_s": 141.3} |
| {"update": 30, "steps_done": 61440, "mean_step_reward": 0.3231, "min_step_reward": -10.5865, "max_step_reward": 2.5, "n_episodes_finished": 13, "mean_ep_return": 54.153, "mean_ep_length": 151.5, "mean_collisions": 3.69, "mean_intent_rate": 0.369, "policy_loss": 0.005068, "value_loss": 45.223323, "entropy": 0.891951, "total_loss": 22.607811, "clip_frac": 0.158569, "sps": 440.4, "elapsed_s": 146.0} |
| {"update": 31, "steps_done": 63488, "mean_step_reward": 0.4801, "min_step_reward": -10.5826, "max_step_reward": 2.5, "n_episodes_finished": 13, "mean_ep_return": 72.98, "mean_ep_length": 171.3, "mean_collisions": 2.92, "mean_intent_rate": 0.369, "policy_loss": 0.001024, "value_loss": 37.813348, "entropy": 0.94773, "total_loss": 18.898221, "clip_frac": 0.117065, "sps": 438.4, "elapsed_s": 150.6} |
| {"update": 32, "steps_done": 65536, "mean_step_reward": 0.3557, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 15, "mean_ep_return": 43.471, "mean_ep_length": 126.9, "mean_collisions": 4.2, "mean_intent_rate": 0.402, "policy_loss": 0.000844, "value_loss": 53.419978, "entropy": 0.923173, "total_loss": 26.701601, "clip_frac": 0.136475, "sps": 436.1, "elapsed_s": 155.3} |
| {"update": 33, "steps_done": 67584, "mean_step_reward": 0.5693, "min_step_reward": -10.5865, "max_step_reward": 2.5, "n_episodes_finished": 15, "mean_ep_return": 83.939, "mean_ep_length": 146.7, "mean_collisions": 1.87, "mean_intent_rate": 0.394, "policy_loss": 0.003718, "value_loss": 44.727893, "entropy": 0.926369, "total_loss": 22.358401, "clip_frac": 0.153687, "sps": 438.8, "elapsed_s": 160.0} |
| {"update": 34, "steps_done": 69632, "mean_step_reward": 0.4593, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 14, "mean_ep_return": 62.324, "mean_ep_length": 137.4, "mean_collisions": 5.29, "mean_intent_rate": 0.469, "policy_loss": -0.00339, "value_loss": 57.108087, "entropy": 0.900207, "total_loss": 28.541652, "clip_frac": 0.071289, "sps": 439.3, "elapsed_s": 164.7} |
| {"update": 35, "steps_done": 71680, "mean_step_reward": 0.4836, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 14, "mean_ep_return": 76.053, "mean_ep_length": 153.5, "mean_collisions": 3.86, "mean_intent_rate": 0.417, "policy_loss": 0.001167, "value_loss": 64.216389, "entropy": 0.902002, "total_loss": 32.100341, "clip_frac": 0.082397, "sps": 431.0, "elapsed_s": 169.4} |
| {"update": 36, "steps_done": 73728, "mean_step_reward": 0.5376, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 103.09, "mean_ep_length": 195.1, "mean_collisions": 3.6, "mean_intent_rate": 0.416, "policy_loss": 0.008763, "value_loss": 46.413742, "entropy": 0.887265, "total_loss": 23.206762, "clip_frac": 0.226196, "sps": 439.1, "elapsed_s": 174.1} |
| {"update": 37, "steps_done": 75776, "mean_step_reward": 0.4146, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 12, "mean_ep_return": 72.28, "mean_ep_length": 173.4, "mean_collisions": 4.5, "mean_intent_rate": 0.403, "policy_loss": 0.003322, "value_loss": 63.556824, "entropy": 0.832208, "total_loss": 31.773412, "clip_frac": 0.073853, "sps": 426.9, "elapsed_s": 178.9} |
| {"update": 38, "steps_done": 77824, "mean_step_reward": 0.442, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 82.089, "mean_ep_length": 188.0, "mean_collisions": 5.0, "mean_intent_rate": 0.406, "policy_loss": -0.000195, "value_loss": 52.361372, "entropy": 0.852122, "total_loss": 26.17197, "clip_frac": 0.047974, "sps": 439.8, "elapsed_s": 183.5} |
| {"update": 39, "steps_done": 79872, "mean_step_reward": 0.4858, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 93.724, "mean_ep_length": 177.7, "mean_collisions": 2.91, "mean_intent_rate": 0.399, "policy_loss": 0.000423, "value_loss": 42.278864, "entropy": 0.847473, "total_loss": 21.13138, "clip_frac": 0.092896, "sps": 433.4, "elapsed_s": 188.3} |
| {"update": 40, "steps_done": 81920, "mean_step_reward": 0.5156, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 12, "mean_ep_return": 90.896, "mean_ep_length": 170.5, "mean_collisions": 2.25, "mean_intent_rate": 0.388, "policy_loss": 0.001282, "value_loss": 49.620241, "entropy": 0.843239, "total_loss": 24.80297, "clip_frac": 0.110352, "sps": 437.3, "elapsed_s": 193.0} |
| {"update": 41, "steps_done": 83968, "mean_step_reward": 0.4472, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 12, "mean_ep_return": 74.515, "mean_ep_length": 182.7, "mean_collisions": 0.92, "mean_intent_rate": 0.32, "policy_loss": 0.004933, "value_loss": 32.398986, "entropy": 0.808579, "total_loss": 16.19634, "clip_frac": 0.055664, "sps": 431.5, "elapsed_s": 197.7} |
| {"update": 42, "steps_done": 86016, "mean_step_reward": 0.424, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 83.893, "mean_ep_length": 200.0, "mean_collisions": 0.9, "mean_intent_rate": 0.333, "policy_loss": -6.3e-05, "value_loss": 37.536108, "entropy": 0.813032, "total_loss": 18.759861, "clip_frac": 0.16687, "sps": 435.1, "elapsed_s": 202.4} |
| {"update": 43, "steps_done": 88064, "mean_step_reward": 0.345, "min_step_reward": -10.5827, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 68.553, "mean_ep_length": 175.5, "mean_collisions": 1.82, "mean_intent_rate": 0.342, "policy_loss": 0.005903, "value_loss": 40.777066, "entropy": 0.761103, "total_loss": 20.386825, "clip_frac": 0.140015, "sps": 436.0, "elapsed_s": 207.1} |
| {"update": 44, "steps_done": 90112, "mean_step_reward": 0.4202, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 69.839, "mean_ep_length": 186.8, "mean_collisions": 1.64, "mean_intent_rate": 0.324, "policy_loss": -0.003439, "value_loss": 35.912429, "entropy": 0.776619, "total_loss": 17.945009, "clip_frac": 0.127075, "sps": 435.0, "elapsed_s": 211.8} |
| {"update": 45, "steps_done": 92160, "mean_step_reward": 0.4979, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 12, "mean_ep_return": 90.78, "mean_ep_length": 183.8, "mean_collisions": 1.17, "mean_intent_rate": 0.357, "policy_loss": 0.000188, "value_loss": 41.10612, "entropy": 0.739093, "total_loss": 20.545857, "clip_frac": 0.054688, "sps": 433.3, "elapsed_s": 216.5} |
| {"update": 46, "steps_done": 94208, "mean_step_reward": 0.4952, "min_step_reward": -10.5614, "max_step_reward": 2.5, "n_episodes_finished": 13, "mean_ep_return": 75.897, "mean_ep_length": 148.9, "mean_collisions": 2.08, "mean_intent_rate": 0.355, "policy_loss": 0.009029, "value_loss": 45.746731, "entropy": 0.7672, "total_loss": 22.874722, "clip_frac": 0.199829, "sps": 427.3, "elapsed_s": 221.3} |
| {"update": 47, "steps_done": 96256, "mean_step_reward": 0.5472, "min_step_reward": -10.5593, "max_step_reward": 2.5, "n_episodes_finished": 13, "mean_ep_return": 83.454, "mean_ep_length": 162.2, "mean_collisions": 1.69, "mean_intent_rate": 0.377, "policy_loss": 0.006133, "value_loss": 43.342461, "entropy": 0.815698, "total_loss": 21.669206, "clip_frac": 0.15918, "sps": 433.5, "elapsed_s": 226.1} |
| {"update": 48, "steps_done": 98304, "mean_step_reward": 0.4869, "min_step_reward": -10.5817, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 95.143, "mean_ep_length": 200.0, "mean_collisions": 3.3, "mean_intent_rate": 0.393, "policy_loss": 0.002777, "value_loss": 49.2188, "entropy": 0.808097, "total_loss": 24.604096, "clip_frac": 0.161011, "sps": 438.0, "elapsed_s": 230.7} |
| {"update": 49, "steps_done": 100352, "mean_step_reward": 0.5016, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 102.76, "mean_ep_length": 195.5, "mean_collisions": 3.0, "mean_intent_rate": 0.405, "policy_loss": 0.003046, "value_loss": 48.824697, "entropy": 0.736507, "total_loss": 24.408029, "clip_frac": 0.106567, "sps": 440.3, "elapsed_s": 235.4} |
| {"update": 50, "steps_done": 102400, "mean_step_reward": 0.5744, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 12, "mean_ep_return": 95.091, "mean_ep_length": 159.4, "mean_collisions": 1.67, "mean_intent_rate": 0.394, "policy_loss": -0.002172, "value_loss": 40.01186, "entropy": 0.852681, "total_loss": 19.995231, "clip_frac": 0.097534, "sps": 439.5, "elapsed_s": 240.1} |
| {"update": 51, "steps_done": 104448, "mean_step_reward": 0.3725, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 12, "mean_ep_return": 65.663, "mean_ep_length": 177.6, "mean_collisions": 5.75, "mean_intent_rate": 0.404, "policy_loss": -0.000127, "value_loss": 56.673666, "entropy": 0.893356, "total_loss": 28.327773, "clip_frac": 0.129395, "sps": 443.9, "elapsed_s": 244.7} |
| {"update": 52, "steps_done": 106496, "mean_step_reward": 0.3456, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 65.381, "mean_ep_length": 193.7, "mean_collisions": 7.8, "mean_intent_rate": 0.423, "policy_loss": 0.000125, "value_loss": 86.751001, "entropy": 0.865089, "total_loss": 43.366975, "clip_frac": 0.140015, "sps": 439.6, "elapsed_s": 249.3} |
| {"update": 53, "steps_done": 108544, "mean_step_reward": 0.5924, "min_step_reward": -10.5818, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 114.195, "mean_ep_length": 200.0, "mean_collisions": 3.27, "mean_intent_rate": 0.422, "policy_loss": 0.006993, "value_loss": 48.283552, "entropy": 0.8885, "total_loss": 24.139884, "clip_frac": 0.128906, "sps": 435.2, "elapsed_s": 254.0} |
| {"update": 54, "steps_done": 110592, "mean_step_reward": 0.5512, "min_step_reward": -10.5848, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 105.368, "mean_ep_length": 185.5, "mean_collisions": 2.82, "mean_intent_rate": 0.417, "policy_loss": 0.004923, "value_loss": 51.992232, "entropy": 0.88264, "total_loss": 25.992213, "clip_frac": 0.073853, "sps": 436.9, "elapsed_s": 258.7} |
| {"update": 55, "steps_done": 112640, "mean_step_reward": 0.6373, "min_step_reward": -10.5874, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 122.23, "mean_ep_length": 200.0, "mean_collisions": 3.0, "mean_intent_rate": 0.432, "policy_loss": 0.006491, "value_loss": 49.37725, "entropy": 0.850036, "total_loss": 24.686615, "clip_frac": 0.200317, "sps": 427.5, "elapsed_s": 263.5} |
| {"update": 56, "steps_done": 114688, "mean_step_reward": 0.653, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 129.138, "mean_ep_length": 177.6, "mean_collisions": 1.91, "mean_intent_rate": 0.439, "policy_loss": -0.000423, "value_loss": 69.728629, "entropy": 0.874839, "total_loss": 34.855143, "clip_frac": 0.132812, "sps": 438.3, "elapsed_s": 268.2} |
| {"update": 57, "steps_done": 116736, "mean_step_reward": 0.6241, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 117.888, "mean_ep_length": 200.0, "mean_collisions": 3.64, "mean_intent_rate": 0.436, "policy_loss": 0.005019, "value_loss": 59.776852, "entropy": 0.85694, "total_loss": 29.884876, "clip_frac": 0.165283, "sps": 436.2, "elapsed_s": 272.9} |
| {"update": 58, "steps_done": 118784, "mean_step_reward": 0.6185, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 101.843, "mean_ep_length": 176.1, "mean_collisions": 4.64, "mean_intent_rate": 0.451, "policy_loss": 0.001816, "value_loss": 74.202941, "entropy": 0.881636, "total_loss": 37.09447, "clip_frac": 0.075195, "sps": 438.9, "elapsed_s": 277.6} |
| {"update": 59, "steps_done": 120832, "mean_step_reward": 0.7006, "min_step_reward": -10.5364, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 136.522, "mean_ep_length": 185.9, "mean_collisions": 4.18, "mean_intent_rate": 0.498, "policy_loss": 0.019172, "value_loss": 78.725432, "entropy": 0.838843, "total_loss": 39.373499, "clip_frac": 0.282104, "sps": 438.7, "elapsed_s": 282.2} |
| {"update": 60, "steps_done": 122880, "mean_step_reward": 0.4953, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 96.951, "mean_ep_length": 200.0, "mean_collisions": 6.9, "mean_intent_rate": 0.456, "policy_loss": 0.00513, "value_loss": 66.670811, "entropy": 0.830392, "total_loss": 33.332232, "clip_frac": 0.214966, "sps": 437.4, "elapsed_s": 286.9} |
| {"update": 61, "steps_done": 124928, "mean_step_reward": 0.6774, "min_step_reward": -10.5833, "max_step_reward": 2.5, "n_episodes_finished": 12, "mean_ep_return": 123.598, "mean_ep_length": 183.9, "mean_collisions": 2.92, "mean_intent_rate": 0.465, "policy_loss": 0.000362, "value_loss": 68.745282, "entropy": 0.803051, "total_loss": 34.364972, "clip_frac": 0.095947, "sps": 439.1, "elapsed_s": 291.6} |
| {"update": 62, "steps_done": 126976, "mean_step_reward": 0.6008, "min_step_reward": -10.5856, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 110.524, "mean_ep_length": 179.5, "mean_collisions": 4.09, "mean_intent_rate": 0.47, "policy_loss": 0.002859, "value_loss": 63.322886, "entropy": 0.799584, "total_loss": 31.656306, "clip_frac": 0.151489, "sps": 440.3, "elapsed_s": 296.2} |
| {"update": 63, "steps_done": 129024, "mean_step_reward": 0.5202, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 100.646, "mean_ep_length": 200.0, "mean_collisions": 5.5, "mean_intent_rate": 0.439, "policy_loss": 0.006497, "value_loss": 66.993075, "entropy": 0.798078, "total_loss": 33.495053, "clip_frac": 0.13855, "sps": 438.4, "elapsed_s": 300.9} |
| {"update": 64, "steps_done": 131072, "mean_step_reward": 0.6248, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 134.843, "mean_ep_length": 200.0, "mean_collisions": 3.0, "mean_intent_rate": 0.453, "policy_loss": 0.004739, "value_loss": 64.06471, "entropy": 0.746717, "total_loss": 32.029627, "clip_frac": 0.173828, "sps": 429.8, "elapsed_s": 305.7} |
| {"update": 65, "steps_done": 133120, "mean_step_reward": 0.6541, "min_step_reward": -10.5809, "max_step_reward": 2.5, "n_episodes_finished": 12, "mean_ep_return": 107.155, "mean_ep_length": 181.3, "mean_collisions": 3.75, "mean_intent_rate": 0.447, "policy_loss": 4.1e-05, "value_loss": 70.377688, "entropy": 0.769007, "total_loss": 35.181194, "clip_frac": 0.0354, "sps": 440.0, "elapsed_s": 310.3} |
| {"update": 66, "steps_done": 135168, "mean_step_reward": 0.8372, "min_step_reward": -10.575, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 168.457, "mean_ep_length": 200.0, "mean_collisions": 2.7, "mean_intent_rate": 0.504, "policy_loss": 0.004166, "value_loss": 72.497719, "entropy": 0.7488, "total_loss": 36.245537, "clip_frac": 0.105347, "sps": 439.0, "elapsed_s": 315.0} |
| {"update": 67, "steps_done": 137216, "mean_step_reward": 0.7056, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 142.381, "mean_ep_length": 200.0, "mean_collisions": 3.4, "mean_intent_rate": 0.472, "policy_loss": 0.003959, "value_loss": 63.049681, "entropy": 0.732245, "total_loss": 31.521477, "clip_frac": 0.143677, "sps": 439.2, "elapsed_s": 319.7} |
| {"update": 68, "steps_done": 139264, "mean_step_reward": 0.6467, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 131.287, "mean_ep_length": 200.0, "mean_collisions": 5.18, "mean_intent_rate": 0.483, "policy_loss": 0.005415, "value_loss": 88.011554, "entropy": 0.746896, "total_loss": 44.003724, "clip_frac": 0.198486, "sps": 438.0, "elapsed_s": 324.3} |
| {"update": 69, "steps_done": 141312, "mean_step_reward": 0.8649, "min_step_reward": -10.585, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 170.757, "mean_ep_length": 200.0, "mean_collisions": 4.1, "mean_intent_rate": 0.53, "policy_loss": 0.01276, "value_loss": 79.798684, "entropy": 0.717985, "total_loss": 39.904922, "clip_frac": 0.254395, "sps": 437.8, "elapsed_s": 329.0} |
| {"update": 70, "steps_done": 143360, "mean_step_reward": 0.5667, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 125.846, "mean_ep_length": 200.0, "mean_collisions": 6.6, "mean_intent_rate": 0.497, "policy_loss": -0.000746, "value_loss": 119.884901, "entropy": 0.733475, "total_loss": 59.93437, "clip_frac": 0.110352, "sps": 433.9, "elapsed_s": 333.7} |
| {"update": 71, "steps_done": 145408, "mean_step_reward": 0.765, "min_step_reward": -10.585, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 140.145, "mean_ep_length": 194.2, "mean_collisions": 4.18, "mean_intent_rate": 0.494, "policy_loss": 0.004457, "value_loss": 83.074171, "entropy": 0.672234, "total_loss": 41.534821, "clip_frac": 0.169434, "sps": 437.8, "elapsed_s": 338.4} |
| {"update": 72, "steps_done": 147456, "mean_step_reward": 0.5297, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 103.17, "mean_ep_length": 200.0, "mean_collisions": 6.8, "mean_intent_rate": 0.462, "policy_loss": 0.013964, "value_loss": 88.067253, "entropy": 0.704896, "total_loss": 44.040541, "clip_frac": 0.213745, "sps": 437.6, "elapsed_s": 343.1} |
| {"update": 73, "steps_done": 149504, "mean_step_reward": 0.5538, "min_step_reward": -10.5793, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 116.084, "mean_ep_length": 200.0, "mean_collisions": 4.9, "mean_intent_rate": 0.451, "policy_loss": 0.009164, "value_loss": 74.202499, "entropy": 0.666524, "total_loss": 37.103747, "clip_frac": 0.208374, "sps": 425.5, "elapsed_s": 347.9} |
| {"update": 74, "steps_done": 151552, "mean_step_reward": 0.6742, "min_step_reward": -10.5274, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 125.085, "mean_ep_length": 200.0, "mean_collisions": 2.4, "mean_intent_rate": 0.424, "policy_loss": 0.01705, "value_loss": 64.878484, "entropy": 0.684011, "total_loss": 32.449452, "clip_frac": 0.258423, "sps": 435.6, "elapsed_s": 352.6} |
| {"update": 75, "steps_done": 153600, "mean_step_reward": 0.4129, "min_step_reward": -10.5836, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 89.887, "mean_ep_length": 200.0, "mean_collisions": 1.82, "mean_intent_rate": 0.359, "policy_loss": 0.009807, "value_loss": 65.557708, "entropy": 0.471172, "total_loss": 32.78395, "clip_frac": 0.137695, "sps": 436.9, "elapsed_s": 357.3} |
| {"update": 76, "steps_done": 155648, "mean_step_reward": 0.3914, "min_step_reward": -10.5621, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 79.281, "mean_ep_length": 200.0, "mean_collisions": 1.4, "mean_intent_rate": 0.335, "policy_loss": 0.002553, "value_loss": 50.637056, "entropy": 0.569156, "total_loss": 25.31539, "clip_frac": 0.13855, "sps": 440.8, "elapsed_s": 361.9} |
| {"update": 77, "steps_done": 157696, "mean_step_reward": 0.5484, "min_step_reward": -10.5851, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 108.394, "mean_ep_length": 200.0, "mean_collisions": 1.1, "mean_intent_rate": 0.377, "policy_loss": 0.002416, "value_loss": 50.499614, "entropy": 0.577307, "total_loss": 25.24645, "clip_frac": 0.106079, "sps": 441.7, "elapsed_s": 366.6} |
| {"update": 78, "steps_done": 159744, "mean_step_reward": 0.6441, "min_step_reward": -10.5029, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 122.994, "mean_ep_length": 186.5, "mean_collisions": 1.0, "mean_intent_rate": 0.409, "policy_loss": 0.009536, "value_loss": 66.35578, "entropy": 0.661047, "total_loss": 33.180815, "clip_frac": 0.199585, "sps": 440.2, "elapsed_s": 371.2} |
| {"update": 79, "steps_done": 161792, "mean_step_reward": 0.7056, "min_step_reward": -10.5867, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 132.331, "mean_ep_length": 193.6, "mean_collisions": 1.82, "mean_intent_rate": 0.433, "policy_loss": 0.018327, "value_loss": 70.178122, "entropy": 0.568277, "total_loss": 35.101705, "clip_frac": 0.256226, "sps": 437.3, "elapsed_s": 375.9} |
| {"update": 80, "steps_done": 163840, "mean_step_reward": 0.5496, "min_step_reward": -10.5847, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 117.143, "mean_ep_length": 200.0, "mean_collisions": 2.8, "mean_intent_rate": 0.42, "policy_loss": 0.008829, "value_loss": 66.083047, "entropy": 0.700215, "total_loss": 33.04335, "clip_frac": 0.196655, "sps": 440.2, "elapsed_s": 380.6} |
| {"update": 81, "steps_done": 165888, "mean_step_reward": 0.7959, "min_step_reward": -7.5883, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 149.476, "mean_ep_length": 200.0, "mean_collisions": 1.6, "mean_intent_rate": 0.454, "policy_loss": 0.003558, "value_loss": 69.739732, "entropy": 0.667952, "total_loss": 34.866745, "clip_frac": 0.076538, "sps": 436.9, "elapsed_s": 385.3} |
| {"update": 82, "steps_done": 167936, "mean_step_reward": 0.5715, "min_step_reward": -10.5825, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 116.771, "mean_ep_length": 200.0, "mean_collisions": 2.3, "mean_intent_rate": 0.411, "policy_loss": 0.003309, "value_loss": 60.71387, "entropy": 0.711739, "total_loss": 30.353127, "clip_frac": 0.150391, "sps": 429.7, "elapsed_s": 390.0} |
| {"update": 83, "steps_done": 169984, "mean_step_reward": 0.6219, "min_step_reward": -10.5789, "max_step_reward": 2.5, "n_episodes_finished": 12, "mean_ep_return": 105.24, "mean_ep_length": 176.8, "mean_collisions": 3.25, "mean_intent_rate": 0.434, "policy_loss": 0.009543, "value_loss": 97.593401, "entropy": 0.70282, "total_loss": 48.799215, "clip_frac": 0.189941, "sps": 437.9, "elapsed_s": 394.7} |
| {"update": 84, "steps_done": 172032, "mean_step_reward": 0.5745, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 122.173, "mean_ep_length": 200.0, "mean_collisions": 2.6, "mean_intent_rate": 0.426, "policy_loss": 0.022917, "value_loss": 70.583123, "entropy": 0.70751, "total_loss": 35.307403, "clip_frac": 0.289673, "sps": 438.5, "elapsed_s": 399.4} |
| {"update": 85, "steps_done": 174080, "mean_step_reward": 0.4464, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 80.704, "mean_ep_length": 185.2, "mean_collisions": 1.73, "mean_intent_rate": 0.335, "policy_loss": 0.007056, "value_loss": 47.494692, "entropy": 0.654989, "total_loss": 23.747852, "clip_frac": 0.158569, "sps": 441.9, "elapsed_s": 404.0} |
| {"update": 86, "steps_done": 176128, "mean_step_reward": 0.4903, "min_step_reward": -10.5847, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 95.244, "mean_ep_length": 192.0, "mean_collisions": 3.18, "mean_intent_rate": 0.396, "policy_loss": 0.005267, "value_loss": 57.209159, "entropy": 0.685683, "total_loss": 28.602989, "clip_frac": 0.144409, "sps": 439.1, "elapsed_s": 408.7} |
| {"update": 87, "steps_done": 178176, "mean_step_reward": 0.4181, "min_step_reward": -10.5534, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 82.427, "mean_ep_length": 200.0, "mean_collisions": 2.1, "mean_intent_rate": 0.35, "policy_loss": 0.017267, "value_loss": 44.016353, "entropy": 0.611822, "total_loss": 22.019325, "clip_frac": 0.185547, "sps": 438.3, "elapsed_s": 413.3} |
| {"update": 88, "steps_done": 180224, "mean_step_reward": 0.6446, "min_step_reward": -10.5854, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 127.918, "mean_ep_length": 195.0, "mean_collisions": 3.91, "mean_intent_rate": 0.459, "policy_loss": 0.020766, "value_loss": 78.069201, "entropy": 0.677057, "total_loss": 39.048596, "clip_frac": 0.369507, "sps": 433.9, "elapsed_s": 418.1} |
| {"update": 89, "steps_done": 182272, "mean_step_reward": 0.6862, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 132.054, "mean_ep_length": 200.0, "mean_collisions": 5.3, "mean_intent_rate": 0.486, "policy_loss": 0.014747, "value_loss": 79.711939, "entropy": 0.706854, "total_loss": 39.863647, "clip_frac": 0.265991, "sps": 439.8, "elapsed_s": 422.7} |
| {"update": 90, "steps_done": 184320, "mean_step_reward": 0.7509, "min_step_reward": -10.5851, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 153.939, "mean_ep_length": 199.3, "mean_collisions": 1.7, "mean_intent_rate": 0.464, "policy_loss": -0.000409, "value_loss": 69.354106, "entropy": 0.666713, "total_loss": 34.669977, "clip_frac": 0.114502, "sps": 431.0, "elapsed_s": 427.5} |
| {"update": 91, "steps_done": 186368, "mean_step_reward": 0.935, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 178.897, "mean_ep_length": 200.0, "mean_collisions": 2.5, "mean_intent_rate": 0.518, "policy_loss": 0.000991, "value_loss": 72.434509, "entropy": 0.669515, "total_loss": 36.211551, "clip_frac": 0.125244, "sps": 433.8, "elapsed_s": 432.2} |
| {"update": 92, "steps_done": 188416, "mean_step_reward": 0.4077, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 87.029, "mean_ep_length": 190.4, "mean_collisions": 5.91, "mean_intent_rate": 0.433, "policy_loss": 0.005803, "value_loss": 120.827351, "entropy": 0.596187, "total_loss": 60.413517, "clip_frac": 0.158569, "sps": 437.0, "elapsed_s": 436.9} |
| {"update": 93, "steps_done": 190464, "mean_step_reward": 0.5728, "min_step_reward": -10.5635, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 100.591, "mean_ep_length": 187.1, "mean_collisions": 3.0, "mean_intent_rate": 0.403, "policy_loss": 0.006498, "value_loss": 62.927809, "entropy": 0.655697, "total_loss": 31.463845, "clip_frac": 0.177002, "sps": 443.8, "elapsed_s": 441.5} |
| {"update": 94, "steps_done": 192512, "mean_step_reward": 0.3918, "min_step_reward": -10.5848, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 86.199, "mean_ep_length": 200.0, "mean_collisions": 7.0, "mean_intent_rate": 0.439, "policy_loss": 0.014177, "value_loss": 102.115791, "entropy": 0.626656, "total_loss": 51.065806, "clip_frac": 0.190674, "sps": 440.9, "elapsed_s": 446.2} |
| {"update": 95, "steps_done": 194560, "mean_step_reward": 0.7196, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 139.316, "mean_ep_length": 200.0, "mean_collisions": 6.18, "mean_intent_rate": 0.514, "policy_loss": 0.006206, "value_loss": 105.851328, "entropy": 0.707507, "total_loss": 52.924796, "clip_frac": 0.131714, "sps": 436.3, "elapsed_s": 450.8} |
| {"update": 96, "steps_done": 196608, "mean_step_reward": 0.6608, "min_step_reward": -10.5851, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 134.707, "mean_ep_length": 200.0, "mean_collisions": 3.5, "mean_intent_rate": 0.463, "policy_loss": 0.009179, "value_loss": 65.688648, "entropy": 0.730126, "total_loss": 32.846202, "clip_frac": 0.132324, "sps": 437.8, "elapsed_s": 455.5} |
| {"update": 97, "steps_done": 198656, "mean_step_reward": 0.5854, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 110.835, "mean_ep_length": 192.5, "mean_collisions": 3.09, "mean_intent_rate": 0.423, "policy_loss": 0.005005, "value_loss": 75.566135, "entropy": 0.640748, "total_loss": 37.781665, "clip_frac": 0.142822, "sps": 435.6, "elapsed_s": 460.2} |
| {"update": 98, "steps_done": 200704, "mean_step_reward": 0.554, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 105.564, "mean_ep_length": 184.5, "mean_collisions": 3.36, "mean_intent_rate": 0.436, "policy_loss": 0.016108, "value_loss": 91.931685, "entropy": 0.729287, "total_loss": 45.974658, "clip_frac": 0.21936, "sps": 446.4, "elapsed_s": 464.8} |
| {"update": 99, "steps_done": 202752, "mean_step_reward": 0.5119, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 92.383, "mean_ep_length": 185.6, "mean_collisions": 2.73, "mean_intent_rate": 0.389, "policy_loss": 0.007905, "value_loss": 61.3185, "entropy": 0.527465, "total_loss": 30.66188, "clip_frac": 0.129761, "sps": 432.9, "elapsed_s": 469.5} |
| {"update": 100, "steps_done": 204800, "mean_step_reward": 0.8617, "min_step_reward": -10.567, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 170.036, "mean_ep_length": 200.0, "mean_collisions": 2.3, "mean_intent_rate": 0.501, "policy_loss": 0.007186, "value_loss": 65.147746, "entropy": 0.682458, "total_loss": 32.574234, "clip_frac": 0.139526, "sps": 438.2, "elapsed_s": 474.2} |
| {"update": 101, "steps_done": 206848, "mean_step_reward": 0.591, "min_step_reward": -10.5841, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 117.839, "mean_ep_length": 193.0, "mean_collisions": 2.91, "mean_intent_rate": 0.425, "policy_loss": -0.000716, "value_loss": 75.053616, "entropy": 0.602873, "total_loss": 37.520064, "clip_frac": 0.031372, "sps": 437.8, "elapsed_s": 478.9} |
| {"update": 102, "steps_done": 208896, "mean_step_reward": 0.704, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 138.284, "mean_ep_length": 200.0, "mean_collisions": 2.6, "mean_intent_rate": 0.453, "policy_loss": 0.003361, "value_loss": 72.781519, "entropy": 0.542627, "total_loss": 36.388695, "clip_frac": 0.071899, "sps": 446.6, "elapsed_s": 483.5} |
| {"update": 103, "steps_done": 210944, "mean_step_reward": 0.5786, "min_step_reward": -10.5829, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 122.321, "mean_ep_length": 200.0, "mean_collisions": 2.7, "mean_intent_rate": 0.426, "policy_loss": 0.00184, "value_loss": 65.753759, "entropy": 0.65274, "total_loss": 32.872192, "clip_frac": 0.083862, "sps": 442.5, "elapsed_s": 488.1} |
| {"update": 104, "steps_done": 212992, "mean_step_reward": 0.4823, "min_step_reward": -10.5861, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 84.38, "mean_ep_length": 200.0, "mean_collisions": 3.3, "mean_intent_rate": 0.374, "policy_loss": -0.001783, "value_loss": 68.904147, "entropy": 0.603567, "total_loss": 34.444255, "clip_frac": 0.02771, "sps": 443.3, "elapsed_s": 492.7} |
| {"update": 105, "steps_done": 215040, "mean_step_reward": 0.7341, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 151.514, "mean_ep_length": 200.0, "mean_collisions": 2.36, "mean_intent_rate": 0.47, "policy_loss": 0.011612, "value_loss": 67.255589, "entropy": 0.704839, "total_loss": 33.632358, "clip_frac": 0.134033, "sps": 437.0, "elapsed_s": 497.4} |
| {"update": 106, "steps_done": 217088, "mean_step_reward": 0.6141, "min_step_reward": -10.5867, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 124.794, "mean_ep_length": 200.0, "mean_collisions": 3.2, "mean_intent_rate": 0.44, "policy_loss": 0.002653, "value_loss": 71.468164, "entropy": 0.605735, "total_loss": 35.730678, "clip_frac": 0.099487, "sps": 440.1, "elapsed_s": 502.1} |
| {"update": 107, "steps_done": 219136, "mean_step_reward": 0.4657, "min_step_reward": -10.5887, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 86.977, "mean_ep_length": 189.6, "mean_collisions": 7.45, "mean_intent_rate": 0.461, "policy_loss": 0.007723, "value_loss": 95.554623, "entropy": 0.684679, "total_loss": 47.778188, "clip_frac": 0.214355, "sps": 441.7, "elapsed_s": 506.7} |
| {"update": 108, "steps_done": 221184, "mean_step_reward": 0.5881, "min_step_reward": -10.5856, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 124.383, "mean_ep_length": 200.0, "mean_collisions": 4.9, "mean_intent_rate": 0.467, "policy_loss": -0.001948, "value_loss": 81.981599, "entropy": 0.643758, "total_loss": 40.982414, "clip_frac": 0.069824, "sps": 429.6, "elapsed_s": 511.5} |
| {"update": 109, "steps_done": 223232, "mean_step_reward": 0.6389, "min_step_reward": -10.5836, "max_step_reward": 2.5, "n_episodes_finished": 12, "mean_ep_return": 104.909, "mean_ep_length": 173.4, "mean_collisions": 3.33, "mean_intent_rate": 0.434, "policy_loss": 0.001389, "value_loss": 66.304474, "entropy": 0.603763, "total_loss": 33.147589, "clip_frac": 0.121216, "sps": 442.1, "elapsed_s": 516.1} |
| {"update": 110, "steps_done": 225280, "mean_step_reward": 0.8385, "min_step_reward": -10.5886, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 162.768, "mean_ep_length": 192.1, "mean_collisions": 1.3, "mean_intent_rate": 0.472, "policy_loss": 0.005514, "value_loss": 68.76877, "entropy": 0.553872, "total_loss": 34.38436, "clip_frac": 0.109741, "sps": 438.6, "elapsed_s": 520.8} |
| {"update": 111, "steps_done": 227328, "mean_step_reward": 0.6716, "min_step_reward": -10.5861, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 134.709, "mean_ep_length": 200.0, "mean_collisions": 1.91, "mean_intent_rate": 0.436, "policy_loss": 0.000941, "value_loss": 70.632444, "entropy": 0.608306, "total_loss": 35.31108, "clip_frac": 0.099243, "sps": 438.5, "elapsed_s": 525.5} |
| {"update": 112, "steps_done": 229376, "mean_step_reward": 0.7154, "min_step_reward": -10.5734, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 143.048, "mean_ep_length": 187.4, "mean_collisions": 1.4, "mean_intent_rate": 0.454, "policy_loss": 0.012677, "value_loss": 62.352205, "entropy": 0.563175, "total_loss": 31.183148, "clip_frac": 0.192993, "sps": 444.6, "elapsed_s": 530.1} |
| {"update": 113, "steps_done": 231424, "mean_step_reward": 0.7552, "min_step_reward": -10.5852, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 137.901, "mean_ep_length": 194.7, "mean_collisions": 2.09, "mean_intent_rate": 0.446, "policy_loss": 0.009006, "value_loss": 61.929052, "entropy": 0.590815, "total_loss": 30.967624, "clip_frac": 0.159546, "sps": 440.0, "elapsed_s": 534.7} |
| {"update": 114, "steps_done": 233472, "mean_step_reward": 0.5204, "min_step_reward": -10.5867, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 105.393, "mean_ep_length": 200.0, "mean_collisions": 3.2, "mean_intent_rate": 0.407, "policy_loss": 0.008865, "value_loss": 68.518765, "entropy": 0.605428, "total_loss": 34.262194, "clip_frac": 0.196777, "sps": 435.9, "elapsed_s": 539.4} |
| {"update": 115, "steps_done": 235520, "mean_step_reward": 0.7664, "min_step_reward": -10.586, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 146.742, "mean_ep_length": 187.5, "mean_collisions": 0.73, "mean_intent_rate": 0.441, "policy_loss": 0.005789, "value_loss": 62.496433, "entropy": 0.591041, "total_loss": 31.248095, "clip_frac": 0.127686, "sps": 435.9, "elapsed_s": 544.1} |
| {"update": 116, "steps_done": 237568, "mean_step_reward": 0.6704, "min_step_reward": -10.5875, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 131.876, "mean_ep_length": 200.0, "mean_collisions": 2.0, "mean_intent_rate": 0.43, "policy_loss": 0.001425, "value_loss": 66.842081, "entropy": 0.625731, "total_loss": 33.416208, "clip_frac": 0.083618, "sps": 434.5, "elapsed_s": 548.8} |
| {"update": 117, "steps_done": 239616, "mean_step_reward": 0.4811, "min_step_reward": -7.5874, "max_step_reward": 2.5, "n_episodes_finished": 12, "mean_ep_return": 83.74, "mean_ep_length": 177.0, "mean_collisions": 1.83, "mean_intent_rate": 0.365, "policy_loss": 0.006837, "value_loss": 66.576585, "entropy": 0.547485, "total_loss": 33.289654, "clip_frac": 0.141968, "sps": 433.2, "elapsed_s": 553.6} |
| {"update": 118, "steps_done": 241664, "mean_step_reward": 0.6252, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 118.456, "mean_ep_length": 200.0, "mean_collisions": 3.8, "mean_intent_rate": 0.438, "policy_loss": 0.000761, "value_loss": 62.466696, "entropy": 0.587864, "total_loss": 31.22823, "clip_frac": 0.085938, "sps": 439.9, "elapsed_s": 558.2} |
| {"update": 119, "steps_done": 243712, "mean_step_reward": 0.8926, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 189.218, "mean_ep_length": 200.0, "mean_collisions": 2.3, "mean_intent_rate": 0.531, "policy_loss": 0.003821, "value_loss": 79.028006, "entropy": 0.659771, "total_loss": 39.511226, "clip_frac": 0.088867, "sps": 432.9, "elapsed_s": 563.0} |
| {"update": 120, "steps_done": 245760, "mean_step_reward": 0.5612, "min_step_reward": -10.5845, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 109.393, "mean_ep_length": 200.0, "mean_collisions": 3.18, "mean_intent_rate": 0.415, "policy_loss": -0.000954, "value_loss": 69.601302, "entropy": 0.555353, "total_loss": 34.794143, "clip_frac": 0.060791, "sps": 442.3, "elapsed_s": 567.6} |
| {"update": 121, "steps_done": 247808, "mean_step_reward": 0.8497, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 165.153, "mean_ep_length": 191.5, "mean_collisions": 4.4, "mean_intent_rate": 0.545, "policy_loss": 0.006623, "value_loss": 85.870336, "entropy": 0.607535, "total_loss": 42.935715, "clip_frac": 0.098389, "sps": 437.6, "elapsed_s": 572.3} |
| {"update": 122, "steps_done": 249856, "mean_step_reward": 0.7503, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 150.413, "mean_ep_length": 200.0, "mean_collisions": 2.91, "mean_intent_rate": 0.477, "policy_loss": 0.007433, "value_loss": 76.29259, "entropy": 0.636775, "total_loss": 38.14736, "clip_frac": 0.132446, "sps": 439.9, "elapsed_s": 576.9} |
| {"update": 123, "steps_done": 251904, "mean_step_reward": 0.5655, "min_step_reward": -10.584, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 116.736, "mean_ep_length": 189.2, "mean_collisions": 3.9, "mean_intent_rate": 0.454, "policy_loss": 0.011617, "value_loss": 78.317911, "entropy": 0.620899, "total_loss": 39.164364, "clip_frac": 0.209229, "sps": 445.7, "elapsed_s": 581.5} |
| {"update": 124, "steps_done": 253952, "mean_step_reward": 0.6437, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 120.754, "mean_ep_length": 200.0, "mean_collisions": 8.09, "mean_intent_rate": 0.515, "policy_loss": -0.001525, "value_loss": 124.612569, "entropy": 0.614341, "total_loss": 62.298616, "clip_frac": 0.233398, "sps": 437.8, "elapsed_s": 586.2} |
| {"update": 125, "steps_done": 256000, "mean_step_reward": 0.9071, "min_step_reward": -10.5843, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 179.913, "mean_ep_length": 200.0, "mean_collisions": 3.9, "mean_intent_rate": 0.543, "policy_loss": 0.009737, "value_loss": 81.531982, "entropy": 0.590646, "total_loss": 40.769821, "clip_frac": 0.214722, "sps": 438.8, "elapsed_s": 590.9} |
| {"update": 126, "steps_done": 258048, "mean_step_reward": 0.6867, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 133.704, "mean_ep_length": 200.0, "mean_collisions": 8.9, "mean_intent_rate": 0.549, "policy_loss": 0.009931, "value_loss": 98.414801, "entropy": 0.619462, "total_loss": 49.211137, "clip_frac": 0.210083, "sps": 428.7, "elapsed_s": 595.7} |
| {"update": 127, "steps_done": 260096, "mean_step_reward": 0.7962, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 162.804, "mean_ep_length": 200.0, "mean_collisions": 5.1, "mean_intent_rate": 0.533, "policy_loss": 0.012632, "value_loss": 84.448723, "entropy": 0.638645, "total_loss": 42.230607, "clip_frac": 0.190552, "sps": 434.8, "elapsed_s": 600.4} |
| {"update": 128, "steps_done": 262144, "mean_step_reward": 0.5785, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 109.055, "mean_ep_length": 184.6, "mean_collisions": 6.55, "mean_intent_rate": 0.477, "policy_loss": 0.017774, "value_loss": 90.191936, "entropy": 0.560427, "total_loss": 45.108137, "clip_frac": 0.283325, "sps": 439.0, "elapsed_s": 605.0} |
| {"update": 129, "steps_done": 264192, "mean_step_reward": 0.8724, "min_step_reward": -10.5851, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 170.808, "mean_ep_length": 200.0, "mean_collisions": 3.09, "mean_intent_rate": 0.513, "policy_loss": 0.006675, "value_loss": 88.660828, "entropy": 0.586686, "total_loss": 44.331222, "clip_frac": 0.195312, "sps": 439.1, "elapsed_s": 609.7} |
| {"update": 130, "steps_done": 266240, "mean_step_reward": 0.85, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 169.398, "mean_ep_length": 200.0, "mean_collisions": 6.0, "mean_intent_rate": 0.56, "policy_loss": 0.015171, "value_loss": 100.561534, "entropy": 0.645339, "total_loss": 50.289484, "clip_frac": 0.270508, "sps": 435.2, "elapsed_s": 614.4} |
| {"update": 131, "steps_done": 268288, "mean_step_reward": 0.7458, "min_step_reward": -10.5862, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 151.534, "mean_ep_length": 200.0, "mean_collisions": 7.6, "mean_intent_rate": 0.556, "policy_loss": 0.064595, "value_loss": 100.697244, "entropy": 0.598477, "total_loss": 50.407232, "clip_frac": 0.480835, "sps": 439.6, "elapsed_s": 619.1} |
| {"update": 132, "steps_done": 270336, "mean_step_reward": 0.5874, "min_step_reward": -10.587, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 119.522, "mean_ep_length": 200.0, "mean_collisions": 7.3, "mean_intent_rate": 0.499, "policy_loss": 0.003906, "value_loss": 98.892376, "entropy": 0.557342, "total_loss": 49.44452, "clip_frac": 0.102295, "sps": 447.2, "elapsed_s": 623.6} |
| {"update": 133, "steps_done": 272384, "mean_step_reward": 0.4994, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 99.091, "mean_ep_length": 200.0, "mean_collisions": 11.73, "mean_intent_rate": 0.54, "policy_loss": 0.014227, "value_loss": 109.603099, "entropy": 0.6513, "total_loss": 54.809264, "clip_frac": 0.274658, "sps": 443.8, "elapsed_s": 628.3} |
| {"update": 134, "steps_done": 274432, "mean_step_reward": 0.6679, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 131.638, "mean_ep_length": 200.0, "mean_collisions": 4.5, "mean_intent_rate": 0.472, "policy_loss": 0.000638, "value_loss": 68.066889, "entropy": 0.636123, "total_loss": 34.027721, "clip_frac": 0.034058, "sps": 439.7, "elapsed_s": 632.9} |
| {"update": 135, "steps_done": 276480, "mean_step_reward": 0.6015, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 118.951, "mean_ep_length": 200.0, "mean_collisions": 8.6, "mean_intent_rate": 0.52, "policy_loss": -0.000968, "value_loss": 150.198789, "entropy": 0.618305, "total_loss": 75.092244, "clip_frac": 0.119141, "sps": 433.8, "elapsed_s": 637.6} |
| {"update": 136, "steps_done": 278528, "mean_step_reward": 0.8329, "min_step_reward": -10.5778, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 167.83, "mean_ep_length": 200.0, "mean_collisions": 2.2, "mean_intent_rate": 0.493, "policy_loss": 0.042492, "value_loss": 63.286295, "entropy": 0.618334, "total_loss": 31.679456, "clip_frac": 0.432861, "sps": 442.4, "elapsed_s": 642.3} |
| {"update": 137, "steps_done": 280576, "mean_step_reward": 0.734, "min_step_reward": -10.5851, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 143.856, "mean_ep_length": 196.3, "mean_collisions": 4.36, "mean_intent_rate": 0.495, "policy_loss": 0.000949, "value_loss": 85.982451, "entropy": 0.583811, "total_loss": 42.986336, "clip_frac": 0.12793, "sps": 439.3, "elapsed_s": 646.9} |
| {"update": 138, "steps_done": 282624, "mean_step_reward": 0.8548, "min_step_reward": -10.5854, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 168.954, "mean_ep_length": 200.0, "mean_collisions": 2.3, "mean_intent_rate": 0.496, "policy_loss": 0.009598, "value_loss": 76.313187, "entropy": 0.651125, "total_loss": 38.159681, "clip_frac": 0.214355, "sps": 437.2, "elapsed_s": 651.6} |
| {"update": 139, "steps_done": 284672, "mean_step_reward": 0.8048, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 171.82, "mean_ep_length": 200.0, "mean_collisions": 4.6, "mean_intent_rate": 0.541, "policy_loss": 0.004666, "value_loss": 78.134189, "entropy": 0.625249, "total_loss": 39.065508, "clip_frac": 0.127319, "sps": 439.7, "elapsed_s": 656.3} |
| {"update": 140, "steps_done": 286720, "mean_step_reward": 0.6453, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 124.8, "mean_ep_length": 200.0, "mean_collisions": 6.0, "mean_intent_rate": 0.484, "policy_loss": 0.0001, "value_loss": 85.26931, "entropy": 0.656196, "total_loss": 42.628194, "clip_frac": 0.055786, "sps": 437.4, "elapsed_s": 661.0} |
| {"update": 141, "steps_done": 288768, "mean_step_reward": 0.8621, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 158.598, "mean_ep_length": 193.5, "mean_collisions": 4.36, "mean_intent_rate": 0.53, "policy_loss": 0.005624, "value_loss": 89.537315, "entropy": 0.595611, "total_loss": 44.768325, "clip_frac": 0.140381, "sps": 437.0, "elapsed_s": 665.6} |
| {"update": 142, "steps_done": 290816, "mean_step_reward": 0.5764, "min_step_reward": -10.5865, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 122.397, "mean_ep_length": 200.0, "mean_collisions": 4.3, "mean_intent_rate": 0.456, "policy_loss": 0.013902, "value_loss": 86.145637, "entropy": 0.531019, "total_loss": 43.081411, "clip_frac": 0.227539, "sps": 442.5, "elapsed_s": 670.3} |
| {"update": 143, "steps_done": 292864, "mean_step_reward": 0.8437, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 166.581, "mean_ep_length": 200.0, "mean_collisions": 2.45, "mean_intent_rate": 0.495, "policy_loss": 0.004528, "value_loss": 84.585704, "entropy": 0.625788, "total_loss": 42.291122, "clip_frac": 0.091919, "sps": 433.5, "elapsed_s": 675.0} |
| {"update": 144, "steps_done": 294912, "mean_step_reward": 0.7992, "min_step_reward": -10.5632, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 161.441, "mean_ep_length": 200.0, "mean_collisions": 4.4, "mean_intent_rate": 0.521, "policy_loss": -4.5e-05, "value_loss": 85.543074, "entropy": 0.56022, "total_loss": 42.76589, "clip_frac": 0.110962, "sps": 439.4, "elapsed_s": 679.7} |
| {"update": 145, "steps_done": 296960, "mean_step_reward": 0.7777, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 156.012, "mean_ep_length": 200.0, "mean_collisions": 5.6, "mean_intent_rate": 0.532, "policy_loss": 0.004967, "value_loss": 103.169205, "entropy": 0.623077, "total_loss": 51.583338, "clip_frac": 0.091064, "sps": 434.5, "elapsed_s": 684.4} |
| {"update": 146, "steps_done": 299008, "mean_step_reward": 0.6475, "min_step_reward": -10.5877, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 136.926, "mean_ep_length": 200.0, "mean_collisions": 6.0, "mean_intent_rate": 0.507, "policy_loss": 0.005819, "value_loss": 89.298962, "entropy": 0.597847, "total_loss": 44.649322, "clip_frac": 0.176025, "sps": 437.8, "elapsed_s": 689.1} |
| {"update": 147, "steps_done": 301056, "mean_step_reward": 0.7753, "min_step_reward": -10.5878, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 145.279, "mean_ep_length": 200.0, "mean_collisions": 5.27, "mean_intent_rate": 0.508, "policy_loss": 0.003426, "value_loss": 83.002499, "entropy": 0.607303, "total_loss": 41.498602, "clip_frac": 0.104248, "sps": 441.9, "elapsed_s": 693.7} |
| {"update": 148, "steps_done": 303104, "mean_step_reward": 0.7183, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 148.147, "mean_ep_length": 200.0, "mean_collisions": 4.1, "mean_intent_rate": 0.493, "policy_loss": 0.002706, "value_loss": 87.60759, "entropy": 0.580164, "total_loss": 43.800699, "clip_frac": 0.097778, "sps": 441.9, "elapsed_s": 698.3} |
| {"update": 149, "steps_done": 305152, "mean_step_reward": 0.6228, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 125.135, "mean_ep_length": 200.0, "mean_collisions": 9.5, "mean_intent_rate": 0.546, "policy_loss": 0.009486, "value_loss": 119.395531, "entropy": 0.617782, "total_loss": 59.701074, "clip_frac": 0.176758, "sps": 440.9, "elapsed_s": 703.0} |
| {"update": 150, "steps_done": 307200, "mean_step_reward": 0.7526, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 136.401, "mean_ep_length": 191.8, "mean_collisions": 8.09, "mean_intent_rate": 0.558, "policy_loss": -0.001194, "value_loss": 82.933614, "entropy": 0.637053, "total_loss": 41.459242, "clip_frac": 0.058716, "sps": 439.7, "elapsed_s": 707.6} |
| {"update": 151, "steps_done": 309248, "mean_step_reward": 0.9471, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 197.476, "mean_ep_length": 200.0, "mean_collisions": 4.6, "mean_intent_rate": 0.582, "policy_loss": 0.011834, "value_loss": 88.079314, "entropy": 0.581795, "total_loss": 44.045673, "clip_frac": 0.203979, "sps": 435.8, "elapsed_s": 712.3} |
| {"update": 152, "steps_done": 311296, "mean_step_reward": 0.7352, "min_step_reward": -10.5871, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 141.725, "mean_ep_length": 200.0, "mean_collisions": 6.1, "mean_intent_rate": 0.516, "policy_loss": 0.010214, "value_loss": 86.519546, "entropy": 0.566384, "total_loss": 43.264323, "clip_frac": 0.188965, "sps": 428.8, "elapsed_s": 717.1} |
| {"update": 153, "steps_done": 313344, "mean_step_reward": 0.9079, "min_step_reward": -10.5498, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 171.731, "mean_ep_length": 184.5, "mean_collisions": 2.27, "mean_intent_rate": 0.522, "policy_loss": 0.007231, "value_loss": 88.11861, "entropy": 0.574951, "total_loss": 44.060786, "clip_frac": 0.137573, "sps": 435.9, "elapsed_s": 721.8} |
| {"update": 154, "steps_done": 315392, "mean_step_reward": 0.5575, "min_step_reward": -10.5845, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 109.213, "mean_ep_length": 200.0, "mean_collisions": 4.2, "mean_intent_rate": 0.431, "policy_loss": 0.012172, "value_loss": 86.017163, "entropy": 0.530686, "total_loss": 43.015446, "clip_frac": 0.163574, "sps": 441.3, "elapsed_s": 726.5} |
| {"update": 155, "steps_done": 317440, "mean_step_reward": 0.774, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 153.854, "mean_ep_length": 200.0, "mean_collisions": 2.64, "mean_intent_rate": 0.479, "policy_loss": 0.001078, "value_loss": 89.28776, "entropy": 0.530431, "total_loss": 44.639653, "clip_frac": 0.083984, "sps": 438.4, "elapsed_s": 731.1} |
| {"update": 156, "steps_done": 319488, "mean_step_reward": 0.7857, "min_step_reward": -10.5844, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 162.643, "mean_ep_length": 200.0, "mean_collisions": 3.4, "mean_intent_rate": 0.506, "policy_loss": 0.003568, "value_loss": 88.967791, "entropy": 0.580054, "total_loss": 44.481664, "clip_frac": 0.115723, "sps": 440.1, "elapsed_s": 735.8} |
| {"update": 157, "steps_done": 321536, "mean_step_reward": 0.7873, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 147.375, "mean_ep_length": 200.0, "mean_collisions": 3.9, "mean_intent_rate": 0.487, "policy_loss": 0.003321, "value_loss": 75.56827, "entropy": 0.573442, "total_loss": 37.781721, "clip_frac": 0.13623, "sps": 437.5, "elapsed_s": 740.5} |
| {"update": 158, "steps_done": 323584, "mean_step_reward": 0.8257, "min_step_reward": -10.5616, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 161.298, "mean_ep_length": 200.0, "mean_collisions": 4.0, "mean_intent_rate": 0.515, "policy_loss": 0.005043, "value_loss": 97.484934, "entropy": 0.544488, "total_loss": 48.742065, "clip_frac": 0.104248, "sps": 430.7, "elapsed_s": 745.2} |
| {"update": 159, "steps_done": 325632, "mean_step_reward": 0.5282, "min_step_reward": -10.5594, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 119.739, "mean_ep_length": 200.0, "mean_collisions": 1.91, "mean_intent_rate": 0.411, "policy_loss": 0.005036, "value_loss": 74.579267, "entropy": 0.44048, "total_loss": 37.290265, "clip_frac": 0.128418, "sps": 444.5, "elapsed_s": 749.8} |
| {"update": 160, "steps_done": 327680, "mean_step_reward": 0.5756, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 101.182, "mean_ep_length": 192.2, "mean_collisions": 2.4, "mean_intent_rate": 0.396, "policy_loss": 0.007226, "value_loss": 67.360596, "entropy": 0.460247, "total_loss": 33.682921, "clip_frac": 0.123901, "sps": 440.4, "elapsed_s": 754.5} |
| {"update": 161, "steps_done": 329728, "mean_step_reward": 0.5609, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 117.674, "mean_ep_length": 200.0, "mean_collisions": 3.73, "mean_intent_rate": 0.436, "policy_loss": 0.000831, "value_loss": 89.85004, "entropy": 0.525929, "total_loss": 44.920591, "clip_frac": 0.075195, "sps": 431.3, "elapsed_s": 759.2} |
| {"update": 162, "steps_done": 331776, "mean_step_reward": 0.9683, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 192.982, "mean_ep_length": 200.0, "mean_collisions": 3.2, "mean_intent_rate": 0.552, "policy_loss": 0.002808, "value_loss": 88.721722, "entropy": 0.526334, "total_loss": 44.358406, "clip_frac": 0.057007, "sps": 432.5, "elapsed_s": 764.0} |
| {"update": 163, "steps_done": 333824, "mean_step_reward": 0.6568, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 125.199, "mean_ep_length": 200.0, "mean_collisions": 5.4, "mean_intent_rate": 0.477, "policy_loss": 0.012036, "value_loss": 95.857734, "entropy": 0.455776, "total_loss": 47.936345, "clip_frac": 0.185669, "sps": 441.7, "elapsed_s": 768.6} |
| {"update": 164, "steps_done": 335872, "mean_step_reward": 0.6166, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 129.186, "mean_ep_length": 200.0, "mean_collisions": 2.9, "mean_intent_rate": 0.443, "policy_loss": 0.003757, "value_loss": 77.830862, "entropy": 0.451968, "total_loss": 38.914668, "clip_frac": 0.088745, "sps": 440.7, "elapsed_s": 773.3} |
| {"update": 165, "steps_done": 337920, "mean_step_reward": 0.8046, "min_step_reward": -10.5483, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 161.164, "mean_ep_length": 200.0, "mean_collisions": 3.27, "mean_intent_rate": 0.5, "policy_loss": 0.008041, "value_loss": 82.954591, "entropy": 0.536806, "total_loss": 41.479969, "clip_frac": 0.148315, "sps": 436.8, "elapsed_s": 777.9} |
| {"update": 166, "steps_done": 339968, "mean_step_reward": 0.4995, "min_step_reward": -10.5869, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 102.119, "mean_ep_length": 199.3, "mean_collisions": 2.0, "mean_intent_rate": 0.382, "policy_loss": 0.000546, "value_loss": 55.606316, "entropy": 0.445274, "total_loss": 27.799252, "clip_frac": 0.046265, "sps": 439.0, "elapsed_s": 782.6} |
| {"update": 167, "steps_done": 342016, "mean_step_reward": 0.8969, "min_step_reward": -10.5474, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 169.062, "mean_ep_length": 200.0, "mean_collisions": 3.4, "mean_intent_rate": 0.517, "policy_loss": 0.005668, "value_loss": 81.864178, "entropy": 0.470425, "total_loss": 40.933052, "clip_frac": 0.129639, "sps": 439.7, "elapsed_s": 787.3} |
| {"update": 168, "steps_done": 344064, "mean_step_reward": 0.8642, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 172.039, "mean_ep_length": 200.0, "mean_collisions": 2.1, "mean_intent_rate": 0.499, "policy_loss": 0.0151, "value_loss": 82.726406, "entropy": 0.528983, "total_loss": 41.373013, "clip_frac": 0.231567, "sps": 434.2, "elapsed_s": 792.0} |
| {"update": 169, "steps_done": 346112, "mean_step_reward": 0.7394, "min_step_reward": -10.587, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 155.323, "mean_ep_length": 200.0, "mean_collisions": 2.09, "mean_intent_rate": 0.473, "policy_loss": -0.000662, "value_loss": 88.975626, "entropy": 0.467993, "total_loss": 44.482472, "clip_frac": 0.078003, "sps": 439.7, "elapsed_s": 796.6} |
| {"update": 170, "steps_done": 348160, "mean_step_reward": 0.7235, "min_step_reward": -10.5859, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 142.581, "mean_ep_length": 200.0, "mean_collisions": 4.3, "mean_intent_rate": 0.488, "policy_loss": 0.003425, "value_loss": 94.540597, "entropy": 0.489192, "total_loss": 47.268831, "clip_frac": 0.069336, "sps": 430.0, "elapsed_s": 801.4} |
| {"update": 171, "steps_done": 350208, "mean_step_reward": 0.7992, "min_step_reward": -10.5377, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 159.852, "mean_ep_length": 200.0, "mean_collisions": 2.2, "mean_intent_rate": 0.48, "policy_loss": -0.000954, "value_loss": 71.187014, "entropy": 0.498008, "total_loss": 35.587573, "clip_frac": 0.109985, "sps": 437.7, "elapsed_s": 806.1} |
| {"update": 172, "steps_done": 352256, "mean_step_reward": 0.7817, "min_step_reward": -10.5465, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 166.924, "mean_ep_length": 200.0, "mean_collisions": 2.9, "mean_intent_rate": 0.504, "policy_loss": 0.004524, "value_loss": 89.738858, "entropy": 0.560915, "total_loss": 44.868344, "clip_frac": 0.085327, "sps": 440.6, "elapsed_s": 810.7} |
| {"update": 173, "steps_done": 354304, "mean_step_reward": 0.6394, "min_step_reward": -10.5419, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 121.191, "mean_ep_length": 200.0, "mean_collisions": 4.0, "mean_intent_rate": 0.446, "policy_loss": 0.013276, "value_loss": 86.923171, "entropy": 0.486013, "total_loss": 43.470001, "clip_frac": 0.247437, "sps": 440.2, "elapsed_s": 815.4} |
| {"update": 174, "steps_done": 356352, "mean_step_reward": 0.739, "min_step_reward": -10.5839, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 149.181, "mean_ep_length": 200.0, "mean_collisions": 3.36, "mean_intent_rate": 0.483, "policy_loss": 0.002926, "value_loss": 88.508465, "entropy": 0.471878, "total_loss": 44.25244, "clip_frac": 0.105591, "sps": 442.9, "elapsed_s": 820.0} |
| {"update": 175, "steps_done": 358400, "mean_step_reward": 0.6592, "min_step_reward": -10.5839, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 120.399, "mean_ep_length": 200.0, "mean_collisions": 4.7, "mean_intent_rate": 0.459, "policy_loss": -0.001856, "value_loss": 86.309261, "entropy": 0.453165, "total_loss": 43.148243, "clip_frac": 0.073608, "sps": 442.6, "elapsed_s": 824.6} |
| {"update": 176, "steps_done": 360448, "mean_step_reward": 0.6745, "min_step_reward": -10.588, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 141.316, "mean_ep_length": 200.0, "mean_collisions": 2.6, "mean_intent_rate": 0.458, "policy_loss": 0.005138, "value_loss": 84.775427, "entropy": 0.432526, "total_loss": 42.388525, "clip_frac": 0.091309, "sps": 440.8, "elapsed_s": 829.3} |
| {"update": 177, "steps_done": 362496, "mean_step_reward": 0.6792, "min_step_reward": -10.5821, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 139.393, "mean_ep_length": 200.0, "mean_collisions": 3.0, "mean_intent_rate": 0.459, "policy_loss": 0.006036, "value_loss": 79.339543, "entropy": 0.470106, "total_loss": 39.671107, "clip_frac": 0.112305, "sps": 436.1, "elapsed_s": 834.0} |
| {"update": 178, "steps_done": 364544, "mean_step_reward": 0.8364, "min_step_reward": -10.5854, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 165.669, "mean_ep_length": 200.0, "mean_collisions": 1.55, "mean_intent_rate": 0.48, "policy_loss": 0.008668, "value_loss": 97.258203, "entropy": 0.494502, "total_loss": 48.632825, "clip_frac": 0.16333, "sps": 438.3, "elapsed_s": 838.7} |
| {"update": 179, "steps_done": 366592, "mean_step_reward": 0.6556, "min_step_reward": -10.5871, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 122.449, "mean_ep_length": 200.0, "mean_collisions": 2.6, "mean_intent_rate": 0.423, "policy_loss": 0.006814, "value_loss": 73.341171, "entropy": 0.435582, "total_loss": 36.673044, "clip_frac": 0.148071, "sps": 427.5, "elapsed_s": 843.5} |
| {"update": 180, "steps_done": 368640, "mean_step_reward": 0.73, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 157.701, "mean_ep_length": 200.0, "mean_collisions": 1.6, "mean_intent_rate": 0.467, "policy_loss": 0.016318, "value_loss": 81.404147, "entropy": 0.530742, "total_loss": 40.713084, "clip_frac": 0.239258, "sps": 436.5, "elapsed_s": 848.1} |
| {"update": 181, "steps_done": 370688, "mean_step_reward": 0.6309, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 120.247, "mean_ep_length": 200.0, "mean_collisions": 1.4, "mean_intent_rate": 0.401, "policy_loss": 0.00442, "value_loss": 68.561655, "entropy": 0.434768, "total_loss": 34.280899, "clip_frac": 0.076782, "sps": 439.3, "elapsed_s": 852.8} |
| {"update": 182, "steps_done": 372736, "mean_step_reward": 0.734, "min_step_reward": -10.5112, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 145.8, "mean_ep_length": 200.0, "mean_collisions": 2.0, "mean_intent_rate": 0.454, "policy_loss": 0.000385, "value_loss": 76.071585, "entropy": 0.456259, "total_loss": 38.031615, "clip_frac": 0.048828, "sps": 434.0, "elapsed_s": 857.5} |
| {"update": 183, "steps_done": 374784, "mean_step_reward": 0.4445, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 90.787, "mean_ep_length": 200.0, "mean_collisions": 1.1, "mean_intent_rate": 0.348, "policy_loss": 0.01227, "value_loss": 59.99216, "entropy": 0.363357, "total_loss": 30.004716, "clip_frac": 0.101685, "sps": 439.9, "elapsed_s": 862.2} |
| {"update": 184, "steps_done": 376832, "mean_step_reward": 0.5817, "min_step_reward": -10.5832, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 116.679, "mean_ep_length": 200.0, "mean_collisions": 2.0, "mean_intent_rate": 0.404, "policy_loss": 0.003587, "value_loss": 65.683745, "entropy": 0.485092, "total_loss": 32.840608, "clip_frac": 0.079834, "sps": 443.8, "elapsed_s": 866.8} |
| {"update": 185, "steps_done": 378880, "mean_step_reward": 0.5991, "min_step_reward": -10.584, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 124.458, "mean_ep_length": 200.0, "mean_collisions": 2.8, "mean_intent_rate": 0.432, "policy_loss": 0.00872, "value_loss": 63.897652, "entropy": 0.439237, "total_loss": 31.953153, "clip_frac": 0.144531, "sps": 442.7, "elapsed_s": 871.4} |
| {"update": 186, "steps_done": 380928, "mean_step_reward": 0.6741, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 126.321, "mean_ep_length": 197.5, "mean_collisions": 2.45, "mean_intent_rate": 0.43, "policy_loss": -0.001552, "value_loss": 75.250908, "entropy": 0.497433, "total_loss": 37.618928, "clip_frac": 0.08374, "sps": 440.2, "elapsed_s": 876.1} |
| {"update": 187, "steps_done": 382976, "mean_step_reward": 0.8122, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 166.4, "mean_ep_length": 200.0, "mean_collisions": 4.3, "mean_intent_rate": 0.527, "policy_loss": 0.003785, "value_loss": 85.42898, "entropy": 0.494941, "total_loss": 42.713326, "clip_frac": 0.092651, "sps": 440.9, "elapsed_s": 880.7} |
| {"update": 188, "steps_done": 385024, "mean_step_reward": 0.7574, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 146.665, "mean_ep_length": 200.0, "mean_collisions": 3.9, "mean_intent_rate": 0.486, "policy_loss": 0.004756, "value_loss": 71.417986, "entropy": 0.537152, "total_loss": 35.708377, "clip_frac": 0.085327, "sps": 433.4, "elapsed_s": 885.5} |
| {"update": 189, "steps_done": 387072, "mean_step_reward": 0.5814, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 116.416, "mean_ep_length": 197.2, "mean_collisions": 9.64, "mean_intent_rate": 0.536, "policy_loss": 0.016974, "value_loss": 103.716687, "entropy": 0.569681, "total_loss": 51.869621, "clip_frac": 0.285156, "sps": 439.2, "elapsed_s": 890.1} |
| {"update": 190, "steps_done": 389120, "mean_step_reward": 0.8488, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 166.534, "mean_ep_length": 200.0, "mean_collisions": 7.9, "mean_intent_rate": 0.588, "policy_loss": 0.009322, "value_loss": 131.753117, "entropy": 0.547793, "total_loss": 65.880402, "clip_frac": 0.189941, "sps": 435.6, "elapsed_s": 894.8} |
| {"update": 191, "steps_done": 391168, "mean_step_reward": 0.8529, "min_step_reward": -10.5861, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 181.362, "mean_ep_length": 200.0, "mean_collisions": 3.7, "mean_intent_rate": 0.542, "policy_loss": 0.009114, "value_loss": 96.090922, "entropy": 0.552465, "total_loss": 48.04905, "clip_frac": 0.177002, "sps": 437.6, "elapsed_s": 899.5} |
| {"update": 192, "steps_done": 393216, "mean_step_reward": 0.8818, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 163.964, "mean_ep_length": 200.0, "mean_collisions": 5.9, "mean_intent_rate": 0.548, "policy_loss": 0.001577, "value_loss": 86.686103, "entropy": 0.567162, "total_loss": 43.338957, "clip_frac": 0.1073, "sps": 439.2, "elapsed_s": 904.2} |
| {"update": 193, "steps_done": 395264, "mean_step_reward": 0.8066, "min_step_reward": -10.5358, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 166.838, "mean_ep_length": 200.0, "mean_collisions": 5.27, "mean_intent_rate": 0.542, "policy_loss": 0.006319, "value_loss": 90.980781, "entropy": 0.573576, "total_loss": 45.490973, "clip_frac": 0.193115, "sps": 437.0, "elapsed_s": 908.8} |
| {"update": 194, "steps_done": 397312, "mean_step_reward": 0.8102, "min_step_reward": -10.5871, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 164.998, "mean_ep_length": 200.0, "mean_collisions": 3.2, "mean_intent_rate": 0.506, "policy_loss": 0.00812, "value_loss": 79.892043, "entropy": 0.497021, "total_loss": 39.949172, "clip_frac": 0.160034, "sps": 435.4, "elapsed_s": 913.6} |
| {"update": 195, "steps_done": 399360, "mean_step_reward": 0.6063, "min_step_reward": -7.5825, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 126.697, "mean_ep_length": 200.0, "mean_collisions": 1.3, "mean_intent_rate": 0.41, "policy_loss": 0.013749, "value_loss": 75.868267, "entropy": 0.459382, "total_loss": 37.943289, "clip_frac": 0.17395, "sps": 436.1, "elapsed_s": 918.3} |
| {"update": 196, "steps_done": 401408, "mean_step_reward": 0.7616, "min_step_reward": -10.5846, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 151.092, "mean_ep_length": 200.0, "mean_collisions": 1.3, "mean_intent_rate": 0.452, "policy_loss": 0.002523, "value_loss": 76.684189, "entropy": 0.425207, "total_loss": 38.340366, "clip_frac": 0.12793, "sps": 437.0, "elapsed_s": 922.9} |
| {"update": 197, "steps_done": 403456, "mean_step_reward": 0.8102, "min_step_reward": -7.5857, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 153.97, "mean_ep_length": 200.0, "mean_collisions": 1.0, "mean_intent_rate": 0.451, "policy_loss": 0.00215, "value_loss": 84.66958, "entropy": 0.50549, "total_loss": 42.331885, "clip_frac": 0.020752, "sps": 426.3, "elapsed_s": 927.7} |
| {"update": 198, "steps_done": 405504, "mean_step_reward": 0.6748, "min_step_reward": -10.5364, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 130.628, "mean_ep_length": 198.9, "mean_collisions": 1.8, "mean_intent_rate": 0.427, "policy_loss": 0.001271, "value_loss": 68.790699, "entropy": 0.480343, "total_loss": 34.391816, "clip_frac": 0.081909, "sps": 435.6, "elapsed_s": 932.4} |
| {"update": 199, "steps_done": 407552, "mean_step_reward": 0.7064, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 146.441, "mean_ep_length": 200.0, "mean_collisions": 1.1, "mean_intent_rate": 0.439, "policy_loss": 0.010618, "value_loss": 73.741461, "entropy": 0.489876, "total_loss": 36.87645, "clip_frac": 0.166382, "sps": 434.0, "elapsed_s": 937.2} |
| {"update": 200, "steps_done": 409600, "mean_step_reward": 0.7866, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 164.789, "mean_ep_length": 200.0, "mean_collisions": 2.7, "mean_intent_rate": 0.496, "policy_loss": 0.012709, "value_loss": 82.053394, "entropy": 0.530321, "total_loss": 41.034103, "clip_frac": 0.177246, "sps": 437.1, "elapsed_s": 941.9} |
| {"update": 201, "steps_done": 411648, "mean_step_reward": 0.7317, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 138.867, "mean_ep_length": 200.0, "mean_collisions": 3.27, "mean_intent_rate": 0.464, "policy_loss": 0.008226, "value_loss": 95.503623, "entropy": 0.582516, "total_loss": 47.754213, "clip_frac": 0.196289, "sps": 436.6, "elapsed_s": 946.6} |
| {"update": 202, "steps_done": 413696, "mean_step_reward": 0.5619, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 105.602, "mean_ep_length": 192.4, "mean_collisions": 6.9, "mean_intent_rate": 0.475, "policy_loss": 0.001207, "value_loss": 114.39791, "entropy": 0.601757, "total_loss": 57.194145, "clip_frac": 0.206665, "sps": 438.8, "elapsed_s": 951.2} |
| {"update": 203, "steps_done": 415744, "mean_step_reward": 0.7204, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 136.207, "mean_ep_length": 193.4, "mean_collisions": 6.18, "mean_intent_rate": 0.522, "policy_loss": 0.019225, "value_loss": 93.594405, "entropy": 0.616691, "total_loss": 46.810261, "clip_frac": 0.190918, "sps": 437.3, "elapsed_s": 955.9} |
| {"update": 204, "steps_done": 417792, "mean_step_reward": 0.8702, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 179.316, "mean_ep_length": 200.0, "mean_collisions": 4.3, "mean_intent_rate": 0.548, "policy_loss": 0.015278, "value_loss": 77.987459, "entropy": 0.592901, "total_loss": 39.003079, "clip_frac": 0.231812, "sps": 436.3, "elapsed_s": 960.6} |
| {"update": 205, "steps_done": 419840, "mean_step_reward": 0.6014, "min_step_reward": -10.5866, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 130.357, "mean_ep_length": 200.0, "mean_collisions": 7.0, "mean_intent_rate": 0.511, "policy_loss": 0.008974, "value_loss": 103.332752, "entropy": 0.608188, "total_loss": 51.669269, "clip_frac": 0.172363, "sps": 431.8, "elapsed_s": 965.3} |
| {"update": 206, "steps_done": 421888, "mean_step_reward": 0.8242, "min_step_reward": -10.5855, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 151.616, "mean_ep_length": 200.0, "mean_collisions": 4.9, "mean_intent_rate": 0.512, "policy_loss": 0.003555, "value_loss": 77.077864, "entropy": 0.570208, "total_loss": 38.536785, "clip_frac": 0.08606, "sps": 440.2, "elapsed_s": 970.0} |
| {"update": 207, "steps_done": 423936, "mean_step_reward": 0.7763, "min_step_reward": -10.5578, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 155.041, "mean_ep_length": 200.0, "mean_collisions": 4.09, "mean_intent_rate": 0.504, "policy_loss": 0.000241, "value_loss": 80.382202, "entropy": 0.559039, "total_loss": 40.185752, "clip_frac": 0.094849, "sps": 436.8, "elapsed_s": 974.7} |
| {"update": 208, "steps_done": 425984, "mean_step_reward": 0.7816, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 153.752, "mean_ep_length": 200.0, "mean_collisions": 5.4, "mean_intent_rate": 0.524, "policy_loss": 0.000844, "value_loss": 103.095121, "entropy": 0.541399, "total_loss": 51.542991, "clip_frac": 0.113647, "sps": 437.7, "elapsed_s": 979.4} |
| {"update": 209, "steps_done": 428032, "mean_step_reward": 0.6886, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 141.109, "mean_ep_length": 200.0, "mean_collisions": 2.4, "mean_intent_rate": 0.454, "policy_loss": 0.006833, "value_loss": 72.879185, "entropy": 0.508127, "total_loss": 36.441344, "clip_frac": 0.166748, "sps": 435.3, "elapsed_s": 984.1} |
| {"update": 210, "steps_done": 430080, "mean_step_reward": 0.7115, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 140.118, "mean_ep_length": 200.0, "mean_collisions": 3.0, "mean_intent_rate": 0.461, "policy_loss": 0.002389, "value_loss": 83.320297, "entropy": 0.530128, "total_loss": 41.657236, "clip_frac": 0.071777, "sps": 436.6, "elapsed_s": 988.8} |
| {"update": 211, "steps_done": 432128, "mean_step_reward": 0.9242, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 182.885, "mean_ep_length": 200.0, "mean_collisions": 2.55, "mean_intent_rate": 0.526, "policy_loss": 0.002112, "value_loss": 88.587013, "entropy": 0.528077, "total_loss": 44.290338, "clip_frac": 0.104492, "sps": 436.7, "elapsed_s": 993.5} |
| {"update": 212, "steps_done": 434176, "mean_step_reward": 0.8366, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 169.114, "mean_ep_length": 200.0, "mean_collisions": 4.0, "mean_intent_rate": 0.527, "policy_loss": 0.003652, "value_loss": 109.582285, "entropy": 0.532477, "total_loss": 54.78947, "clip_frac": 0.0802, "sps": 434.8, "elapsed_s": 998.2} |
| {"update": 213, "steps_done": 436224, "mean_step_reward": 0.8951, "min_step_reward": -10.519, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 181.834, "mean_ep_length": 200.0, "mean_collisions": 3.9, "mean_intent_rate": 0.544, "policy_loss": 0.001428, "value_loss": 96.826653, "entropy": 0.534656, "total_loss": 48.409407, "clip_frac": 0.081665, "sps": 434.1, "elapsed_s": 1002.9} |
| {"update": 214, "steps_done": 438272, "mean_step_reward": 0.7823, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 146.704, "mean_ep_length": 200.0, "mean_collisions": 4.8, "mean_intent_rate": 0.503, "policy_loss": 0.023359, "value_loss": 101.339829, "entropy": 0.492072, "total_loss": 50.688353, "clip_frac": 0.313721, "sps": 428.6, "elapsed_s": 1007.7} |
| {"update": 215, "steps_done": 440320, "mean_step_reward": 0.8172, "min_step_reward": -10.5847, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 172.409, "mean_ep_length": 200.0, "mean_collisions": 1.36, "mean_intent_rate": 0.489, "policy_loss": 0.00671, "value_loss": 86.234903, "entropy": 0.536947, "total_loss": 43.118792, "clip_frac": 0.167603, "sps": 436.9, "elapsed_s": 1012.4} |
| {"update": 216, "steps_done": 442368, "mean_step_reward": 0.882, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 157.889, "mean_ep_length": 187.9, "mean_collisions": 5.3, "mean_intent_rate": 0.543, "policy_loss": 0.007703, "value_loss": 90.422619, "entropy": 0.493226, "total_loss": 45.21408, "clip_frac": 0.174927, "sps": 438.2, "elapsed_s": 1017.0} |
| {"update": 217, "steps_done": 444416, "mean_step_reward": 0.633, "min_step_reward": -10.5873, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 137.281, "mean_ep_length": 200.0, "mean_collisions": 4.55, "mean_intent_rate": 0.483, "policy_loss": 0.00752, "value_loss": 88.412067, "entropy": 0.520795, "total_loss": 44.208345, "clip_frac": 0.150879, "sps": 436.1, "elapsed_s": 1021.7} |
| {"update": 218, "steps_done": 446464, "mean_step_reward": 0.8181, "min_step_reward": -10.5885, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 155.006, "mean_ep_length": 200.0, "mean_collisions": 5.0, "mean_intent_rate": 0.52, "policy_loss": 0.016522, "value_loss": 95.022897, "entropy": 0.49629, "total_loss": 47.523008, "clip_frac": 0.2229, "sps": 442.6, "elapsed_s": 1026.4} |
| {"update": 219, "steps_done": 448512, "mean_step_reward": 0.7368, "min_step_reward": -10.5835, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 162.227, "mean_ep_length": 200.0, "mean_collisions": 3.7, "mean_intent_rate": 0.511, "policy_loss": 0.006497, "value_loss": 82.0838, "entropy": 0.501842, "total_loss": 41.043378, "clip_frac": 0.118652, "sps": 440.9, "elapsed_s": 1031.0} |
| {"update": 220, "steps_done": 450560, "mean_step_reward": 0.8196, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 150.04, "mean_ep_length": 200.0, "mean_collisions": 1.4, "mean_intent_rate": 0.451, "policy_loss": 0.001309, "value_loss": 75.304947, "entropy": 0.495398, "total_loss": 37.648829, "clip_frac": 0.109619, "sps": 438.8, "elapsed_s": 1035.7} |
| {"update": 221, "steps_done": 452608, "mean_step_reward": 0.822, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 168.975, "mean_ep_length": 200.0, "mean_collisions": 6.55, "mean_intent_rate": 0.568, "policy_loss": 0.009283, "value_loss": 113.10594, "entropy": 0.523898, "total_loss": 56.557014, "clip_frac": 0.230225, "sps": 434.8, "elapsed_s": 1040.4} |
| {"update": 222, "steps_done": 454656, "mean_step_reward": 0.7439, "min_step_reward": -10.5288, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 147.581, "mean_ep_length": 200.0, "mean_collisions": 2.7, "mean_intent_rate": 0.471, "policy_loss": 0.009619, "value_loss": 84.484498, "entropy": 0.506198, "total_loss": 42.246806, "clip_frac": 0.127319, "sps": 436.5, "elapsed_s": 1045.1} |
| {"update": 223, "steps_done": 456704, "mean_step_reward": 0.6882, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 138.378, "mean_ep_length": 200.0, "mean_collisions": 6.2, "mean_intent_rate": 0.511, "policy_loss": 0.007016, "value_loss": 102.17161, "entropy": 0.555274, "total_loss": 51.087269, "clip_frac": 0.162354, "sps": 424.4, "elapsed_s": 1049.9} |
| {"update": 224, "steps_done": 458752, "mean_step_reward": 0.6918, "min_step_reward": -10.5808, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 136.965, "mean_ep_length": 200.0, "mean_collisions": 5.4, "mean_intent_rate": 0.495, "policy_loss": -0.000315, "value_loss": 96.69145, "entropy": 0.557925, "total_loss": 48.339831, "clip_frac": 0.091309, "sps": 438.4, "elapsed_s": 1054.6} |
| {"update": 225, "steps_done": 460800, "mean_step_reward": 0.9654, "min_step_reward": -10.5596, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 189.249, "mean_ep_length": 200.0, "mean_collisions": 3.91, "mean_intent_rate": 0.559, "policy_loss": 0.001067, "value_loss": 101.380837, "entropy": 0.534225, "total_loss": 50.686144, "clip_frac": 0.0979, "sps": 434.7, "elapsed_s": 1059.3} |
| {"update": 226, "steps_done": 462848, "mean_step_reward": 0.9683, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 188.492, "mean_ep_length": 193.4, "mean_collisions": 3.3, "mean_intent_rate": 0.562, "policy_loss": 0.004227, "value_loss": 105.489475, "entropy": 0.54364, "total_loss": 52.743528, "clip_frac": 0.070068, "sps": 432.5, "elapsed_s": 1064.0} |
| {"update": 227, "steps_done": 464896, "mean_step_reward": 0.7354, "min_step_reward": -10.5846, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 154.555, "mean_ep_length": 200.0, "mean_collisions": 3.1, "mean_intent_rate": 0.486, "policy_loss": 0.019078, "value_loss": 82.703459, "entropy": 0.530783, "total_loss": 41.365499, "clip_frac": 0.295044, "sps": 438.4, "elapsed_s": 1068.7} |
| {"update": 228, "steps_done": 466944, "mean_step_reward": 0.8503, "min_step_reward": -10.5187, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 162.893, "mean_ep_length": 200.0, "mean_collisions": 4.73, "mean_intent_rate": 0.529, "policy_loss": 0.012124, "value_loss": 111.491661, "entropy": 0.546192, "total_loss": 55.752493, "clip_frac": 0.250366, "sps": 434.8, "elapsed_s": 1073.4} |
| {"update": 229, "steps_done": 468992, "mean_step_reward": 0.8825, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 182.602, "mean_ep_length": 200.0, "mean_collisions": 2.5, "mean_intent_rate": 0.523, "policy_loss": -7.6e-05, "value_loss": 89.579675, "entropy": 0.520053, "total_loss": 44.784562, "clip_frac": 0.059082, "sps": 442.1, "elapsed_s": 1078.0} |
| {"update": 230, "steps_done": 471040, "mean_step_reward": 0.6006, "min_step_reward": -10.5605, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 124.01, "mean_ep_length": 200.0, "mean_collisions": 8.1, "mean_intent_rate": 0.518, "policy_loss": 0.005402, "value_loss": 106.164536, "entropy": 0.550624, "total_loss": 53.082163, "clip_frac": 0.145508, "sps": 437.7, "elapsed_s": 1082.7} |
| {"update": 231, "steps_done": 473088, "mean_step_reward": 0.9181, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 166.326, "mean_ep_length": 200.0, "mean_collisions": 5.3, "mean_intent_rate": 0.544, "policy_loss": 0.011212, "value_loss": 100.534341, "entropy": 0.494004, "total_loss": 50.273443, "clip_frac": 0.154541, "sps": 437.9, "elapsed_s": 1087.4} |
| {"update": 232, "steps_done": 475136, "mean_step_reward": 0.8434, "min_step_reward": -10.5571, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 175.697, "mean_ep_length": 200.0, "mean_collisions": 2.73, "mean_intent_rate": 0.517, "policy_loss": 0.01003, "value_loss": 108.707129, "entropy": 0.522619, "total_loss": 54.358369, "clip_frac": 0.181763, "sps": 427.6, "elapsed_s": 1092.2} |
| {"update": 233, "steps_done": 477184, "mean_step_reward": 0.9625, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 190.311, "mean_ep_length": 200.0, "mean_collisions": 1.8, "mean_intent_rate": 0.524, "policy_loss": 0.005012, "value_loss": 92.269559, "entropy": 0.501702, "total_loss": 46.134775, "clip_frac": 0.158813, "sps": 433.8, "elapsed_s": 1096.9} |
| {"update": 234, "steps_done": 479232, "mean_step_reward": 0.6425, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 131.133, "mean_ep_length": 200.0, "mean_collisions": 2.5, "mean_intent_rate": 0.439, "policy_loss": 0.013994, "value_loss": 85.880755, "entropy": 0.451814, "total_loss": 42.949853, "clip_frac": 0.133179, "sps": 439.3, "elapsed_s": 1101.6} |
| {"update": 235, "steps_done": 481280, "mean_step_reward": 0.8293, "min_step_reward": -10.5736, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 174.592, "mean_ep_length": 200.0, "mean_collisions": 3.0, "mean_intent_rate": 0.519, "policy_loss": 0.009337, "value_loss": 92.390123, "entropy": 0.504419, "total_loss": 46.199354, "clip_frac": 0.120483, "sps": 437.5, "elapsed_s": 1106.3} |
| {"update": 236, "steps_done": 483328, "mean_step_reward": 0.8117, "min_step_reward": -10.584, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 155.26, "mean_ep_length": 200.0, "mean_collisions": 1.09, "mean_intent_rate": 0.455, "policy_loss": 0.004124, "value_loss": 89.791521, "entropy": 0.478783, "total_loss": 44.895097, "clip_frac": 0.113403, "sps": 439.3, "elapsed_s": 1110.9} |
| {"update": 237, "steps_done": 485376, "mean_step_reward": 0.8385, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 166.893, "mean_ep_length": 200.0, "mean_collisions": 3.6, "mean_intent_rate": 0.516, "policy_loss": 0.019362, "value_loss": 105.5083, "entropy": 0.463604, "total_loss": 52.768876, "clip_frac": 0.182983, "sps": 438.3, "elapsed_s": 1115.6} |
| {"update": 238, "steps_done": 487424, "mean_step_reward": 0.916, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 188.938, "mean_ep_length": 200.0, "mean_collisions": 4.4, "mean_intent_rate": 0.565, "policy_loss": 0.00499, "value_loss": 105.188854, "entropy": 0.497189, "total_loss": 52.594446, "clip_frac": 0.120972, "sps": 435.7, "elapsed_s": 1120.3} |
| {"update": 239, "steps_done": 489472, "mean_step_reward": 1.0463, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 195.894, "mean_ep_length": 200.0, "mean_collisions": 3.8, "mean_intent_rate": 0.568, "policy_loss": 0.004936, "value_loss": 118.794709, "entropy": 0.503782, "total_loss": 59.397254, "clip_frac": 0.132324, "sps": 432.8, "elapsed_s": 1125.0} |
| {"update": 240, "steps_done": 491520, "mean_step_reward": 0.7326, "min_step_reward": -10.5806, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 157.231, "mean_ep_length": 200.0, "mean_collisions": 5.18, "mean_intent_rate": 0.527, "policy_loss": 0.022882, "value_loss": 113.537678, "entropy": 0.501848, "total_loss": 56.786702, "clip_frac": 0.286499, "sps": 436.8, "elapsed_s": 1129.7} |
| {"update": 241, "steps_done": 493568, "mean_step_reward": 0.8055, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 159.418, "mean_ep_length": 200.0, "mean_collisions": 4.8, "mean_intent_rate": 0.524, "policy_loss": 0.015258, "value_loss": 102.401557, "entropy": 0.460747, "total_loss": 51.211429, "clip_frac": 0.252686, "sps": 433.1, "elapsed_s": 1134.4} |
| {"update": 242, "steps_done": 495616, "mean_step_reward": 0.8828, "min_step_reward": -7.5873, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 179.56, "mean_ep_length": 200.0, "mean_collisions": 1.2, "mean_intent_rate": 0.495, "policy_loss": 0.003077, "value_loss": 88.755168, "entropy": 0.478404, "total_loss": 44.375877, "clip_frac": 0.104126, "sps": 423.1, "elapsed_s": 1139.3} |
| {"update": 243, "steps_done": 497664, "mean_step_reward": 0.6919, "min_step_reward": -10.5786, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 136.83, "mean_ep_length": 200.0, "mean_collisions": 4.1, "mean_intent_rate": 0.474, "policy_loss": 0.008341, "value_loss": 91.10239, "entropy": 0.477585, "total_loss": 45.55476, "clip_frac": 0.16333, "sps": 429.3, "elapsed_s": 1144.1} |
| {"update": 244, "steps_done": 499712, "mean_step_reward": 0.9495, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 186.712, "mean_ep_length": 200.0, "mean_collisions": 3.36, "mean_intent_rate": 0.546, "policy_loss": 0.002571, "value_loss": 105.881665, "entropy": 0.484895, "total_loss": 52.938554, "clip_frac": 0.072388, "sps": 438.3, "elapsed_s": 1148.7} |
| {"update": 245, "steps_done": 501760, "mean_step_reward": 0.9068, "min_step_reward": -10.5192, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 184.283, "mean_ep_length": 200.0, "mean_collisions": 3.7, "mean_intent_rate": 0.546, "policy_loss": 0.001091, "value_loss": 100.754241, "entropy": 0.499542, "total_loss": 50.373216, "clip_frac": 0.071289, "sps": 433.2, "elapsed_s": 1153.5} |
| {"update": 246, "steps_done": 503808, "mean_step_reward": 0.7889, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 148.45, "mean_ep_length": 200.0, "mean_collisions": 3.9, "mean_intent_rate": 0.489, "policy_loss": 0.001563, "value_loss": 102.546503, "entropy": 0.515633, "total_loss": 51.269658, "clip_frac": 0.108032, "sps": 440.5, "elapsed_s": 1158.1} |
| {"update": 247, "steps_done": 505856, "mean_step_reward": 0.8571, "min_step_reward": -10.5845, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 175.102, "mean_ep_length": 200.0, "mean_collisions": 5.0, "mean_intent_rate": 0.552, "policy_loss": -0.000294, "value_loss": 86.677044, "entropy": 0.49984, "total_loss": 43.333229, "clip_frac": 0.045898, "sps": 437.3, "elapsed_s": 1162.8} |
| {"update": 248, "steps_done": 507904, "mean_step_reward": 0.8258, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 158.057, "mean_ep_length": 200.0, "mean_collisions": 5.9, "mean_intent_rate": 0.54, "policy_loss": 0.013565, "value_loss": 95.36911, "entropy": 0.477169, "total_loss": 47.693349, "clip_frac": 0.209717, "sps": 436.6, "elapsed_s": 1167.5} |
| {"update": 249, "steps_done": 509952, "mean_step_reward": 0.8378, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 175.122, "mean_ep_length": 200.0, "mean_collisions": 9.73, "mean_intent_rate": 0.634, "policy_loss": 0.0073, "value_loss": 160.660799, "entropy": 0.496016, "total_loss": 80.332739, "clip_frac": 0.172119, "sps": 425.8, "elapsed_s": 1172.3} |
| {"update": 250, "steps_done": 512000, "mean_step_reward": 0.9966, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 196.674, "mean_ep_length": 200.0, "mean_collisions": 5.5, "mean_intent_rate": 0.597, "policy_loss": 0.005285, "value_loss": 102.109416, "entropy": 0.498936, "total_loss": 51.055002, "clip_frac": 0.113281, "sps": 418.4, "elapsed_s": 1177.2} |
| {"update": 251, "steps_done": 514048, "mean_step_reward": 1.0146, "min_step_reward": -10.5466, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 200.129, "mean_ep_length": 193.7, "mean_collisions": 4.09, "mean_intent_rate": 0.594, "policy_loss": 0.000351, "value_loss": 99.424313, "entropy": 0.488693, "total_loss": 49.70762, "clip_frac": 0.103027, "sps": 431.5, "elapsed_s": 1181.9} |
| {"update": 252, "steps_done": 516096, "mean_step_reward": 0.8669, "min_step_reward": -10.5158, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 173.342, "mean_ep_length": 200.0, "mean_collisions": 4.5, "mean_intent_rate": 0.54, "policy_loss": 0.010277, "value_loss": 93.377761, "entropy": 0.453054, "total_loss": 46.694628, "clip_frac": 0.177246, "sps": 439.3, "elapsed_s": 1186.6} |
| {"update": 253, "steps_done": 518144, "mean_step_reward": 0.762, "min_step_reward": -10.5868, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 150.954, "mean_ep_length": 200.0, "mean_collisions": 1.9, "mean_intent_rate": 0.461, "policy_loss": 0.003917, "value_loss": 76.936226, "entropy": 0.446466, "total_loss": 38.467565, "clip_frac": 0.080688, "sps": 440.9, "elapsed_s": 1191.3} |
| {"update": 254, "steps_done": 520192, "mean_step_reward": 0.7164, "min_step_reward": -10.5764, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 136.201, "mean_ep_length": 182.7, "mean_collisions": 3.36, "mean_intent_rate": 0.498, "policy_loss": 0.003038, "value_loss": 89.209943, "entropy": 0.426721, "total_loss": 44.603743, "clip_frac": 0.067383, "sps": 442.9, "elapsed_s": 1195.9} |
| {"update": 255, "steps_done": 522240, "mean_step_reward": 0.96, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 183.841, "mean_ep_length": 200.0, "mean_collisions": 0.7, "mean_intent_rate": 0.493, "policy_loss": 0.021557, "value_loss": 80.981442, "entropy": 0.458496, "total_loss": 40.507693, "clip_frac": 0.299194, "sps": 437.2, "elapsed_s": 1200.6} |
| {"update": 256, "steps_done": 524288, "mean_step_reward": 0.8942, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 179.187, "mean_ep_length": 200.0, "mean_collisions": 6.73, "mean_intent_rate": 0.588, "policy_loss": 0.009046, "value_loss": 117.620785, "entropy": 0.493699, "total_loss": 58.814502, "clip_frac": 0.213257, "sps": 435.7, "elapsed_s": 1205.3} |
| {"update": 257, "steps_done": 526336, "mean_step_reward": 0.8888, "min_step_reward": -10.5426, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 175.234, "mean_ep_length": 200.0, "mean_collisions": 4.4, "mean_intent_rate": 0.542, "policy_loss": 0.000536, "value_loss": 88.645977, "entropy": 0.527992, "total_loss": 44.318244, "clip_frac": 0.11792, "sps": 434.1, "elapsed_s": 1210.0} |
| {"update": 258, "steps_done": 528384, "mean_step_reward": 0.9065, "min_step_reward": -10.53, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 198.043, "mean_ep_length": 200.0, "mean_collisions": 5.2, "mean_intent_rate": 0.594, "policy_loss": 0.006708, "value_loss": 110.10925, "entropy": 0.488375, "total_loss": 55.056449, "clip_frac": 0.196045, "sps": 429.4, "elapsed_s": 1214.8} |
| {"update": 259, "steps_done": 530432, "mean_step_reward": 0.7457, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 142.537, "mean_ep_length": 200.0, "mean_collisions": 12.6, "mean_intent_rate": 0.625, "policy_loss": -0.001099, "value_loss": 126.604704, "entropy": 0.495218, "total_loss": 63.296301, "clip_frac": 0.084717, "sps": 426.1, "elapsed_s": 1219.6} |
| {"update": 260, "steps_done": 532480, "mean_step_reward": 0.8926, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 12, "mean_ep_return": 158.192, "mean_ep_length": 183.5, "mean_collisions": 6.83, "mean_intent_rate": 0.54, "policy_loss": 0.0064, "value_loss": 130.389517, "entropy": 0.506735, "total_loss": 65.196091, "clip_frac": 0.106689, "sps": 441.0, "elapsed_s": 1224.2} |
| {"update": 261, "steps_done": 534528, "mean_step_reward": 0.7203, "min_step_reward": -10.5801, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 146.999, "mean_ep_length": 200.0, "mean_collisions": 8.5, "mean_intent_rate": 0.565, "policy_loss": 0.030229, "value_loss": 119.901496, "entropy": 0.458092, "total_loss": 59.976396, "clip_frac": 0.310303, "sps": 439.9, "elapsed_s": 1228.9} |
| {"update": 262, "steps_done": 536576, "mean_step_reward": 0.7243, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 142.945, "mean_ep_length": 200.0, "mean_collisions": 3.8, "mean_intent_rate": 0.48, "policy_loss": 0.010237, "value_loss": 129.283211, "entropy": 0.471273, "total_loss": 64.64713, "clip_frac": 0.161011, "sps": 437.7, "elapsed_s": 1233.5} |
| {"update": 263, "steps_done": 538624, "mean_step_reward": 1.1278, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 219.675, "mean_ep_length": 200.0, "mean_collisions": 3.4, "mean_intent_rate": 0.599, "policy_loss": 0.014958, "value_loss": 106.420498, "entropy": 0.454909, "total_loss": 53.220658, "clip_frac": 0.226074, "sps": 430.9, "elapsed_s": 1238.3} |
| {"update": 264, "steps_done": 540672, "mean_step_reward": 0.865, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 175.6, "mean_ep_length": 200.0, "mean_collisions": 2.45, "mean_intent_rate": 0.511, "policy_loss": 0.001608, "value_loss": 113.309663, "entropy": 0.456113, "total_loss": 56.651878, "clip_frac": 0.141968, "sps": 431.5, "elapsed_s": 1243.0} |
| {"update": 265, "steps_done": 542720, "mean_step_reward": 0.9259, "min_step_reward": -10.5491, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 183.84, "mean_ep_length": 200.0, "mean_collisions": 5.1, "mean_intent_rate": 0.571, "policy_loss": 0.008154, "value_loss": 117.718343, "entropy": 0.460901, "total_loss": 58.862716, "clip_frac": 0.155029, "sps": 432.3, "elapsed_s": 1247.8} |
| {"update": 266, "steps_done": 544768, "mean_step_reward": 0.8881, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 173.916, "mean_ep_length": 200.0, "mean_collisions": 4.4, "mean_intent_rate": 0.541, "policy_loss": 0.007657, "value_loss": 104.703843, "entropy": 0.385971, "total_loss": 52.355719, "clip_frac": 0.153809, "sps": 440.9, "elapsed_s": 1252.4} |
| {"update": 267, "steps_done": 546816, "mean_step_reward": 0.9634, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 200.741, "mean_ep_length": 200.0, "mean_collisions": 1.3, "mean_intent_rate": 0.533, "policy_loss": 0.007549, "value_loss": 101.769512, "entropy": 0.419876, "total_loss": 50.888106, "clip_frac": 0.140137, "sps": 432.3, "elapsed_s": 1257.2} |
| {"update": 268, "steps_done": 548864, "mean_step_reward": 0.7305, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 148.181, "mean_ep_length": 200.0, "mean_collisions": 5.64, "mean_intent_rate": 0.519, "policy_loss": 0.023733, "value_loss": 118.46856, "entropy": 0.462311, "total_loss": 59.25339, "clip_frac": 0.281616, "sps": 436.4, "elapsed_s": 1261.9} |
| {"update": 269, "steps_done": 550912, "mean_step_reward": 0.9901, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 195.942, "mean_ep_length": 197.9, "mean_collisions": 3.9, "mean_intent_rate": 0.571, "policy_loss": 0.006491, "value_loss": 98.387148, "entropy": 0.453394, "total_loss": 49.195532, "clip_frac": 0.148193, "sps": 432.3, "elapsed_s": 1266.6} |
| {"update": 270, "steps_done": 552960, "mean_step_reward": 0.8728, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 182.468, "mean_ep_length": 200.0, "mean_collisions": 4.1, "mean_intent_rate": 0.549, "policy_loss": 0.013894, "value_loss": 102.515797, "entropy": 0.437271, "total_loss": 51.26742, "clip_frac": 0.246216, "sps": 439.5, "elapsed_s": 1271.3} |
| {"update": 271, "steps_done": 555008, "mean_step_reward": 1.0077, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 185.831, "mean_ep_length": 200.0, "mean_collisions": 6.8, "mean_intent_rate": 0.601, "policy_loss": 0.007687, "value_loss": 135.583676, "entropy": 0.444301, "total_loss": 67.795081, "clip_frac": 0.14624, "sps": 438.3, "elapsed_s": 1275.9} |
| {"update": 272, "steps_done": 557056, "mean_step_reward": 0.9082, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 189.12, "mean_ep_length": 200.0, "mean_collisions": 9.0, "mean_intent_rate": 0.643, "policy_loss": 0.011284, "value_loss": 115.255822, "entropy": 0.448215, "total_loss": 57.634714, "clip_frac": 0.176025, "sps": 432.9, "elapsed_s": 1280.7} |
| {"update": 273, "steps_done": 559104, "mean_step_reward": 0.8656, "min_step_reward": -10.5806, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 171.848, "mean_ep_length": 200.0, "mean_collisions": 7.3, "mean_intent_rate": 0.588, "policy_loss": 0.00479, "value_loss": 110.741543, "entropy": 0.440905, "total_loss": 55.371152, "clip_frac": 0.090454, "sps": 441.5, "elapsed_s": 1285.3} |
| {"update": 274, "steps_done": 561152, "mean_step_reward": 1.0359, "min_step_reward": -10.5694, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 205.373, "mean_ep_length": 200.0, "mean_collisions": 3.8, "mean_intent_rate": 0.583, "policy_loss": 0.010478, "value_loss": 109.298987, "entropy": 0.448606, "total_loss": 54.655486, "clip_frac": 0.185059, "sps": 439.7, "elapsed_s": 1290.0} |
| {"update": 275, "steps_done": 563200, "mean_step_reward": 0.983, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 200.611, "mean_ep_length": 200.0, "mean_collisions": 4.7, "mean_intent_rate": 0.588, "policy_loss": 0.006173, "value_loss": 100.716499, "entropy": 0.453494, "total_loss": 50.359887, "clip_frac": 0.178101, "sps": 433.8, "elapsed_s": 1294.7} |
| {"update": 276, "steps_done": 565248, "mean_step_reward": 0.9765, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 193.075, "mean_ep_length": 200.0, "mean_collisions": 7.0, "mean_intent_rate": 0.616, "policy_loss": 0.013942, "value_loss": 130.851978, "entropy": 0.450563, "total_loss": 65.435425, "clip_frac": 0.194458, "sps": 423.9, "elapsed_s": 1299.5} |
| {"update": 277, "steps_done": 567296, "mean_step_reward": 0.7768, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 152.72, "mean_ep_length": 200.0, "mean_collisions": 10.7, "mean_intent_rate": 0.612, "policy_loss": 0.014132, "value_loss": 139.702194, "entropy": 0.457275, "total_loss": 69.860656, "clip_frac": 0.191895, "sps": 436.9, "elapsed_s": 1304.2} |
| {"update": 278, "steps_done": 569344, "mean_step_reward": 0.9381, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 184.09, "mean_ep_length": 200.0, "mean_collisions": 8.2, "mean_intent_rate": 0.622, "policy_loss": 0.01145, "value_loss": 108.549584, "entropy": 0.436159, "total_loss": 54.281881, "clip_frac": 0.200073, "sps": 430.6, "elapsed_s": 1309.0} |
| {"update": 279, "steps_done": 571392, "mean_step_reward": 1.1737, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 237.804, "mean_ep_length": 200.0, "mean_collisions": 3.9, "mean_intent_rate": 0.638, "policy_loss": 0.003533, "value_loss": 111.184307, "entropy": 0.478251, "total_loss": 55.590904, "clip_frac": 0.077026, "sps": 439.9, "elapsed_s": 1313.6} |
| {"update": 280, "steps_done": 573440, "mean_step_reward": 0.8187, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 172.701, "mean_ep_length": 200.0, "mean_collisions": 9.18, "mean_intent_rate": 0.619, "policy_loss": 0.026637, "value_loss": 157.705946, "entropy": 0.519189, "total_loss": 78.874418, "clip_frac": 0.262573, "sps": 447.4, "elapsed_s": 1318.2} |
| {"update": 281, "steps_done": 575488, "mean_step_reward": 0.4946, "min_step_reward": -10.5762, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 97.716, "mean_ep_length": 200.0, "mean_collisions": 17.4, "mean_intent_rate": 0.633, "policy_loss": 0.001966, "value_loss": 129.403053, "entropy": 0.519988, "total_loss": 64.698292, "clip_frac": 0.153809, "sps": 441.0, "elapsed_s": 1322.8} |
| {"update": 282, "steps_done": 577536, "mean_step_reward": 0.7995, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 161.202, "mean_ep_length": 200.0, "mean_collisions": 13.0, "mean_intent_rate": 0.665, "policy_loss": 0.001049, "value_loss": 119.862521, "entropy": 0.507194, "total_loss": 59.927238, "clip_frac": 0.135376, "sps": 437.7, "elapsed_s": 1327.5} |
| {"update": 283, "steps_done": 579584, "mean_step_reward": 0.7988, "min_step_reward": -10.5876, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 155.118, "mean_ep_length": 200.0, "mean_collisions": 11.5, "mean_intent_rate": 0.629, "policy_loss": 0.008063, "value_loss": 109.622442, "entropy": 0.528248, "total_loss": 54.814002, "clip_frac": 0.161133, "sps": 432.4, "elapsed_s": 1332.3} |
| {"update": 284, "steps_done": 581632, "mean_step_reward": 0.9754, "min_step_reward": -10.535, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 192.527, "mean_ep_length": 200.0, "mean_collisions": 10.09, "mean_intent_rate": 0.669, "policy_loss": 0.018411, "value_loss": 114.838543, "entropy": 0.475996, "total_loss": 57.432923, "clip_frac": 0.218018, "sps": 435.7, "elapsed_s": 1337.0} |
| {"update": 285, "steps_done": 583680, "mean_step_reward": 0.8287, "min_step_reward": -10.5147, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 165.49, "mean_ep_length": 199.9, "mean_collisions": 11.8, "mean_intent_rate": 0.651, "policy_loss": 0.01083, "value_loss": 113.141587, "entropy": 0.482382, "total_loss": 56.576801, "clip_frac": 0.203369, "sps": 428.0, "elapsed_s": 1341.7} |
| {"update": 286, "steps_done": 585728, "mean_step_reward": 0.724, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 145.869, "mean_ep_length": 200.0, "mean_collisions": 9.7, "mean_intent_rate": 0.584, "policy_loss": 0.005524, "value_loss": 99.337375, "entropy": 0.538628, "total_loss": 49.668825, "clip_frac": 0.119629, "sps": 433.2, "elapsed_s": 1346.5} |
| {"update": 287, "steps_done": 587776, "mean_step_reward": 1.0445, "min_step_reward": -10.5722, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 209.318, "mean_ep_length": 200.0, "mean_collisions": 5.3, "mean_intent_rate": 0.614, "policy_loss": 0.024157, "value_loss": 105.684106, "entropy": 0.490117, "total_loss": 52.861308, "clip_frac": 0.258911, "sps": 435.4, "elapsed_s": 1351.2} |
| {"update": 288, "steps_done": 589824, "mean_step_reward": 0.9192, "min_step_reward": -10.5394, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 180.304, "mean_ep_length": 200.0, "mean_collisions": 7.2, "mean_intent_rate": 0.597, "policy_loss": -0.000224, "value_loss": 114.856252, "entropy": 0.485916, "total_loss": 57.423043, "clip_frac": 0.029785, "sps": 438.0, "elapsed_s": 1355.9} |
| {"update": 289, "steps_done": 591872, "mean_step_reward": 0.8909, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 180.314, "mean_ep_length": 200.0, "mean_collisions": 6.36, "mean_intent_rate": 0.584, "policy_loss": 0.014025, "value_loss": 112.360902, "entropy": 0.489211, "total_loss": 56.189584, "clip_frac": 0.195312, "sps": 440.5, "elapsed_s": 1360.5} |
| {"update": 290, "steps_done": 593920, "mean_step_reward": 0.9248, "min_step_reward": -10.5187, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 186.318, "mean_ep_length": 200.0, "mean_collisions": 4.2, "mean_intent_rate": 0.557, "policy_loss": -0.002404, "value_loss": 103.008513, "entropy": 0.502749, "total_loss": 51.496825, "clip_frac": 0.051636, "sps": 442.9, "elapsed_s": 1365.1} |
| {"update": 291, "steps_done": 595968, "mean_step_reward": 1.0015, "min_step_reward": -10.5766, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 199.888, "mean_ep_length": 200.0, "mean_collisions": 5.6, "mean_intent_rate": 0.603, "policy_loss": 0.001244, "value_loss": 105.73334, "entropy": 0.505762, "total_loss": 52.862856, "clip_frac": 0.046631, "sps": 441.2, "elapsed_s": 1369.8} |
| {"update": 292, "steps_done": 598016, "mean_step_reward": 1.0558, "min_step_reward": -10.5651, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 206.381, "mean_ep_length": 200.0, "mean_collisions": 4.0, "mean_intent_rate": 0.589, "policy_loss": 0.013425, "value_loss": 110.501718, "entropy": 0.444191, "total_loss": 55.259842, "clip_frac": 0.183472, "sps": 437.1, "elapsed_s": 1374.5} |
| {"update": 293, "steps_done": 600064, "mean_step_reward": 1.0408, "min_step_reward": -7.5863, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 210.566, "mean_ep_length": 200.0, "mean_collisions": 2.55, "mean_intent_rate": 0.571, "policy_loss": 0.002219, "value_loss": 108.25904, "entropy": 0.437636, "total_loss": 54.127363, "clip_frac": 0.118042, "sps": 437.5, "elapsed_s": 1379.1} |
| {"update": 294, "steps_done": 602112, "mean_step_reward": 1.0972, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 220.421, "mean_ep_length": 200.0, "mean_collisions": 2.7, "mean_intent_rate": 0.588, "policy_loss": 0.020789, "value_loss": 101.313844, "entropy": 0.475215, "total_loss": 50.672959, "clip_frac": 0.276611, "sps": 425.2, "elapsed_s": 1384.0} |
| {"update": 295, "steps_done": 604160, "mean_step_reward": 0.9344, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 179.632, "mean_ep_length": 200.0, "mean_collisions": 1.3, "mean_intent_rate": 0.497, "policy_loss": 0.004638, "value_loss": 102.677285, "entropy": 0.474395, "total_loss": 51.338536, "clip_frac": 0.199341, "sps": 440.3, "elapsed_s": 1388.6} |
| {"update": 296, "steps_done": 606208, "mean_step_reward": 1.1661, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 246.037, "mean_ep_length": 200.0, "mean_collisions": 2.7, "mean_intent_rate": 0.634, "policy_loss": 0.009165, "value_loss": 125.633858, "entropy": 0.441694, "total_loss": 62.821677, "clip_frac": 0.128418, "sps": 437.6, "elapsed_s": 1393.3} |
| {"update": 297, "steps_done": 608256, "mean_step_reward": 1.0083, "min_step_reward": -10.5194, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 195.515, "mean_ep_length": 200.0, "mean_collisions": 4.82, "mean_intent_rate": 0.584, "policy_loss": 0.018521, "value_loss": 123.420049, "entropy": 0.454154, "total_loss": 61.724004, "clip_frac": 0.278198, "sps": 435.4, "elapsed_s": 1398.0} |
| {"update": 298, "steps_done": 610304, "mean_step_reward": 0.9556, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 190.778, "mean_ep_length": 200.0, "mean_collisions": 5.6, "mean_intent_rate": 0.588, "policy_loss": 0.020692, "value_loss": 126.367929, "entropy": 0.461697, "total_loss": 63.20004, "clip_frac": 0.239624, "sps": 437.9, "elapsed_s": 1402.7} |
| {"update": 299, "steps_done": 612352, "mean_step_reward": 0.9479, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 186.207, "mean_ep_length": 200.0, "mean_collisions": 3.7, "mean_intent_rate": 0.549, "policy_loss": 0.005386, "value_loss": 115.621491, "entropy": 0.466434, "total_loss": 57.811467, "clip_frac": 0.144409, "sps": 435.2, "elapsed_s": 1407.4} |
| {"update": 300, "steps_done": 614400, "mean_step_reward": 1.0343, "min_step_reward": -10.5622, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 215.738, "mean_ep_length": 200.0, "mean_collisions": 2.8, "mean_intent_rate": 0.583, "policy_loss": 0.002559, "value_loss": 103.872114, "entropy": 0.457093, "total_loss": 51.934045, "clip_frac": 0.099487, "sps": 431.2, "elapsed_s": 1412.1} |
| {"update": 301, "steps_done": 616448, "mean_step_reward": 0.9666, "min_step_reward": -10.587, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 191.151, "mean_ep_length": 200.0, "mean_collisions": 4.45, "mean_intent_rate": 0.57, "policy_loss": 0.016573, "value_loss": 118.651376, "entropy": 0.446977, "total_loss": 59.337791, "clip_frac": 0.204956, "sps": 440.6, "elapsed_s": 1416.8} |
| {"update": 302, "steps_done": 618496, "mean_step_reward": 1.0458, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 206.102, "mean_ep_length": 200.0, "mean_collisions": 4.8, "mean_intent_rate": 0.601, "policy_loss": 0.014147, "value_loss": 115.135825, "entropy": 0.436771, "total_loss": 57.577692, "clip_frac": 0.2229, "sps": 442.2, "elapsed_s": 1421.4} |
| {"update": 303, "steps_done": 620544, "mean_step_reward": 0.9181, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 179.148, "mean_ep_length": 200.0, "mean_collisions": 8.6, "mean_intent_rate": 0.619, "policy_loss": 0.040095, "value_loss": 134.928749, "entropy": 0.461162, "total_loss": 67.499858, "clip_frac": 0.337158, "sps": 427.5, "elapsed_s": 1426.2} |
| {"update": 304, "steps_done": 622592, "mean_step_reward": 0.9043, "min_step_reward": -10.5744, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 178.425, "mean_ep_length": 200.0, "mean_collisions": 7.6, "mean_intent_rate": 0.601, "policy_loss": 0.003584, "value_loss": 112.870434, "entropy": 0.461096, "total_loss": 56.43419, "clip_frac": 0.13916, "sps": 441.2, "elapsed_s": 1430.9} |
| {"update": 305, "steps_done": 624640, "mean_step_reward": 0.9068, "min_step_reward": -10.5588, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 179.448, "mean_ep_length": 190.4, "mean_collisions": 8.55, "mean_intent_rate": 0.645, "policy_loss": 0.007667, "value_loss": 121.566038, "entropy": 0.494126, "total_loss": 60.785745, "clip_frac": 0.138916, "sps": 438.1, "elapsed_s": 1435.5} |
| {"update": 306, "steps_done": 626688, "mean_step_reward": 0.8759, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 176.399, "mean_ep_length": 200.0, "mean_collisions": 10.8, "mean_intent_rate": 0.651, "policy_loss": 0.011734, "value_loss": 136.590761, "entropy": 0.467536, "total_loss": 68.302439, "clip_frac": 0.153809, "sps": 437.1, "elapsed_s": 1440.2} |
| {"update": 307, "steps_done": 628736, "mean_step_reward": 0.8386, "min_step_reward": -10.5801, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 163.966, "mean_ep_length": 193.8, "mean_collisions": 9.91, "mean_intent_rate": 0.629, "policy_loss": 0.009129, "value_loss": 129.418419, "entropy": 0.469384, "total_loss": 64.713644, "clip_frac": 0.166992, "sps": 439.3, "elapsed_s": 1444.9} |
| {"update": 308, "steps_done": 630784, "mean_step_reward": 0.8691, "min_step_reward": -10.5191, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 166.489, "mean_ep_length": 200.0, "mean_collisions": 10.6, "mean_intent_rate": 0.633, "policy_loss": 0.000568, "value_loss": 120.551909, "entropy": 0.431557, "total_loss": 60.272207, "clip_frac": 0.122437, "sps": 438.1, "elapsed_s": 1449.6} |
| {"update": 309, "steps_done": 632832, "mean_step_reward": 0.9029, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 193.08, "mean_ep_length": 200.0, "mean_collisions": 8.7, "mean_intent_rate": 0.647, "policy_loss": 0.000141, "value_loss": 110.363559, "entropy": 0.437322, "total_loss": 55.177547, "clip_frac": 0.089355, "sps": 438.1, "elapsed_s": 1454.2} |
| {"update": 310, "steps_done": 634880, "mean_step_reward": 1.1219, "min_step_reward": -10.5291, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 215.141, "mean_ep_length": 200.0, "mean_collisions": 3.91, "mean_intent_rate": 0.601, "policy_loss": 0.00755, "value_loss": 113.068936, "entropy": 0.43749, "total_loss": 56.537643, "clip_frac": 0.124756, "sps": 432.9, "elapsed_s": 1459.0} |
| {"update": 311, "steps_done": 636928, "mean_step_reward": 0.9393, "min_step_reward": -10.54, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 186.564, "mean_ep_length": 200.0, "mean_collisions": 5.0, "mean_intent_rate": 0.572, "policy_loss": 0.014244, "value_loss": 100.632625, "entropy": 0.41992, "total_loss": 50.326358, "clip_frac": 0.219116, "sps": 433.1, "elapsed_s": 1463.7} |
| {"update": 312, "steps_done": 638976, "mean_step_reward": 1.0735, "min_step_reward": -7.5853, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 214.455, "mean_ep_length": 200.0, "mean_collisions": 3.4, "mean_intent_rate": 0.592, "policy_loss": 0.007671, "value_loss": 114.51165, "entropy": 0.431499, "total_loss": 57.259181, "clip_frac": 0.12085, "sps": 425.3, "elapsed_s": 1468.5} |
| {"update": 313, "steps_done": 641024, "mean_step_reward": 0.8215, "min_step_reward": -10.586, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 159.914, "mean_ep_length": 200.0, "mean_collisions": 0.9, "mean_intent_rate": 0.46, "policy_loss": 0.009669, "value_loss": 81.806349, "entropy": 0.378686, "total_loss": 40.909057, "clip_frac": 0.171021, "sps": 439.8, "elapsed_s": 1473.2} |
| {"update": 314, "steps_done": 643072, "mean_step_reward": 0.8985, "min_step_reward": -10.5391, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 186.672, "mean_ep_length": 200.0, "mean_collisions": 3.64, "mean_intent_rate": 0.55, "policy_loss": 0.001805, "value_loss": 107.624458, "entropy": 0.421539, "total_loss": 53.809819, "clip_frac": 0.127319, "sps": 440.2, "elapsed_s": 1477.8} |
| {"update": 315, "steps_done": 645120, "mean_step_reward": 0.9675, "min_step_reward": -7.5882, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 176.482, "mean_ep_length": 183.5, "mean_collisions": 1.91, "mean_intent_rate": 0.533, "policy_loss": 0.004404, "value_loss": 102.355608, "entropy": 0.414205, "total_loss": 51.178067, "clip_frac": 0.123291, "sps": 438.3, "elapsed_s": 1482.5} |
| {"update": 316, "steps_done": 647168, "mean_step_reward": 1.1144, "min_step_reward": -7.5781, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 228.304, "mean_ep_length": 200.0, "mean_collisions": 1.3, "mean_intent_rate": 0.579, "policy_loss": 0.01303, "value_loss": 106.863434, "entropy": 0.403927, "total_loss": 53.440707, "clip_frac": 0.209961, "sps": 435.7, "elapsed_s": 1487.2} |
| {"update": 317, "steps_done": 649216, "mean_step_reward": 1.0397, "min_step_reward": -10.5201, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 201.74, "mean_ep_length": 200.0, "mean_collisions": 4.3, "mean_intent_rate": 0.586, "policy_loss": 0.005698, "value_loss": 112.076076, "entropy": 0.409353, "total_loss": 56.039642, "clip_frac": 0.120117, "sps": 438.5, "elapsed_s": 1491.9} |
| {"update": 318, "steps_done": 651264, "mean_step_reward": 0.9045, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 179.916, "mean_ep_length": 200.0, "mean_collisions": 5.4, "mean_intent_rate": 0.568, "policy_loss": 0.001927, "value_loss": 110.091162, "entropy": 0.430338, "total_loss": 55.043204, "clip_frac": 0.093872, "sps": 437.1, "elapsed_s": 1496.6} |
| {"update": 319, "steps_done": 653312, "mean_step_reward": 1.0308, "min_step_reward": -10.5535, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 206.496, "mean_ep_length": 200.0, "mean_collisions": 3.09, "mean_intent_rate": 0.573, "policy_loss": 0.005057, "value_loss": 120.671333, "entropy": 0.416743, "total_loss": 60.336556, "clip_frac": 0.129272, "sps": 435.7, "elapsed_s": 1501.3} |
| {"update": 320, "steps_done": 655360, "mean_step_reward": 1.1435, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 228.27, "mean_ep_length": 200.0, "mean_collisions": 3.1, "mean_intent_rate": 0.609, "policy_loss": 0.007252, "value_loss": 119.135374, "entropy": 0.407888, "total_loss": 59.57086, "clip_frac": 0.113892, "sps": 435.9, "elapsed_s": 1506.0} |
| {"update": 321, "steps_done": 657408, "mean_step_reward": 1.0779, "min_step_reward": -10.5428, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 220.48, "mean_ep_length": 200.0, "mean_collisions": 2.1, "mean_intent_rate": 0.58, "policy_loss": 0.014256, "value_loss": 122.67465, "entropy": 0.412856, "total_loss": 61.347452, "clip_frac": 0.175049, "sps": 426.5, "elapsed_s": 1510.8} |
| {"update": 322, "steps_done": 659456, "mean_step_reward": 0.9702, "min_step_reward": -7.5872, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 193.303, "mean_ep_length": 200.0, "mean_collisions": 2.1, "mean_intent_rate": 0.534, "policy_loss": -1.1e-05, "value_loss": 116.064726, "entropy": 0.39134, "total_loss": 58.028439, "clip_frac": 0.14209, "sps": 440.1, "elapsed_s": 1515.4} |
| {"update": 323, "steps_done": 661504, "mean_step_reward": 0.9031, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 182.531, "mean_ep_length": 200.0, "mean_collisions": 0.91, "mean_intent_rate": 0.495, "policy_loss": 0.01592, "value_loss": 109.965951, "entropy": 0.435911, "total_loss": 54.994537, "clip_frac": 0.195557, "sps": 437.6, "elapsed_s": 1520.1} |
| {"update": 324, "steps_done": 663552, "mean_step_reward": 0.7983, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 157.492, "mean_ep_length": 200.0, "mean_collisions": 1.5, "mean_intent_rate": 0.464, "policy_loss": 0.004197, "value_loss": 89.957475, "entropy": 0.393954, "total_loss": 44.978996, "clip_frac": 0.10376, "sps": 439.6, "elapsed_s": 1524.8} |
| {"update": 325, "steps_done": 665600, "mean_step_reward": 0.8004, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 156.928, "mean_ep_length": 200.0, "mean_collisions": 1.9, "mean_intent_rate": 0.47, "policy_loss": 0.018635, "value_loss": 89.838698, "entropy": 0.387589, "total_loss": 44.934108, "clip_frac": 0.199341, "sps": 435.9, "elapsed_s": 1529.5} |
| {"update": 326, "steps_done": 667648, "mean_step_reward": 0.6395, "min_step_reward": -10.5401, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 141.874, "mean_ep_length": 200.0, "mean_collisions": 2.2, "mean_intent_rate": 0.452, "policy_loss": 0.003538, "value_loss": 85.90544, "entropy": 0.376772, "total_loss": 42.952491, "clip_frac": 0.078125, "sps": 439.0, "elapsed_s": 1534.1} |
| {"update": 327, "steps_done": 669696, "mean_step_reward": 0.8563, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 163.185, "mean_ep_length": 200.0, "mean_collisions": 1.36, "mean_intent_rate": 0.473, "policy_loss": 0.003808, "value_loss": 98.813687, "entropy": 0.403301, "total_loss": 49.406619, "clip_frac": 0.130371, "sps": 439.0, "elapsed_s": 1538.8} |
| {"update": 328, "steps_done": 671744, "mean_step_reward": 0.932, "min_step_reward": -10.5285, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 182.443, "mean_ep_length": 200.0, "mean_collisions": 2.8, "mean_intent_rate": 0.527, "policy_loss": 0.003027, "value_loss": 89.281274, "entropy": 0.433773, "total_loss": 44.639326, "clip_frac": 0.082642, "sps": 436.0, "elapsed_s": 1543.5} |
| {"update": 329, "steps_done": 673792, "mean_step_reward": 1.0604, "min_step_reward": -10.5401, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 217.339, "mean_ep_length": 200.0, "mean_collisions": 2.9, "mean_intent_rate": 0.588, "policy_loss": 0.002772, "value_loss": 104.03946, "entropy": 0.423295, "total_loss": 52.018268, "clip_frac": 0.101685, "sps": 428.9, "elapsed_s": 1548.3} |
| {"update": 330, "steps_done": 675840, "mean_step_reward": 0.8658, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 174.263, "mean_ep_length": 200.0, "mean_collisions": 6.4, "mean_intent_rate": 0.575, "policy_loss": 0.047956, "value_loss": 120.406631, "entropy": 0.420475, "total_loss": 60.247066, "clip_frac": 0.409668, "sps": 438.9, "elapsed_s": 1552.9} |
| {"update": 331, "steps_done": 677888, "mean_step_reward": 0.8701, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 170.193, "mean_ep_length": 200.0, "mean_collisions": 10.0, "mean_intent_rate": 0.63, "policy_loss": 0.004793, "value_loss": 122.50468, "entropy": 0.442121, "total_loss": 61.252711, "clip_frac": 0.144897, "sps": 437.7, "elapsed_s": 1557.6} |
| {"update": 332, "steps_done": 679936, "mean_step_reward": 1.017, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 203.026, "mean_ep_length": 200.0, "mean_collisions": 6.1, "mean_intent_rate": 0.617, "policy_loss": 0.018759, "value_loss": 115.442649, "entropy": 0.418874, "total_loss": 57.735896, "clip_frac": 0.20105, "sps": 439.2, "elapsed_s": 1562.3} |
| {"update": 333, "steps_done": 681984, "mean_step_reward": 0.996, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 199.064, "mean_ep_length": 200.0, "mean_collisions": 4.7, "mean_intent_rate": 0.588, "policy_loss": 0.004285, "value_loss": 109.876192, "entropy": 0.416037, "total_loss": 54.938221, "clip_frac": 0.124512, "sps": 436.9, "elapsed_s": 1567.0} |
| {"update": 334, "steps_done": 684032, "mean_step_reward": 1.005, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 197.605, "mean_ep_length": 200.0, "mean_collisions": 2.5, "mean_intent_rate": 0.547, "policy_loss": 0.003698, "value_loss": 107.712811, "entropy": 0.422886, "total_loss": 53.855875, "clip_frac": 0.078125, "sps": 438.7, "elapsed_s": 1571.6} |
| {"update": 335, "steps_done": 686080, "mean_step_reward": 0.9338, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 188.454, "mean_ep_length": 185.7, "mean_collisions": 2.0, "mean_intent_rate": 0.548, "policy_loss": 0.013905, "value_loss": 106.762042, "entropy": 0.389606, "total_loss": 53.391031, "clip_frac": 0.196533, "sps": 439.0, "elapsed_s": 1576.3} |
| {"update": 336, "steps_done": 688128, "mean_step_reward": 1.0406, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 198.024, "mean_ep_length": 200.0, "mean_collisions": 2.0, "mean_intent_rate": 0.54, "policy_loss": 0.004057, "value_loss": 109.728448, "entropy": 0.381843, "total_loss": 54.864463, "clip_frac": 0.148071, "sps": 443.2, "elapsed_s": 1580.9} |
| {"update": 337, "steps_done": 690176, "mean_step_reward": 1.0925, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 223.833, "mean_ep_length": 200.0, "mean_collisions": 4.2, "mean_intent_rate": 0.62, "policy_loss": 0.007204, "value_loss": 122.421379, "entropy": 0.433679, "total_loss": 61.213556, "clip_frac": 0.118164, "sps": 442.0, "elapsed_s": 1585.6} |
| {"update": 338, "steps_done": 692224, "mean_step_reward": 0.9635, "min_step_reward": -10.584, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 182.909, "mean_ep_length": 200.0, "mean_collisions": 9.1, "mean_intent_rate": 0.636, "policy_loss": 0.009407, "value_loss": 145.241934, "entropy": 0.416479, "total_loss": 72.62621, "clip_frac": 0.162964, "sps": 431.9, "elapsed_s": 1590.3} |
| {"update": 339, "steps_done": 694272, "mean_step_reward": 0.9418, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 214.714, "mean_ep_length": 200.0, "mean_collisions": 6.9, "mean_intent_rate": 0.65, "policy_loss": 0.022283, "value_loss": 159.35937, "entropy": 0.388415, "total_loss": 79.698084, "clip_frac": 0.259521, "sps": 441.0, "elapsed_s": 1594.9} |
| {"update": 340, "steps_done": 696320, "mean_step_reward": 0.9024, "min_step_reward": -10.5872, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 160.878, "mean_ep_length": 200.0, "mean_collisions": 11.45, "mean_intent_rate": 0.637, "policy_loss": 0.007775, "value_loss": 107.706784, "entropy": 0.389167, "total_loss": 53.857276, "clip_frac": 0.108398, "sps": 446.3, "elapsed_s": 1599.5} |
| {"update": 341, "steps_done": 698368, "mean_step_reward": 1.0757, "min_step_reward": -10.5349, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 216.446, "mean_ep_length": 200.0, "mean_collisions": 7.9, "mean_intent_rate": 0.671, "policy_loss": 0.000869, "value_loss": 122.714047, "entropy": 0.390389, "total_loss": 61.353988, "clip_frac": 0.100708, "sps": 437.4, "elapsed_s": 1604.2} |
| {"update": 342, "steps_done": 700416, "mean_step_reward": 1.1293, "min_step_reward": -10.5787, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 220.02, "mean_ep_length": 200.0, "mean_collisions": 6.7, "mean_intent_rate": 0.657, "policy_loss": 0.007131, "value_loss": 129.241344, "entropy": 0.374663, "total_loss": 64.624056, "clip_frac": 0.133301, "sps": 436.5, "elapsed_s": 1608.9} |
| {"update": 343, "steps_done": 702464, "mean_step_reward": 1.1201, "min_step_reward": -10.5334, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 227.695, "mean_ep_length": 200.0, "mean_collisions": 5.8, "mean_intent_rate": 0.654, "policy_loss": 0.015401, "value_loss": 136.14787, "entropy": 0.369969, "total_loss": 68.085637, "clip_frac": 0.183838, "sps": 437.8, "elapsed_s": 1613.6} |
| {"update": 344, "steps_done": 704512, "mean_step_reward": 1.1548, "min_step_reward": -10.529, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 230.606, "mean_ep_length": 200.0, "mean_collisions": 5.1, "mean_intent_rate": 0.646, "policy_loss": 0.008109, "value_loss": 117.718083, "entropy": 0.406094, "total_loss": 58.86309, "clip_frac": 0.151245, "sps": 437.0, "elapsed_s": 1618.3} |
| {"update": 345, "steps_done": 706560, "mean_step_reward": 1.0611, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 210.381, "mean_ep_length": 200.0, "mean_collisions": 5.09, "mean_intent_rate": 0.614, "policy_loss": 0.005656, "value_loss": 141.250235, "entropy": 0.38177, "total_loss": 70.626956, "clip_frac": 0.102783, "sps": 434.0, "elapsed_s": 1623.0} |
| {"update": 346, "steps_done": 708608, "mean_step_reward": 1.0013, "min_step_reward": -10.5798, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 204.167, "mean_ep_length": 200.0, "mean_collisions": 7.3, "mean_intent_rate": 0.641, "policy_loss": 0.006845, "value_loss": 166.398924, "entropy": 0.3608, "total_loss": 83.202698, "clip_frac": 0.144409, "sps": 445.8, "elapsed_s": 1627.6} |
| {"update": 347, "steps_done": 710656, "mean_step_reward": 1.0695, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 216.64, "mean_ep_length": 200.0, "mean_collisions": 8.2, "mean_intent_rate": 0.675, "policy_loss": 0.041306, "value_loss": 156.964024, "entropy": 0.352011, "total_loss": 78.519798, "clip_frac": 0.297485, "sps": 425.4, "elapsed_s": 1632.4} |
| {"update": 348, "steps_done": 712704, "mean_step_reward": 0.6745, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 136.514, "mean_ep_length": 200.0, "mean_collisions": 17.6, "mean_intent_rate": 0.702, "policy_loss": 0.017531, "value_loss": 180.852241, "entropy": 0.361951, "total_loss": 90.440031, "clip_frac": 0.181763, "sps": 446.5, "elapsed_s": 1637.0} |
| {"update": 349, "steps_done": 714752, "mean_step_reward": 0.3235, "min_step_reward": -10.5863, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 66.457, "mean_ep_length": 200.0, "mean_collisions": 23.36, "mean_intent_rate": 0.682, "policy_loss": 0.02173, "value_loss": 222.742671, "entropy": 0.416984, "total_loss": 111.388895, "clip_frac": 0.22937, "sps": 445.6, "elapsed_s": 1641.6} |
| {"update": 350, "steps_done": 716800, "mean_step_reward": 0.5866, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 113.783, "mean_ep_length": 200.0, "mean_collisions": 19.2, "mean_intent_rate": 0.691, "policy_loss": 0.011174, "value_loss": 163.550454, "entropy": 0.450362, "total_loss": 81.781897, "clip_frac": 0.176025, "sps": 450.5, "elapsed_s": 1646.1} |
| {"update": 351, "steps_done": 718848, "mean_step_reward": 0.9958, "min_step_reward": -10.5436, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 212.445, "mean_ep_length": 200.0, "mean_collisions": 8.5, "mean_intent_rate": 0.675, "policy_loss": 0.002471, "value_loss": 111.046599, "entropy": 0.466385, "total_loss": 55.521108, "clip_frac": 0.105835, "sps": 452.3, "elapsed_s": 1650.7} |
| {"update": 352, "steps_done": 720896, "mean_step_reward": 0.876, "min_step_reward": -10.5871, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 169.358, "mean_ep_length": 200.0, "mean_collisions": 9.4, "mean_intent_rate": 0.617, "policy_loss": 0.005887, "value_loss": 104.136088, "entropy": 0.464913, "total_loss": 52.069282, "clip_frac": 0.11792, "sps": 452.3, "elapsed_s": 1655.2} |
| {"update": 353, "steps_done": 722944, "mean_step_reward": 1.0341, "min_step_reward": -10.5483, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 198.784, "mean_ep_length": 196.5, "mean_collisions": 8.18, "mean_intent_rate": 0.655, "policy_loss": 0.00575, "value_loss": 117.281694, "entropy": 0.452757, "total_loss": 58.642069, "clip_frac": 0.14978, "sps": 442.9, "elapsed_s": 1659.8} |
| {"update": 354, "steps_done": 724992, "mean_step_reward": 1.0803, "min_step_reward": -10.5872, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 218.861, "mean_ep_length": 200.0, "mean_collisions": 3.9, "mean_intent_rate": 0.608, "policy_loss": 0.009639, "value_loss": 109.811141, "entropy": 0.391751, "total_loss": 54.911293, "clip_frac": 0.166016, "sps": 449.6, "elapsed_s": 1664.4} |
| {"update": 355, "steps_done": 727040, "mean_step_reward": 1.0717, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 205.54, "mean_ep_length": 200.0, "mean_collisions": 6.3, "mean_intent_rate": 0.624, "policy_loss": 0.007436, "value_loss": 106.10977, "entropy": 0.40685, "total_loss": 53.058253, "clip_frac": 0.127808, "sps": 447.7, "elapsed_s": 1669.0} |
| {"update": 356, "steps_done": 729088, "mean_step_reward": 1.0366, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 214.678, "mean_ep_length": 200.0, "mean_collisions": 3.64, "mean_intent_rate": 0.597, "policy_loss": 0.003371, "value_loss": 117.310928, "entropy": 0.435212, "total_loss": 58.654483, "clip_frac": 0.085083, "sps": 441.2, "elapsed_s": 1673.6} |
| {"update": 357, "steps_done": 731136, "mean_step_reward": 0.9736, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 190.397, "mean_ep_length": 200.0, "mean_collisions": 5.6, "mean_intent_rate": 0.588, "policy_loss": 0.007527, "value_loss": 122.696465, "entropy": 0.43304, "total_loss": 61.351429, "clip_frac": 0.159668, "sps": 440.7, "elapsed_s": 1678.2} |
| {"update": 358, "steps_done": 733184, "mean_step_reward": 1.0247, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 209.082, "mean_ep_length": 200.0, "mean_collisions": 6.2, "mean_intent_rate": 0.629, "policy_loss": 0.002859, "value_loss": 109.540006, "entropy": 0.391486, "total_loss": 54.768947, "clip_frac": 0.081299, "sps": 441.6, "elapsed_s": 1682.9} |
| {"update": 359, "steps_done": 735232, "mean_step_reward": 0.7975, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 164.197, "mean_ep_length": 200.0, "mean_collisions": 10.2, "mean_intent_rate": 0.621, "policy_loss": 0.008232, "value_loss": 125.742071, "entropy": 0.415177, "total_loss": 62.875116, "clip_frac": 0.191528, "sps": 434.5, "elapsed_s": 1687.6} |
| {"update": 360, "steps_done": 737280, "mean_step_reward": 0.898, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 174.794, "mean_ep_length": 200.0, "mean_collisions": 7.91, "mean_intent_rate": 0.601, "policy_loss": 0.005788, "value_loss": 107.175286, "entropy": 0.433181, "total_loss": 53.5891, "clip_frac": 0.127686, "sps": 440.0, "elapsed_s": 1692.2} |
| {"update": 361, "steps_done": 739328, "mean_step_reward": 0.9026, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 178.367, "mean_ep_length": 200.0, "mean_collisions": 6.1, "mean_intent_rate": 0.578, "policy_loss": 0.004707, "value_loss": 115.99981, "entropy": 0.413119, "total_loss": 58.000481, "clip_frac": 0.108032, "sps": 442.8, "elapsed_s": 1696.9} |
| {"update": 362, "steps_done": 741376, "mean_step_reward": 0.9931, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 200.418, "mean_ep_length": 200.0, "mean_collisions": 6.1, "mean_intent_rate": 0.613, "policy_loss": 0.000631, "value_loss": 129.160648, "entropy": 0.406147, "total_loss": 64.576893, "clip_frac": 0.100464, "sps": 440.9, "elapsed_s": 1701.5} |
| {"update": 363, "steps_done": 743424, "mean_step_reward": 1.1513, "min_step_reward": -10.5776, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 226.6, "mean_ep_length": 200.0, "mean_collisions": 4.9, "mean_intent_rate": 0.637, "policy_loss": 0.003728, "value_loss": 115.645747, "entropy": 0.386482, "total_loss": 57.822736, "clip_frac": 0.086304, "sps": 437.7, "elapsed_s": 1706.2} |
| {"update": 364, "steps_done": 745472, "mean_step_reward": 1.2618, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 261.516, "mean_ep_length": 200.0, "mean_collisions": 3.4, "mean_intent_rate": 0.669, "policy_loss": 0.002358, "value_loss": 128.314029, "entropy": 0.3749, "total_loss": 64.155623, "clip_frac": 0.087158, "sps": 442.3, "elapsed_s": 1710.8} |
| {"update": 365, "steps_done": 747520, "mean_step_reward": 1.1032, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 214.194, "mean_ep_length": 200.0, "mean_collisions": 8.36, "mean_intent_rate": 0.674, "policy_loss": 0.03357, "value_loss": 135.905742, "entropy": 0.34071, "total_loss": 67.983034, "clip_frac": 0.267212, "sps": 428.6, "elapsed_s": 1715.6} |
| {"update": 366, "steps_done": 749568, "mean_step_reward": 0.0203, "min_step_reward": -10.5513, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": -0.586, "mean_ep_length": 200.0, "mean_collisions": 27.4, "mean_intent_rate": 0.637, "policy_loss": 0.010785, "value_loss": 251.237031, "entropy": 0.438384, "total_loss": 125.624916, "clip_frac": 0.282349, "sps": 437.6, "elapsed_s": 1720.3} |
| {"update": 367, "steps_done": 751616, "mean_step_reward": 0.795, "min_step_reward": -10.5757, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 158.944, "mean_ep_length": 200.0, "mean_collisions": 13.2, "mean_intent_rate": 0.664, "policy_loss": 0.010641, "value_loss": 137.000401, "entropy": 0.38734, "total_loss": 68.506969, "clip_frac": 0.154663, "sps": 438.6, "elapsed_s": 1725.0} |
| {"update": 368, "steps_done": 753664, "mean_step_reward": 1.0697, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 205.488, "mean_ep_length": 200.0, "mean_collisions": 9.3, "mean_intent_rate": 0.674, "policy_loss": 0.009436, "value_loss": 129.921981, "entropy": 0.354609, "total_loss": 64.966881, "clip_frac": 0.143921, "sps": 441.4, "elapsed_s": 1729.6} |
| {"update": 369, "steps_done": 755712, "mean_step_reward": 0.6132, "min_step_reward": -10.5839, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 139.723, "mean_ep_length": 200.0, "mean_collisions": 17.36, "mean_intent_rate": 0.7, "policy_loss": 0.006456, "value_loss": 177.931467, "entropy": 0.364129, "total_loss": 88.968548, "clip_frac": 0.138306, "sps": 437.4, "elapsed_s": 1734.3} |
| {"update": 370, "steps_done": 757760, "mean_step_reward": 0.944, "min_step_reward": -10.558, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 175.043, "mean_ep_length": 186.3, "mean_collisions": 11.09, "mean_intent_rate": 0.689, "policy_loss": 0.008716, "value_loss": 103.202466, "entropy": 0.370385, "total_loss": 51.606246, "clip_frac": 0.129883, "sps": 434.7, "elapsed_s": 1739.0} |
| {"update": 371, "steps_done": 759808, "mean_step_reward": 0.7807, "min_step_reward": -10.5853, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 151.396, "mean_ep_length": 200.0, "mean_collisions": 13.0, "mean_intent_rate": 0.647, "policy_loss": 0.000232, "value_loss": 122.35761, "entropy": 0.37054, "total_loss": 61.175331, "clip_frac": 0.040649, "sps": 438.9, "elapsed_s": 1743.7} |
| {"update": 372, "steps_done": 761856, "mean_step_reward": 0.9423, "min_step_reward": -10.5862, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 187.113, "mean_ep_length": 200.0, "mean_collisions": 10.7, "mean_intent_rate": 0.668, "policy_loss": 0.00694, "value_loss": 115.953343, "entropy": 0.370811, "total_loss": 57.979903, "clip_frac": 0.105591, "sps": 447.3, "elapsed_s": 1748.2} |
| {"update": 373, "steps_done": 763904, "mean_step_reward": 1.0332, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 203.424, "mean_ep_length": 194.4, "mean_collisions": 8.09, "mean_intent_rate": 0.663, "policy_loss": 0.008095, "value_loss": 99.242667, "entropy": 0.380642, "total_loss": 49.625622, "clip_frac": 0.127808, "sps": 442.2, "elapsed_s": 1752.9} |
| {"update": 374, "steps_done": 765952, "mean_step_reward": 0.992, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 203.159, "mean_ep_length": 200.0, "mean_collisions": 9.1, "mean_intent_rate": 0.669, "policy_loss": 0.00982, "value_loss": 115.453876, "entropy": 0.364782, "total_loss": 57.73311, "clip_frac": 0.136597, "sps": 444.4, "elapsed_s": 1757.5} |
| {"update": 375, "steps_done": 768000, "mean_step_reward": 0.8778, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 173.578, "mean_ep_length": 200.0, "mean_collisions": 13.0, "mean_intent_rate": 0.683, "policy_loss": 0.004022, "value_loss": 133.671948, "entropy": 0.374121, "total_loss": 66.836254, "clip_frac": 0.133179, "sps": 439.2, "elapsed_s": 1762.1} |
| {"update": 376, "steps_done": 770048, "mean_step_reward": 0.4254, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 85.289, "mean_ep_length": 200.0, "mean_collisions": 21.8, "mean_intent_rate": 0.685, "policy_loss": 0.007631, "value_loss": 192.961325, "entropy": 0.387458, "total_loss": 96.484419, "clip_frac": 0.160767, "sps": 444.9, "elapsed_s": 1766.8} |
| {"update": 377, "steps_done": 772096, "mean_step_reward": 0.8226, "min_step_reward": -10.5139, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 163.546, "mean_ep_length": 200.0, "mean_collisions": 11.82, "mean_intent_rate": 0.648, "policy_loss": 0.01032, "value_loss": 137.092765, "entropy": 0.379569, "total_loss": 68.552907, "clip_frac": 0.133911, "sps": 442.1, "elapsed_s": 1771.4} |
| {"update": 378, "steps_done": 774144, "mean_step_reward": 0.8552, "min_step_reward": -10.5784, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 167.19, "mean_ep_length": 200.0, "mean_collisions": 11.7, "mean_intent_rate": 0.654, "policy_loss": 0.002164, "value_loss": 109.223775, "entropy": 0.377678, "total_loss": 54.610274, "clip_frac": 0.071411, "sps": 447.1, "elapsed_s": 1776.0} |
| {"update": 379, "steps_done": 776192, "mean_step_reward": 0.891, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 173.87, "mean_ep_length": 200.0, "mean_collisions": 11.0, "mean_intent_rate": 0.651, "policy_loss": 0.00299, "value_loss": 114.807254, "entropy": 0.358249, "total_loss": 57.403035, "clip_frac": 0.133545, "sps": 446.2, "elapsed_s": 1780.6} |
| {"update": 380, "steps_done": 778240, "mean_step_reward": 0.9737, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 202.498, "mean_ep_length": 200.0, "mean_collisions": 10.3, "mean_intent_rate": 0.687, "policy_loss": 0.014734, "value_loss": 128.323453, "entropy": 0.362323, "total_loss": 64.172837, "clip_frac": 0.202026, "sps": 448.0, "elapsed_s": 1785.1} |
| {"update": 381, "steps_done": 780288, "mean_step_reward": 1.1097, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 217.877, "mean_ep_length": 200.0, "mean_collisions": 6.91, "mean_intent_rate": 0.656, "policy_loss": 0.003281, "value_loss": 143.205479, "entropy": 0.376555, "total_loss": 71.602255, "clip_frac": 0.101929, "sps": 454.2, "elapsed_s": 1789.6} |
| {"update": 382, "steps_done": 782336, "mean_step_reward": 0.9963, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 200.295, "mean_ep_length": 200.0, "mean_collisions": 10.1, "mean_intent_rate": 0.68, "policy_loss": 0.006026, "value_loss": 147.965193, "entropy": 0.377965, "total_loss": 73.984842, "clip_frac": 0.113647, "sps": 450.2, "elapsed_s": 1794.2} |
| {"update": 383, "steps_done": 784384, "mean_step_reward": 1.0453, "min_step_reward": -10.5507, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 211.82, "mean_ep_length": 200.0, "mean_collisions": 6.6, "mean_intent_rate": 0.641, "policy_loss": 0.0091, "value_loss": 128.247993, "entropy": 0.369318, "total_loss": 64.129403, "clip_frac": 0.102783, "sps": 445.2, "elapsed_s": 1798.8} |
| {"update": 384, "steps_done": 786432, "mean_step_reward": 1.1121, "min_step_reward": -10.535, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 206.333, "mean_ep_length": 200.0, "mean_collisions": 8.5, "mean_intent_rate": 0.664, "policy_loss": 0.013659, "value_loss": 141.298262, "entropy": 0.360906, "total_loss": 70.659181, "clip_frac": 0.163452, "sps": 440.1, "elapsed_s": 1803.4} |
| {"update": 385, "steps_done": 788480, "mean_step_reward": 1.0401, "min_step_reward": -10.5835, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 219.228, "mean_ep_length": 200.0, "mean_collisions": 7.36, "mean_intent_rate": 0.665, "policy_loss": 0.006725, "value_loss": 148.498594, "entropy": 0.403177, "total_loss": 74.25199, "clip_frac": 0.144043, "sps": 441.6, "elapsed_s": 1808.1} |
| {"update": 386, "steps_done": 790528, "mean_step_reward": 1.125, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 221.555, "mean_ep_length": 200.0, "mean_collisions": 4.6, "mean_intent_rate": 0.623, "policy_loss": 0.002879, "value_loss": 122.831347, "entropy": 0.370736, "total_loss": 61.414845, "clip_frac": 0.101562, "sps": 442.0, "elapsed_s": 1812.7} |
| {"update": 387, "steps_done": 792576, "mean_step_reward": 1.0268, "min_step_reward": -10.5447, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 204.902, "mean_ep_length": 200.0, "mean_collisions": 7.4, "mean_intent_rate": 0.642, "policy_loss": 0.006963, "value_loss": 131.300326, "entropy": 0.380161, "total_loss": 65.653325, "clip_frac": 0.120605, "sps": 437.2, "elapsed_s": 1817.4} |
| {"update": 388, "steps_done": 794624, "mean_step_reward": 1.0812, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 218.43, "mean_ep_length": 200.0, "mean_collisions": 4.3, "mean_intent_rate": 0.614, "policy_loss": 0.004506, "value_loss": 107.009345, "entropy": 0.401559, "total_loss": 53.505163, "clip_frac": 0.099121, "sps": 437.7, "elapsed_s": 1822.1} |
| {"update": 389, "steps_done": 796672, "mean_step_reward": 0.9884, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 199.844, "mean_ep_length": 200.0, "mean_collisions": 2.55, "mean_intent_rate": 0.553, "policy_loss": 0.004784, "value_loss": 108.277246, "entropy": 0.426421, "total_loss": 54.139143, "clip_frac": 0.174683, "sps": 439.9, "elapsed_s": 1826.7} |
| {"update": 390, "steps_done": 798720, "mean_step_reward": 0.9887, "min_step_reward": -7.587, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 196.525, "mean_ep_length": 200.0, "mean_collisions": 2.8, "mean_intent_rate": 0.551, "policy_loss": 0.010163, "value_loss": 104.368646, "entropy": 0.403128, "total_loss": 52.190455, "clip_frac": 0.177246, "sps": 437.4, "elapsed_s": 1831.4} |
| {"update": 391, "steps_done": 800768, "mean_step_reward": 0.8938, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 180.399, "mean_ep_length": 200.0, "mean_collisions": 5.6, "mean_intent_rate": 0.572, "policy_loss": 0.000187, "value_loss": 119.799614, "entropy": 0.381336, "total_loss": 59.89618, "clip_frac": 0.075439, "sps": 437.0, "elapsed_s": 1836.1} |
| {"update": 392, "steps_done": 802816, "mean_step_reward": 0.9716, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 196.439, "mean_ep_length": 200.0, "mean_collisions": 6.7, "mean_intent_rate": 0.618, "policy_loss": 0.001332, "value_loss": 134.06277, "entropy": 0.368055, "total_loss": 67.029036, "clip_frac": 0.072144, "sps": 430.7, "elapsed_s": 1840.9} |
| {"update": 393, "steps_done": 804864, "mean_step_reward": 1.0515, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 206.334, "mean_ep_length": 200.0, "mean_collisions": 4.91, "mean_intent_rate": 0.603, "policy_loss": 0.006116, "value_loss": 139.233097, "entropy": 0.373013, "total_loss": 69.618935, "clip_frac": 0.135132, "sps": 438.6, "elapsed_s": 1845.5} |
| {"update": 394, "steps_done": 806912, "mean_step_reward": 1.0369, "min_step_reward": -10.5452, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 207.281, "mean_ep_length": 200.0, "mean_collisions": 4.5, "mean_intent_rate": 0.597, "policy_loss": 0.002081, "value_loss": 112.590326, "entropy": 0.377588, "total_loss": 56.293468, "clip_frac": 0.084351, "sps": 434.8, "elapsed_s": 1850.2} |
| {"update": 395, "steps_done": 808960, "mean_step_reward": 1.0789, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 213.9, "mean_ep_length": 200.0, "mean_collisions": 3.7, "mean_intent_rate": 0.594, "policy_loss": 0.003073, "value_loss": 114.041626, "entropy": 0.380988, "total_loss": 57.020077, "clip_frac": 0.095581, "sps": 442.6, "elapsed_s": 1854.9} |
| {"update": 396, "steps_done": 811008, "mean_step_reward": 1.1333, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 231.288, "mean_ep_length": 200.0, "mean_collisions": 3.7, "mean_intent_rate": 0.625, "policy_loss": 0.005888, "value_loss": 141.024977, "entropy": 0.369929, "total_loss": 70.514678, "clip_frac": 0.103027, "sps": 433.1, "elapsed_s": 1859.6} |
| {"update": 397, "steps_done": 813056, "mean_step_reward": 1.1688, "min_step_reward": -7.5861, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 222.168, "mean_ep_length": 200.0, "mean_collisions": 4.7, "mean_intent_rate": 0.625, "policy_loss": 0.024808, "value_loss": 131.044296, "entropy": 0.350087, "total_loss": 65.543455, "clip_frac": 0.270386, "sps": 436.8, "elapsed_s": 1864.3} |
| {"update": 398, "steps_done": 815104, "mean_step_reward": 1.1004, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 231.572, "mean_ep_length": 200.0, "mean_collisions": 5.45, "mean_intent_rate": 0.655, "policy_loss": 0.002924, "value_loss": 137.709877, "entropy": 0.353728, "total_loss": 68.854326, "clip_frac": 0.123779, "sps": 441.2, "elapsed_s": 1868.9} |
| {"update": 399, "steps_done": 817152, "mean_step_reward": 0.9469, "min_step_reward": -10.5572, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 186.358, "mean_ep_length": 200.0, "mean_collisions": 10.2, "mean_intent_rate": 0.66, "policy_loss": 0.01113, "value_loss": 142.539528, "entropy": 0.334771, "total_loss": 71.277546, "clip_frac": 0.140015, "sps": 440.6, "elapsed_s": 1873.6} |
| {"update": 400, "steps_done": 819200, "mean_step_reward": 0.8568, "min_step_reward": -10.5192, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 171.053, "mean_ep_length": 200.0, "mean_collisions": 12.3, "mean_intent_rate": 0.669, "policy_loss": 0.005769, "value_loss": 159.877854, "entropy": 0.35343, "total_loss": 79.941162, "clip_frac": 0.123169, "sps": 426.1, "elapsed_s": 1878.4} |
| {"update": 401, "steps_done": 821248, "mean_step_reward": 1.1917, "min_step_reward": -10.5009, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 227.813, "mean_ep_length": 200.0, "mean_collisions": 6.0, "mean_intent_rate": 0.657, "policy_loss": 0.006005, "value_loss": 134.209599, "entropy": 0.349237, "total_loss": 67.107312, "clip_frac": 0.107666, "sps": 437.2, "elapsed_s": 1883.1} |
| {"update": 402, "steps_done": 823296, "mean_step_reward": 1.069, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 221.886, "mean_ep_length": 200.0, "mean_collisions": 3.91, "mean_intent_rate": 0.613, "policy_loss": 0.004968, "value_loss": 132.769326, "entropy": 0.353471, "total_loss": 66.386097, "clip_frac": 0.127197, "sps": 439.3, "elapsed_s": 1887.8} |
| {"update": 403, "steps_done": 825344, "mean_step_reward": 1.1039, "min_step_reward": -10.519, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 218.498, "mean_ep_length": 200.0, "mean_collisions": 7.5, "mean_intent_rate": 0.666, "policy_loss": 0.020507, "value_loss": 149.143636, "entropy": 0.342579, "total_loss": 74.588899, "clip_frac": 0.182007, "sps": 438.8, "elapsed_s": 1892.4} |
| {"update": 404, "steps_done": 827392, "mean_step_reward": 0.8601, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 173.261, "mean_ep_length": 200.0, "mean_collisions": 11.4, "mean_intent_rate": 0.658, "policy_loss": 0.010221, "value_loss": 134.857337, "entropy": 0.363824, "total_loss": 67.435251, "clip_frac": 0.180664, "sps": 436.9, "elapsed_s": 1897.1} |
| {"update": 405, "steps_done": 829440, "mean_step_reward": 1.0592, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 215.839, "mean_ep_length": 200.0, "mean_collisions": 8.6, "mean_intent_rate": 0.682, "policy_loss": 0.005625, "value_loss": 134.250203, "entropy": 0.35, "total_loss": 67.127228, "clip_frac": 0.131348, "sps": 437.2, "elapsed_s": 1901.8} |
| {"update": 406, "steps_done": 831488, "mean_step_reward": 0.9345, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 186.837, "mean_ep_length": 200.0, "mean_collisions": 11.18, "mean_intent_rate": 0.675, "policy_loss": 0.009219, "value_loss": 163.238849, "entropy": 0.345983, "total_loss": 81.625183, "clip_frac": 0.114258, "sps": 436.9, "elapsed_s": 1906.5} |
| {"update": 407, "steps_done": 833536, "mean_step_reward": 0.8588, "min_step_reward": -10.5545, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 172.059, "mean_ep_length": 200.0, "mean_collisions": 12.5, "mean_intent_rate": 0.673, "policy_loss": 0.014482, "value_loss": 144.74973, "entropy": 0.369744, "total_loss": 72.385649, "clip_frac": 0.181396, "sps": 441.7, "elapsed_s": 1911.1} |
| {"update": 408, "steps_done": 835584, "mean_step_reward": 1.0483, "min_step_reward": -10.5437, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 212.281, "mean_ep_length": 200.0, "mean_collisions": 9.3, "mean_intent_rate": 0.687, "policy_loss": 0.012048, "value_loss": 132.972221, "entropy": 0.344115, "total_loss": 66.494717, "clip_frac": 0.140015, "sps": 442.2, "elapsed_s": 1915.7} |
| {"update": 409, "steps_done": 837632, "mean_step_reward": 0.6945, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 140.283, "mean_ep_length": 200.0, "mean_collisions": 15.9, "mean_intent_rate": 0.678, "policy_loss": 0.019383, "value_loss": 176.050024, "entropy": 0.350374, "total_loss": 88.040891, "clip_frac": 0.2229, "sps": 429.6, "elapsed_s": 1920.5} |
| {"update": 410, "steps_done": 839680, "mean_step_reward": 1.0196, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 195.373, "mean_ep_length": 200.0, "mean_collisions": 13.18, "mean_intent_rate": 0.725, "policy_loss": 0.00568, "value_loss": 153.785106, "entropy": 0.327479, "total_loss": 76.894958, "clip_frac": 0.088135, "sps": 439.3, "elapsed_s": 1925.2} |
| {"update": 411, "steps_done": 841728, "mean_step_reward": 0.793, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 160.575, "mean_ep_length": 200.0, "mean_collisions": 15.8, "mean_intent_rate": 0.71, "policy_loss": 0.006899, "value_loss": 160.550207, "entropy": 0.324189, "total_loss": 80.278761, "clip_frac": 0.089966, "sps": 441.2, "elapsed_s": 1929.8} |
| {"update": 412, "steps_done": 843776, "mean_step_reward": 0.4449, "min_step_reward": -10.5877, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 86.292, "mean_ep_length": 200.0, "mean_collisions": 22.4, "mean_intent_rate": 0.697, "policy_loss": 0.008666, "value_loss": 203.856944, "entropy": 0.37133, "total_loss": 101.933425, "clip_frac": 0.157349, "sps": 439.4, "elapsed_s": 1934.5} |
| {"update": 413, "steps_done": 845824, "mean_step_reward": 0.6784, "min_step_reward": -10.5749, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 129.636, "mean_ep_length": 190.9, "mean_collisions": 17.09, "mean_intent_rate": 0.7, "policy_loss": 0.015135, "value_loss": 176.287967, "entropy": 0.367615, "total_loss": 88.155443, "clip_frac": 0.169678, "sps": 437.0, "elapsed_s": 1939.2} |
| {"update": 414, "steps_done": 847872, "mean_step_reward": 0.8526, "min_step_reward": -10.5628, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 171.19, "mean_ep_length": 197.4, "mean_collisions": 11.2, "mean_intent_rate": 0.658, "policy_loss": 0.009896, "value_loss": 131.052781, "entropy": 0.373147, "total_loss": 65.532556, "clip_frac": 0.119995, "sps": 436.2, "elapsed_s": 1943.9} |
| {"update": 415, "steps_done": 849920, "mean_step_reward": 0.8938, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 171.062, "mean_ep_length": 200.0, "mean_collisions": 12.8, "mean_intent_rate": 0.68, "policy_loss": 0.004894, "value_loss": 107.556865, "entropy": 0.360031, "total_loss": 53.779726, "clip_frac": 0.127441, "sps": 440.2, "elapsed_s": 1948.5} |
| {"update": 416, "steps_done": 851968, "mean_step_reward": 0.8958, "min_step_reward": -10.575, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 183.926, "mean_ep_length": 200.0, "mean_collisions": 12.09, "mean_intent_rate": 0.688, "policy_loss": -0.00031, "value_loss": 138.195315, "entropy": 0.364711, "total_loss": 69.093701, "clip_frac": 0.124023, "sps": 435.4, "elapsed_s": 1953.2} |
| {"update": 417, "steps_done": 854016, "mean_step_reward": 0.882, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 175.328, "mean_ep_length": 200.0, "mean_collisions": 13.3, "mean_intent_rate": 0.692, "policy_loss": 0.001866, "value_loss": 158.580045, "entropy": 0.340177, "total_loss": 79.288486, "clip_frac": 0.083984, "sps": 440.6, "elapsed_s": 1957.9} |
| {"update": 418, "steps_done": 856064, "mean_step_reward": 0.8056, "min_step_reward": -10.5499, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 158.165, "mean_ep_length": 200.0, "mean_collisions": 13.4, "mean_intent_rate": 0.668, "policy_loss": 0.002399, "value_loss": 164.389309, "entropy": 0.38049, "total_loss": 82.193249, "clip_frac": 0.100464, "sps": 424.7, "elapsed_s": 1962.7} |
| {"update": 419, "steps_done": 858112, "mean_step_reward": 0.8357, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 163.038, "mean_ep_length": 200.0, "mean_collisions": 15.9, "mean_intent_rate": 0.716, "policy_loss": 0.002874, "value_loss": 148.225999, "entropy": 0.366536, "total_loss": 74.112207, "clip_frac": 0.122559, "sps": 434.6, "elapsed_s": 1967.4} |
| {"update": 420, "steps_done": 860160, "mean_step_reward": 1.1906, "min_step_reward": -10.5855, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 239.45, "mean_ep_length": 200.0, "mean_collisions": 5.64, "mean_intent_rate": 0.672, "policy_loss": 0.009068, "value_loss": 122.939826, "entropy": 0.3509, "total_loss": 61.475472, "clip_frac": 0.134766, "sps": 440.2, "elapsed_s": 1972.1} |
| {"update": 421, "steps_done": 862208, "mean_step_reward": 0.9878, "min_step_reward": -10.559, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 195.869, "mean_ep_length": 200.0, "mean_collisions": 9.0, "mean_intent_rate": 0.654, "policy_loss": 0.008355, "value_loss": 160.578096, "entropy": 0.358602, "total_loss": 80.293817, "clip_frac": 0.153564, "sps": 447.6, "elapsed_s": 1976.6} |
| {"update": 422, "steps_done": 864256, "mean_step_reward": 0.6498, "min_step_reward": -10.5207, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 145.713, "mean_ep_length": 198.7, "mean_collisions": 14.1, "mean_intent_rate": 0.661, "policy_loss": -0.001931, "value_loss": 143.38786, "entropy": 0.378008, "total_loss": 71.688219, "clip_frac": 0.097534, "sps": 440.8, "elapsed_s": 1981.3} |
| {"update": 423, "steps_done": 866304, "mean_step_reward": 0.8265, "min_step_reward": -10.5316, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 148.472, "mean_ep_length": 200.0, "mean_collisions": 14.6, "mean_intent_rate": 0.671, "policy_loss": 0.010645, "value_loss": 133.489578, "entropy": 0.357012, "total_loss": 66.751863, "clip_frac": 0.159668, "sps": 446.7, "elapsed_s": 1985.9} |
| {"update": 424, "steps_done": 868352, "mean_step_reward": 1.0425, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 207.697, "mean_ep_length": 200.0, "mean_collisions": 10.0, "mean_intent_rate": 0.694, "policy_loss": 0.009136, "value_loss": 130.320457, "entropy": 0.346232, "total_loss": 65.165902, "clip_frac": 0.137451, "sps": 436.3, "elapsed_s": 1990.6} |
| {"update": 425, "steps_done": 870400, "mean_step_reward": 0.8233, "min_step_reward": -10.535, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 163.278, "mean_ep_length": 200.0, "mean_collisions": 13.8, "mean_intent_rate": 0.681, "policy_loss": -0.001778, "value_loss": 160.763688, "entropy": 0.365313, "total_loss": 80.376412, "clip_frac": 0.069458, "sps": 437.7, "elapsed_s": 1995.2} |
| {"update": 426, "steps_done": 872448, "mean_step_reward": 0.9047, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 178.286, "mean_ep_length": 200.0, "mean_collisions": 12.2, "mean_intent_rate": 0.678, "policy_loss": 0.024193, "value_loss": 123.979293, "entropy": 0.386407, "total_loss": 62.009975, "clip_frac": 0.192139, "sps": 434.1, "elapsed_s": 2000.0} |
| {"update": 427, "steps_done": 874496, "mean_step_reward": 0.6613, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 138.215, "mean_ep_length": 200.0, "mean_collisions": 16.0, "mean_intent_rate": 0.676, "policy_loss": 0.004845, "value_loss": 122.213252, "entropy": 0.388632, "total_loss": 61.107585, "clip_frac": 0.173462, "sps": 429.6, "elapsed_s": 2004.7} |
| {"update": 428, "steps_done": 876544, "mean_step_reward": 0.8652, "min_step_reward": -10.5685, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 172.826, "mean_ep_length": 200.0, "mean_collisions": 11.55, "mean_intent_rate": 0.66, "policy_loss": 0.006664, "value_loss": 122.428907, "entropy": 0.386023, "total_loss": 61.217257, "clip_frac": 0.099854, "sps": 434.6, "elapsed_s": 2009.4} |
| {"update": 429, "steps_done": 878592, "mean_step_reward": 0.9433, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 199.458, "mean_ep_length": 200.0, "mean_collisions": 10.4, "mean_intent_rate": 0.685, "policy_loss": 0.00711, "value_loss": 122.489396, "entropy": 0.362762, "total_loss": 61.24818, "clip_frac": 0.099121, "sps": 436.8, "elapsed_s": 2014.1} |
| {"update": 430, "steps_done": 880640, "mean_step_reward": 0.7436, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 146.421, "mean_ep_length": 200.0, "mean_collisions": 16.3, "mean_intent_rate": 0.696, "policy_loss": 0.001836, "value_loss": 159.680895, "entropy": 0.358616, "total_loss": 79.838697, "clip_frac": 0.114868, "sps": 438.1, "elapsed_s": 2018.8} |
| {"update": 431, "steps_done": 882688, "mean_step_reward": 0.699, "min_step_reward": -10.5597, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 131.742, "mean_ep_length": 200.0, "mean_collisions": 14.5, "mean_intent_rate": 0.641, "policy_loss": 0.025497, "value_loss": 139.949809, "entropy": 0.374911, "total_loss": 69.996652, "clip_frac": 0.253662, "sps": 437.9, "elapsed_s": 2023.5} |
| {"update": 432, "steps_done": 884736, "mean_step_reward": 0.4369, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 90.959, "mean_ep_length": 200.0, "mean_collisions": 21.73, "mean_intent_rate": 0.694, "policy_loss": 0.013195, "value_loss": 161.637036, "entropy": 0.41301, "total_loss": 80.827583, "clip_frac": 0.195435, "sps": 435.3, "elapsed_s": 2028.2} |
| {"update": 433, "steps_done": 886784, "mean_step_reward": 0.6013, "min_step_reward": -10.5795, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 117.591, "mean_ep_length": 200.0, "mean_collisions": 17.6, "mean_intent_rate": 0.671, "policy_loss": 0.003448, "value_loss": 144.219857, "entropy": 0.384702, "total_loss": 72.109529, "clip_frac": 0.073486, "sps": 437.8, "elapsed_s": 2032.9} |
| {"update": 434, "steps_done": 888832, "mean_step_reward": 0.6413, "min_step_reward": -10.557, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 121.451, "mean_ep_length": 196.9, "mean_collisions": 15.8, "mean_intent_rate": 0.653, "policy_loss": 0.008401, "value_loss": 125.542158, "entropy": 0.376783, "total_loss": 62.775712, "clip_frac": 0.117554, "sps": 436.0, "elapsed_s": 2037.6} |
| {"update": 435, "steps_done": 890880, "mean_step_reward": 0.688, "min_step_reward": -10.5802, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 135.677, "mean_ep_length": 200.0, "mean_collisions": 15.1, "mean_intent_rate": 0.66, "policy_loss": 0.008319, "value_loss": 127.068195, "entropy": 0.371769, "total_loss": 63.5387, "clip_frac": 0.107544, "sps": 436.7, "elapsed_s": 2042.3} |
| {"update": 436, "steps_done": 892928, "mean_step_reward": 0.4136, "min_step_reward": -10.583, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 99.011, "mean_ep_length": 200.0, "mean_collisions": 21.64, "mean_intent_rate": 0.706, "policy_loss": 0.004255, "value_loss": 192.520502, "entropy": 0.353113, "total_loss": 96.260974, "clip_frac": 0.101196, "sps": 424.4, "elapsed_s": 2047.1} |
| {"update": 437, "steps_done": 894976, "mean_step_reward": 0.5884, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 108.801, "mean_ep_length": 200.0, "mean_collisions": 21.4, "mean_intent_rate": 0.718, "policy_loss": 0.011123, "value_loss": 178.97475, "entropy": 0.354761, "total_loss": 89.494951, "clip_frac": 0.226562, "sps": 444.1, "elapsed_s": 2051.7} |
| {"update": 438, "steps_done": 897024, "mean_step_reward": 0.2681, "min_step_reward": -10.5829, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 57.959, "mean_ep_length": 200.0, "mean_collisions": 28.3, "mean_intent_rate": 0.75, "policy_loss": 0.005546, "value_loss": 199.037577, "entropy": 0.352519, "total_loss": 99.52081, "clip_frac": 0.121704, "sps": 443.6, "elapsed_s": 2056.3} |
| {"update": 439, "steps_done": 899072, "mean_step_reward": 0.5103, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 113.359, "mean_ep_length": 200.0, "mean_collisions": 19.8, "mean_intent_rate": 0.7, "policy_loss": 0.002257, "value_loss": 178.157242, "entropy": 0.397445, "total_loss": 89.076904, "clip_frac": 0.078735, "sps": 441.2, "elapsed_s": 2061.0} |
| {"update": 440, "steps_done": 901120, "mean_step_reward": 0.9026, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 162.609, "mean_ep_length": 200.0, "mean_collisions": 15.55, "mean_intent_rate": 0.71, "policy_loss": 0.005177, "value_loss": 110.937686, "entropy": 0.367705, "total_loss": 55.470343, "clip_frac": 0.122925, "sps": 433.9, "elapsed_s": 2065.7} |
| {"update": 441, "steps_done": 903168, "mean_step_reward": 0.656, "min_step_reward": -10.5837, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 133.742, "mean_ep_length": 200.0, "mean_collisions": 17.1, "mean_intent_rate": 0.686, "policy_loss": 0.004621, "value_loss": 118.374881, "entropy": 0.35267, "total_loss": 59.188536, "clip_frac": 0.101562, "sps": 437.8, "elapsed_s": 2070.4} |
| {"update": 442, "steps_done": 905216, "mean_step_reward": 0.7342, "min_step_reward": -10.5399, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 142.95, "mean_ep_length": 200.0, "mean_collisions": 17.2, "mean_intent_rate": 0.706, "policy_loss": 0.019637, "value_loss": 178.602437, "entropy": 0.380688, "total_loss": 89.317049, "clip_frac": 0.123047, "sps": 438.8, "elapsed_s": 2075.0} |
| {"update": 443, "steps_done": 907264, "mean_step_reward": 1.0363, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 207.019, "mean_ep_length": 200.0, "mean_collisions": 9.8, "mean_intent_rate": 0.688, "policy_loss": 0.029705, "value_loss": 113.09721, "entropy": 0.345144, "total_loss": 56.574858, "clip_frac": 0.262817, "sps": 439.7, "elapsed_s": 2079.7} |
| {"update": 444, "steps_done": 909312, "mean_step_reward": 0.6377, "min_step_reward": -10.5859, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 131.283, "mean_ep_length": 200.0, "mean_collisions": 15.91, "mean_intent_rate": 0.664, "policy_loss": 0.011409, "value_loss": 156.955523, "entropy": 0.358871, "total_loss": 78.485583, "clip_frac": 0.129883, "sps": 445.6, "elapsed_s": 2084.3} |
| {"update": 445, "steps_done": 911360, "mean_step_reward": 0.9324, "min_step_reward": -10.5189, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 188.111, "mean_ep_length": 200.0, "mean_collisions": 11.0, "mean_intent_rate": 0.676, "policy_loss": 0.007996, "value_loss": 112.996325, "entropy": 0.347754, "total_loss": 56.502681, "clip_frac": 0.127808, "sps": 434.2, "elapsed_s": 2089.0} |
| {"update": 446, "steps_done": 913408, "mean_step_reward": 0.923, "min_step_reward": -10.5808, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 176.833, "mean_ep_length": 200.0, "mean_collisions": 11.9, "mean_intent_rate": 0.673, "policy_loss": 0.014162, "value_loss": 136.124472, "entropy": 0.315094, "total_loss": 68.073246, "clip_frac": 0.159912, "sps": 442.9, "elapsed_s": 2093.6} |
| {"update": 447, "steps_done": 915456, "mean_step_reward": 0.8452, "min_step_reward": -10.5869, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 179.346, "mean_ep_length": 200.0, "mean_collisions": 11.8, "mean_intent_rate": 0.675, "policy_loss": 0.010207, "value_loss": 148.607715, "entropy": 0.313773, "total_loss": 74.310928, "clip_frac": 0.174927, "sps": 439.0, "elapsed_s": 2098.3} |
| {"update": 448, "steps_done": 917504, "mean_step_reward": 1.127, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 217.678, "mean_ep_length": 200.0, "mean_collisions": 10.45, "mean_intent_rate": 0.715, "policy_loss": 0.013139, "value_loss": 142.296951, "entropy": 0.312675, "total_loss": 71.158489, "clip_frac": 0.143677, "sps": 437.5, "elapsed_s": 2103.0} |
| {"update": 449, "steps_done": 919552, "mean_step_reward": 0.7473, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 147.347, "mean_ep_length": 200.0, "mean_collisions": 16.1, "mean_intent_rate": 0.696, "policy_loss": 0.004831, "value_loss": 143.104218, "entropy": 0.319831, "total_loss": 71.553742, "clip_frac": 0.113281, "sps": 433.0, "elapsed_s": 2107.7} |
| {"update": 450, "steps_done": 921600, "mean_step_reward": 0.7745, "min_step_reward": -10.5788, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 155.518, "mean_ep_length": 200.0, "mean_collisions": 16.5, "mean_intent_rate": 0.714, "policy_loss": -0.000793, "value_loss": 118.250124, "entropy": 0.314894, "total_loss": 59.12112, "clip_frac": 0.084473, "sps": 442.7, "elapsed_s": 2112.3} |
| {"update": 451, "steps_done": 923648, "mean_step_reward": 0.7318, "min_step_reward": -10.5864, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 140.911, "mean_ep_length": 200.0, "mean_collisions": 17.0, "mean_intent_rate": 0.699, "policy_loss": 0.003312, "value_loss": 150.157055, "entropy": 0.333079, "total_loss": 75.078509, "clip_frac": 0.079712, "sps": 442.3, "elapsed_s": 2117.0} |
| {"update": 452, "steps_done": 925696, "mean_step_reward": 0.6683, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 142.646, "mean_ep_length": 200.0, "mean_collisions": 17.82, "mean_intent_rate": 0.715, "policy_loss": 0.026795, "value_loss": 175.903306, "entropy": 0.307047, "total_loss": 87.975378, "clip_frac": 0.2323, "sps": 441.6, "elapsed_s": 2121.6} |
| {"update": 453, "steps_done": 927744, "mean_step_reward": 0.5964, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 118.965, "mean_ep_length": 200.0, "mean_collisions": 19.4, "mean_intent_rate": 0.704, "policy_loss": 0.002645, "value_loss": 133.160335, "entropy": 0.339207, "total_loss": 66.57942, "clip_frac": 0.13855, "sps": 434.2, "elapsed_s": 2126.3} |
| {"update": 454, "steps_done": 929792, "mean_step_reward": 0.6696, "min_step_reward": -10.5757, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 131.651, "mean_ep_length": 200.0, "mean_collisions": 17.5, "mean_intent_rate": 0.692, "policy_loss": 0.002008, "value_loss": 130.412376, "entropy": 0.338579, "total_loss": 65.20481, "clip_frac": 0.068604, "sps": 428.0, "elapsed_s": 2131.1} |
| {"update": 455, "steps_done": 931840, "mean_step_reward": 0.6463, "min_step_reward": -10.5794, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 130.618, "mean_ep_length": 200.0, "mean_collisions": 18.5, "mean_intent_rate": 0.706, "policy_loss": -0.000246, "value_loss": 201.417509, "entropy": 0.352759, "total_loss": 100.704982, "clip_frac": 0.107788, "sps": 434.6, "elapsed_s": 2135.8} |
| {"update": 456, "steps_done": 933888, "mean_step_reward": 0.8606, "min_step_reward": -10.5094, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 163.572, "mean_ep_length": 200.0, "mean_collisions": 14.3, "mean_intent_rate": 0.69, "policy_loss": 0.000763, "value_loss": 120.275285, "entropy": 0.340862, "total_loss": 60.134996, "clip_frac": 0.077393, "sps": 436.8, "elapsed_s": 2140.5} |
| {"update": 457, "steps_done": 935936, "mean_step_reward": 0.854, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 176.143, "mean_ep_length": 200.0, "mean_collisions": 12.36, "mean_intent_rate": 0.68, "policy_loss": 0.005395, "value_loss": 122.169693, "entropy": 0.333692, "total_loss": 61.086905, "clip_frac": 0.12146, "sps": 439.9, "elapsed_s": 2145.2} |
| {"update": 458, "steps_done": 937984, "mean_step_reward": 0.8732, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 172.355, "mean_ep_length": 200.0, "mean_collisions": 9.4, "mean_intent_rate": 0.622, "policy_loss": 0.004267, "value_loss": 105.964046, "entropy": 0.33112, "total_loss": 52.982979, "clip_frac": 0.077759, "sps": 443.2, "elapsed_s": 2149.8} |
| {"update": 459, "steps_done": 940032, "mean_step_reward": 1.159, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 232.351, "mean_ep_length": 200.0, "mean_collisions": 5.2, "mean_intent_rate": 0.654, "policy_loss": 0.019106, "value_loss": 107.220993, "entropy": 0.308345, "total_loss": 53.626519, "clip_frac": 0.190674, "sps": 441.6, "elapsed_s": 2154.4} |
| {"update": 460, "steps_done": 942080, "mean_step_reward": 1.235, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 244.933, "mean_ep_length": 200.0, "mean_collisions": 4.8, "mean_intent_rate": 0.666, "policy_loss": 0.017858, "value_loss": 115.935996, "entropy": 0.317839, "total_loss": 57.982677, "clip_frac": 0.175659, "sps": 436.8, "elapsed_s": 2159.1} |
| {"update": 461, "steps_done": 944128, "mean_step_reward": 0.7913, "min_step_reward": -10.5831, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 165.998, "mean_ep_length": 200.0, "mean_collisions": 13.18, "mean_intent_rate": 0.675, "policy_loss": -0.002462, "value_loss": 121.573338, "entropy": 0.320808, "total_loss": 60.780999, "clip_frac": 0.0896, "sps": 436.1, "elapsed_s": 2163.8} |
| {"update": 462, "steps_done": 946176, "mean_step_reward": 1.0303, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 201.48, "mean_ep_length": 200.0, "mean_collisions": 10.2, "mean_intent_rate": 0.685, "policy_loss": 0.004817, "value_loss": 129.137254, "entropy": 0.313214, "total_loss": 64.570312, "clip_frac": 0.094971, "sps": 427.2, "elapsed_s": 2168.6} |
| {"update": 463, "steps_done": 948224, "mean_step_reward": 1.1005, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 221.587, "mean_ep_length": 200.0, "mean_collisions": 5.8, "mean_intent_rate": 0.645, "policy_loss": 0.010979, "value_loss": 116.818882, "entropy": 0.292465, "total_loss": 58.417496, "clip_frac": 0.11792, "sps": 438.9, "elapsed_s": 2173.3} |
| {"update": 464, "steps_done": 950272, "mean_step_reward": 0.8884, "min_step_reward": -10.5448, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 181.814, "mean_ep_length": 200.0, "mean_collisions": 11.5, "mean_intent_rate": 0.672, "policy_loss": 0.010778, "value_loss": 149.046622, "entropy": 0.30131, "total_loss": 74.531077, "clip_frac": 0.172852, "sps": 439.0, "elapsed_s": 2177.9} |
| {"update": 465, "steps_done": 952320, "mean_step_reward": 0.9711, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 190.511, "mean_ep_length": 200.0, "mean_collisions": 11.18, "mean_intent_rate": 0.682, "policy_loss": 0.008706, "value_loss": 145.145392, "entropy": 0.29806, "total_loss": 72.578421, "clip_frac": 0.120361, "sps": 448.4, "elapsed_s": 2182.5} |
| {"update": 466, "steps_done": 954368, "mean_step_reward": 0.8575, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 167.652, "mean_ep_length": 200.0, "mean_collisions": 13.3, "mean_intent_rate": 0.679, "policy_loss": 0.004069, "value_loss": 151.762316, "entropy": 0.317468, "total_loss": 75.882053, "clip_frac": 0.114746, "sps": 443.0, "elapsed_s": 2187.1} |
| {"update": 467, "steps_done": 956416, "mean_step_reward": 0.8965, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 192.21, "mean_ep_length": 200.0, "mean_collisions": 10.5, "mean_intent_rate": 0.675, "policy_loss": 0.012907, "value_loss": 129.823665, "entropy": 0.305592, "total_loss": 64.921684, "clip_frac": 0.150757, "sps": 432.7, "elapsed_s": 2191.9} |
| {"update": 468, "steps_done": 958464, "mean_step_reward": 1.01, "min_step_reward": -10.5702, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 200.109, "mean_ep_length": 200.0, "mean_collisions": 11.3, "mean_intent_rate": 0.701, "policy_loss": 0.000608, "value_loss": 138.005966, "entropy": 0.28392, "total_loss": 69.000751, "clip_frac": 0.046265, "sps": 436.8, "elapsed_s": 2196.6} |
| {"update": 469, "steps_done": 960512, "mean_step_reward": 1.1125, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 217.021, "mean_ep_length": 200.0, "mean_collisions": 10.82, "mean_intent_rate": 0.721, "policy_loss": 0.005378, "value_loss": 126.521531, "entropy": 0.278598, "total_loss": 63.263358, "clip_frac": 0.083008, "sps": 436.9, "elapsed_s": 2201.3} |
| {"update": 470, "steps_done": 962560, "mean_step_reward": 0.6597, "min_step_reward": -10.5804, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 131.128, "mean_ep_length": 200.0, "mean_collisions": 15.6, "mean_intent_rate": 0.658, "policy_loss": 0.005363, "value_loss": 138.268769, "entropy": 0.310906, "total_loss": 69.136639, "clip_frac": 0.098877, "sps": 436.9, "elapsed_s": 2205.9} |
| {"update": 471, "steps_done": 964608, "mean_step_reward": 0.7569, "min_step_reward": -10.5477, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 153.694, "mean_ep_length": 200.0, "mean_collisions": 14.3, "mean_intent_rate": 0.673, "policy_loss": 0.01358, "value_loss": 137.376871, "entropy": 0.316555, "total_loss": 68.69885, "clip_frac": 0.156006, "sps": 427.0, "elapsed_s": 2210.7} |
| {"update": 472, "steps_done": 966656, "mean_step_reward": 0.8714, "min_step_reward": -10.5678, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 160.038, "mean_ep_length": 200.0, "mean_collisions": 15.0, "mean_intent_rate": 0.697, "policy_loss": 0.004564, "value_loss": 140.709923, "entropy": 0.294865, "total_loss": 70.356577, "clip_frac": 0.120239, "sps": 439.4, "elapsed_s": 2215.4} |
| {"update": 473, "steps_done": 968704, "mean_step_reward": 1.0574, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 219.518, "mean_ep_length": 200.0, "mean_collisions": 8.64, "mean_intent_rate": 0.688, "policy_loss": 0.016757, "value_loss": 116.96631, "entropy": 0.306607, "total_loss": 58.496846, "clip_frac": 0.145142, "sps": 435.5, "elapsed_s": 2220.1} |
| {"update": 474, "steps_done": 970752, "mean_step_reward": 1.075, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 214.044, "mean_ep_length": 200.0, "mean_collisions": 11.3, "mean_intent_rate": 0.724, "policy_loss": 0.03465, "value_loss": 138.553503, "entropy": 0.274613, "total_loss": 69.308656, "clip_frac": 0.231201, "sps": 439.3, "elapsed_s": 2224.8} |
| {"update": 475, "steps_done": 972800, "mean_step_reward": 0.4499, "min_step_reward": -10.5569, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 80.246, "mean_ep_length": 200.0, "mean_collisions": 24.0, "mean_intent_rate": 0.714, "policy_loss": 0.004268, "value_loss": 183.207033, "entropy": 0.290612, "total_loss": 91.604878, "clip_frac": 0.178833, "sps": 434.2, "elapsed_s": 2229.5} |
| {"update": 476, "steps_done": 974848, "mean_step_reward": 0.3053, "min_step_reward": -10.5851, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 70.058, "mean_ep_length": 200.0, "mean_collisions": 16.9, "mean_intent_rate": 0.581, "policy_loss": 0.012222, "value_loss": 108.623943, "entropy": 0.316027, "total_loss": 54.321034, "clip_frac": 0.147217, "sps": 443.8, "elapsed_s": 2234.1} |
| {"update": 477, "steps_done": 976896, "mean_step_reward": 0.6207, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 122.2, "mean_ep_length": 193.3, "mean_collisions": 19.36, "mean_intent_rate": 0.727, "policy_loss": 0.011026, "value_loss": 128.00925, "entropy": 0.28317, "total_loss": 64.012819, "clip_frac": 0.166504, "sps": 439.0, "elapsed_s": 2238.8} |
| {"update": 478, "steps_done": 978944, "mean_step_reward": 0.1155, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 18.242, "mean_ep_length": 200.0, "mean_collisions": 28.2, "mean_intent_rate": 0.683, "policy_loss": 0.012255, "value_loss": 196.115488, "entropy": 0.340193, "total_loss": 98.066597, "clip_frac": 0.226318, "sps": 438.8, "elapsed_s": 2243.4} |
| {"update": 479, "steps_done": 980992, "mean_step_reward": 0.6958, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 147.193, "mean_ep_length": 200.0, "mean_collisions": 19.1, "mean_intent_rate": 0.743, "policy_loss": 0.008088, "value_loss": 143.992565, "entropy": 0.334626, "total_loss": 72.001024, "clip_frac": 0.157471, "sps": 438.2, "elapsed_s": 2248.1} |
| {"update": 480, "steps_done": 983040, "mean_step_reward": 0.8351, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 155.686, "mean_ep_length": 200.0, "mean_collisions": 17.18, "mean_intent_rate": 0.725, "policy_loss": 0.011002, "value_loss": 134.940126, "entropy": 0.296152, "total_loss": 67.478103, "clip_frac": 0.146362, "sps": 431.4, "elapsed_s": 2252.9} |
| {"update": 481, "steps_done": 985088, "mean_step_reward": 0.5514, "min_step_reward": -10.5824, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 105.37, "mean_ep_length": 200.0, "mean_collisions": 22.6, "mean_intent_rate": 0.733, "policy_loss": 0.006978, "value_loss": 149.107638, "entropy": 0.304004, "total_loss": 74.557757, "clip_frac": 0.123047, "sps": 435.7, "elapsed_s": 2257.6} |
| {"update": 482, "steps_done": 987136, "mean_step_reward": 0.6065, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 151.641, "mean_ep_length": 200.0, "mean_collisions": 18.0, "mean_intent_rate": 0.732, "policy_loss": 0.000222, "value_loss": 169.692959, "entropy": 0.289886, "total_loss": 84.843803, "clip_frac": 0.067139, "sps": 443.3, "elapsed_s": 2262.2} |
| {"update": 483, "steps_done": 989184, "mean_step_reward": 0.4394, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 63.839, "mean_ep_length": 200.0, "mean_collisions": 25.9, "mean_intent_rate": 0.719, "policy_loss": 0.002346, "value_loss": 169.706836, "entropy": 0.332957, "total_loss": 84.852434, "clip_frac": 0.123535, "sps": 438.0, "elapsed_s": 2266.8} |
| {"update": 484, "steps_done": 991232, "mean_step_reward": 0.4919, "min_step_reward": -10.5095, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 100.31, "mean_ep_length": 200.0, "mean_collisions": 17.64, "mean_intent_rate": 0.644, "policy_loss": 0.012622, "value_loss": 139.535196, "entropy": 0.348759, "total_loss": 69.776732, "clip_frac": 0.171265, "sps": 441.9, "elapsed_s": 2271.5} |
| {"update": 485, "steps_done": 993280, "mean_step_reward": 0.7575, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 161.202, "mean_ep_length": 200.0, "mean_collisions": 13.8, "mean_intent_rate": 0.678, "policy_loss": 0.017298, "value_loss": 131.467855, "entropy": 0.318339, "total_loss": 65.748042, "clip_frac": 0.200195, "sps": 441.9, "elapsed_s": 2276.1} |
| {"update": 486, "steps_done": 995328, "mean_step_reward": 0.5259, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 89.128, "mean_ep_length": 200.0, "mean_collisions": 22.6, "mean_intent_rate": 0.705, "policy_loss": 0.009901, "value_loss": 198.435022, "entropy": 0.311759, "total_loss": 99.224295, "clip_frac": 0.139404, "sps": 440.7, "elapsed_s": 2280.8} |
| {"update": 487, "steps_done": 997376, "mean_step_reward": 0.37, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 72.665, "mean_ep_length": 182.2, "mean_collisions": 22.09, "mean_intent_rate": 0.72, "policy_loss": 0.012154, "value_loss": 164.714913, "entropy": 0.32408, "total_loss": 82.36637, "clip_frac": 0.138306, "sps": 434.1, "elapsed_s": 2285.5} |
| {"update": 488, "steps_done": 999424, "mean_step_reward": 1.1449, "min_step_reward": -10.5762, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 220.883, "mean_ep_length": 200.0, "mean_collisions": 7.91, "mean_intent_rate": 0.677, "policy_loss": 0.011747, "value_loss": 107.97667, "entropy": 0.316427, "total_loss": 53.996918, "clip_frac": 0.147827, "sps": 435.8, "elapsed_s": 2290.2} |
| {"update": 489, "steps_done": 1001472, "mean_step_reward": 0.9569, "min_step_reward": -10.5813, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 187.401, "mean_ep_length": 200.0, "mean_collisions": 8.9, "mean_intent_rate": 0.641, "policy_loss": 0.009819, "value_loss": 94.971357, "entropy": 0.312943, "total_loss": 47.492368, "clip_frac": 0.121094, "sps": 430.9, "elapsed_s": 2294.9} |
| {"update": 490, "steps_done": 1003520, "mean_step_reward": 0.8604, "min_step_reward": -10.5287, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 180.068, "mean_ep_length": 200.0, "mean_collisions": 9.1, "mean_intent_rate": 0.632, "policy_loss": 0.01397, "value_loss": 101.371847, "entropy": 0.317807, "total_loss": 50.696716, "clip_frac": 0.16333, "sps": 441.2, "elapsed_s": 2299.6} |
| {"update": 491, "steps_done": 1005568, "mean_step_reward": 0.9808, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 190.806, "mean_ep_length": 200.0, "mean_collisions": 11.7, "mean_intent_rate": 0.693, "policy_loss": 0.011061, "value_loss": 128.852579, "entropy": 0.300055, "total_loss": 64.434351, "clip_frac": 0.162964, "sps": 434.0, "elapsed_s": 2304.3} |
| {"update": 492, "steps_done": 1007616, "mean_step_reward": 1.0499, "min_step_reward": -10.5238, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 209.602, "mean_ep_length": 200.0, "mean_collisions": 11.6, "mean_intent_rate": 0.721, "policy_loss": 0.003897, "value_loss": 141.062467, "entropy": 0.287021, "total_loss": 70.53226, "clip_frac": 0.109863, "sps": 442.2, "elapsed_s": 2308.9} |
| {"update": 493, "steps_done": 1009664, "mean_step_reward": 0.8224, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 170.21, "mean_ep_length": 200.0, "mean_collisions": 13.73, "mean_intent_rate": 0.692, "policy_loss": 0.001793, "value_loss": 123.599576, "entropy": 0.300153, "total_loss": 61.798579, "clip_frac": 0.088501, "sps": 440.2, "elapsed_s": 2313.6} |
| {"update": 494, "steps_done": 1011712, "mean_step_reward": 0.6707, "min_step_reward": -10.5862, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 129.512, "mean_ep_length": 200.0, "mean_collisions": 17.3, "mean_intent_rate": 0.685, "policy_loss": 0.004942, "value_loss": 143.093879, "entropy": 0.298036, "total_loss": 71.548901, "clip_frac": 0.147339, "sps": 437.3, "elapsed_s": 2318.3} |
| {"update": 495, "steps_done": 1013760, "mean_step_reward": 1.0095, "min_step_reward": -10.5351, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 200.501, "mean_ep_length": 200.0, "mean_collisions": 11.8, "mean_intent_rate": 0.71, "policy_loss": 0.007365, "value_loss": 132.081853, "entropy": 0.28233, "total_loss": 66.045468, "clip_frac": 0.139771, "sps": 438.1, "elapsed_s": 2323.0} |
| {"update": 496, "steps_done": 1015808, "mean_step_reward": 0.729, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 150.303, "mean_ep_length": 200.0, "mean_collisions": 16.6, "mean_intent_rate": 0.708, "policy_loss": 0.005041, "value_loss": 181.496308, "entropy": 0.296508, "total_loss": 90.75023, "clip_frac": 0.078735, "sps": 437.1, "elapsed_s": 2327.6} |
| {"update": 497, "steps_done": 1017856, "mean_step_reward": 0.7702, "min_step_reward": -10.5379, "max_step_reward": 2.5, "n_episodes_finished": 12, "mean_ep_return": 142.965, "mean_ep_length": 185.6, "mean_collisions": 15.5, "mean_intent_rate": 0.703, "policy_loss": 0.004164, "value_loss": 177.381849, "entropy": 0.296631, "total_loss": 88.692122, "clip_frac": 0.109253, "sps": 438.4, "elapsed_s": 2332.3} |
| {"update": 498, "steps_done": 1019904, "mean_step_reward": 0.7485, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 154.96, "mean_ep_length": 200.0, "mean_collisions": 17.3, "mean_intent_rate": 0.726, "policy_loss": 0.003083, "value_loss": 144.090725, "entropy": 0.28964, "total_loss": 72.045549, "clip_frac": 0.098267, "sps": 433.3, "elapsed_s": 2337.0} |
| {"update": 499, "steps_done": 1021952, "mean_step_reward": 0.5354, "min_step_reward": -10.5353, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 97.575, "mean_ep_length": 200.0, "mean_collisions": 20.5, "mean_intent_rate": 0.687, "policy_loss": 0.003719, "value_loss": 150.912147, "entropy": 0.30516, "total_loss": 75.456741, "clip_frac": 0.072876, "sps": 439.5, "elapsed_s": 2341.7} |
| {"update": 500, "steps_done": 1024000, "mean_step_reward": 0.7357, "min_step_reward": -10.5358, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 157.069, "mean_ep_length": 200.0, "mean_collisions": 16.2, "mean_intent_rate": 0.711, "policy_loss": 0.011255, "value_loss": 116.548649, "entropy": 0.301878, "total_loss": 58.28256, "clip_frac": 0.13623, "sps": 439.1, "elapsed_s": 2346.4} |
| {"update": 501, "steps_done": 1026048, "mean_step_reward": 0.8436, "min_step_reward": -10.5224, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 142.832, "mean_ep_length": 200.0, "mean_collisions": 16.7, "mean_intent_rate": 0.697, "policy_loss": 0.009679, "value_loss": 113.8975, "entropy": 0.303849, "total_loss": 56.955391, "clip_frac": 0.147705, "sps": 445.9, "elapsed_s": 2351.0} |
| {"update": 502, "steps_done": 1028096, "mean_step_reward": 0.6422, "min_step_reward": -10.584, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 143.142, "mean_ep_length": 200.0, "mean_collisions": 19.55, "mean_intent_rate": 0.745, "policy_loss": 0.008782, "value_loss": 158.674717, "entropy": 0.298128, "total_loss": 79.343159, "clip_frac": 0.127075, "sps": 445.5, "elapsed_s": 2355.6} |
| {"update": 503, "steps_done": 1030144, "mean_step_reward": 0.7091, "min_step_reward": -10.5831, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 165.314, "mean_ep_length": 196.3, "mean_collisions": 15.4, "mean_intent_rate": 0.722, "policy_loss": 0.010839, "value_loss": 153.936555, "entropy": 0.313837, "total_loss": 76.975978, "clip_frac": 0.16272, "sps": 443.8, "elapsed_s": 2360.2} |
| {"update": 504, "steps_done": 1032192, "mean_step_reward": 0.6834, "min_step_reward": -10.5229, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 130.026, "mean_ep_length": 200.0, "mean_collisions": 18.7, "mean_intent_rate": 0.707, "policy_loss": 0.024302, "value_loss": 134.666816, "entropy": 0.334141, "total_loss": 67.354369, "clip_frac": 0.210449, "sps": 438.1, "elapsed_s": 2364.9} |
| {"update": 505, "steps_done": 1034240, "mean_step_reward": 1.0004, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 181.608, "mean_ep_length": 200.0, "mean_collisions": 11.18, "mean_intent_rate": 0.67, "policy_loss": 0.011374, "value_loss": 107.446195, "entropy": 0.338187, "total_loss": 53.731089, "clip_frac": 0.123291, "sps": 437.2, "elapsed_s": 2369.5} |
| {"update": 506, "steps_done": 1036288, "mean_step_reward": 1.1444, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 238.558, "mean_ep_length": 200.0, "mean_collisions": 5.0, "mean_intent_rate": 0.658, "policy_loss": 0.018847, "value_loss": 101.37774, "entropy": 0.325882, "total_loss": 50.704458, "clip_frac": 0.170898, "sps": 437.4, "elapsed_s": 2374.2} |
| {"update": 507, "steps_done": 1038336, "mean_step_reward": 1.136, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 213.586, "mean_ep_length": 200.0, "mean_collisions": 9.9, "mean_intent_rate": 0.7, "policy_loss": 0.007189, "value_loss": 139.673759, "entropy": 0.28327, "total_loss": 69.841235, "clip_frac": 0.132935, "sps": 435.4, "elapsed_s": 2378.9} |
| {"update": 508, "steps_done": 1040384, "mean_step_reward": 0.9387, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 213.825, "mean_ep_length": 200.0, "mean_collisions": 7.9, "mean_intent_rate": 0.667, "policy_loss": 0.004619, "value_loss": 131.873367, "entropy": 0.307752, "total_loss": 65.938225, "clip_frac": 0.112671, "sps": 437.7, "elapsed_s": 2383.6} |
| {"update": 509, "steps_done": 1042432, "mean_step_reward": 0.9755, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 177.271, "mean_ep_length": 200.0, "mean_collisions": 5.27, "mean_intent_rate": 0.563, "policy_loss": 0.017063, "value_loss": 99.287748, "entropy": 0.307326, "total_loss": 49.657864, "clip_frac": 0.182129, "sps": 445.6, "elapsed_s": 2388.2} |
| {"update": 510, "steps_done": 1044480, "mean_step_reward": 0.5875, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 110.676, "mean_ep_length": 200.0, "mean_collisions": 16.6, "mean_intent_rate": 0.64, "policy_loss": 0.021366, "value_loss": 181.931861, "entropy": 0.325317, "total_loss": 90.984043, "clip_frac": 0.210205, "sps": 441.3, "elapsed_s": 2392.8} |
| {"update": 511, "steps_done": 1046528, "mean_step_reward": 0.784, "min_step_reward": -10.5872, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 169.65, "mean_ep_length": 200.0, "mean_collisions": 15.1, "mean_intent_rate": 0.714, "policy_loss": 0.010911, "value_loss": 165.045558, "entropy": 0.312894, "total_loss": 82.530561, "clip_frac": 0.191284, "sps": 440.2, "elapsed_s": 2397.5} |
| {"update": 512, "steps_done": 1048576, "mean_step_reward": 0.942, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 176.123, "mean_ep_length": 200.0, "mean_collisions": 12.8, "mean_intent_rate": 0.687, "policy_loss": 0.007353, "value_loss": 150.032238, "entropy": 0.310238, "total_loss": 75.020369, "clip_frac": 0.113159, "sps": 444.7, "elapsed_s": 2402.1} |
| {"update": 513, "steps_done": 1050624, "mean_step_reward": 0.8541, "min_step_reward": -10.5672, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 174.703, "mean_ep_length": 200.0, "mean_collisions": 12.36, "mean_intent_rate": 0.677, "policy_loss": 0.013851, "value_loss": 129.847651, "entropy": 0.297759, "total_loss": 64.934699, "clip_frac": 0.114502, "sps": 443.4, "elapsed_s": 2406.7} |
| {"update": 514, "steps_done": 1052672, "mean_step_reward": 0.7492, "min_step_reward": -10.5444, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 149.971, "mean_ep_length": 200.0, "mean_collisions": 18.9, "mean_intent_rate": 0.744, "policy_loss": 0.004667, "value_loss": 150.072765, "entropy": 0.300182, "total_loss": 75.038049, "clip_frac": 0.139771, "sps": 439.4, "elapsed_s": 2411.4} |
| {"update": 515, "steps_done": 1054720, "mean_step_reward": 0.6077, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 113.264, "mean_ep_length": 183.3, "mean_collisions": 18.09, "mean_intent_rate": 0.691, "policy_loss": 0.005077, "value_loss": 169.472969, "entropy": 0.324892, "total_loss": 84.738313, "clip_frac": 0.116577, "sps": 437.7, "elapsed_s": 2416.1} |
| {"update": 516, "steps_done": 1056768, "mean_step_reward": 0.84, "min_step_reward": -10.5814, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 163.06, "mean_ep_length": 200.0, "mean_collisions": 15.9, "mean_intent_rate": 0.715, "policy_loss": 0.003664, "value_loss": 140.003103, "entropy": 0.329127, "total_loss": 70.001923, "clip_frac": 0.073364, "sps": 432.7, "elapsed_s": 2420.8} |
| {"update": 517, "steps_done": 1058816, "mean_step_reward": 0.6955, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 142.377, "mean_ep_length": 200.0, "mean_collisions": 16.7, "mean_intent_rate": 0.695, "policy_loss": 0.015664, "value_loss": 118.519567, "entropy": 0.32372, "total_loss": 59.27221, "clip_frac": 0.162842, "sps": 443.3, "elapsed_s": 2425.4} |
| {"update": 518, "steps_done": 1060864, "mean_step_reward": 0.927, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 181.517, "mean_ep_length": 200.0, "mean_collisions": 15.0, "mean_intent_rate": 0.731, "policy_loss": -0.00286, "value_loss": 143.220644, "entropy": 0.318153, "total_loss": 71.604281, "clip_frac": 0.106567, "sps": 445.7, "elapsed_s": 2430.0} |
| {"update": 519, "steps_done": 1062912, "mean_step_reward": 0.8557, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 170.278, "mean_ep_length": 200.0, "mean_collisions": 13.7, "mean_intent_rate": 0.692, "policy_loss": 0.004774, "value_loss": 126.265101, "entropy": 0.313334, "total_loss": 63.134191, "clip_frac": 0.118408, "sps": 438.2, "elapsed_s": 2434.7} |
| {"update": 520, "steps_done": 1064960, "mean_step_reward": 1.0231, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 205.927, "mean_ep_length": 200.0, "mean_collisions": 10.2, "mean_intent_rate": 0.693, "policy_loss": 0.005548, "value_loss": 106.74672, "entropy": 0.290693, "total_loss": 53.376001, "clip_frac": 0.109741, "sps": 440.6, "elapsed_s": 2439.3} |
| {"update": 521, "steps_done": 1067008, "mean_step_reward": 0.9397, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 198.742, "mean_ep_length": 200.0, "mean_collisions": 10.3, "mean_intent_rate": 0.682, "policy_loss": 0.00323, "value_loss": 140.836416, "entropy": 0.321712, "total_loss": 70.418222, "clip_frac": 0.066772, "sps": 439.3, "elapsed_s": 2444.0} |
| {"update": 522, "steps_done": 1069056, "mean_step_reward": 0.9252, "min_step_reward": -10.5801, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 173.814, "mean_ep_length": 200.0, "mean_collisions": 12.91, "mean_intent_rate": 0.683, "policy_loss": -0.000951, "value_loss": 136.691721, "entropy": 0.317198, "total_loss": 68.341738, "clip_frac": 0.069946, "sps": 442.0, "elapsed_s": 2448.6} |
| {"update": 523, "steps_done": 1071104, "mean_step_reward": 0.9619, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 195.061, "mean_ep_length": 200.0, "mean_collisions": 11.6, "mean_intent_rate": 0.697, "policy_loss": 0.0044, "value_loss": 139.397263, "entropy": 0.329311, "total_loss": 69.699738, "clip_frac": 0.167236, "sps": 444.4, "elapsed_s": 2453.2} |
| {"update": 524, "steps_done": 1073152, "mean_step_reward": 0.9354, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 184.294, "mean_ep_length": 200.0, "mean_collisions": 12.9, "mean_intent_rate": 0.701, "policy_loss": 0.021331, "value_loss": 121.838248, "entropy": 0.3296, "total_loss": 60.937159, "clip_frac": 0.19043, "sps": 435.6, "elapsed_s": 2457.9} |
| {"update": 525, "steps_done": 1075200, "mean_step_reward": 0.8891, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 181.878, "mean_ep_length": 200.0, "mean_collisions": 14.3, "mean_intent_rate": 0.721, "policy_loss": 0.012112, "value_loss": 154.57399, "entropy": 0.308872, "total_loss": 77.296018, "clip_frac": 0.197388, "sps": 443.2, "elapsed_s": 2462.6} |
| {"update": 526, "steps_done": 1077248, "mean_step_reward": 0.8825, "min_step_reward": -10.5568, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 175.53, "mean_ep_length": 200.0, "mean_collisions": 9.64, "mean_intent_rate": 0.633, "policy_loss": 0.000411, "value_loss": 132.917198, "entropy": 0.324729, "total_loss": 66.455763, "clip_frac": 0.097778, "sps": 438.7, "elapsed_s": 2467.2} |
| {"update": 527, "steps_done": 1079296, "mean_step_reward": 0.9746, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 191.307, "mean_ep_length": 200.0, "mean_collisions": 10.0, "mean_intent_rate": 0.664, "policy_loss": 0.008442, "value_loss": 129.206056, "entropy": 0.314668, "total_loss": 64.608323, "clip_frac": 0.151245, "sps": 436.4, "elapsed_s": 2471.9} |
| {"update": 528, "steps_done": 1081344, "mean_step_reward": 0.808, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 170.292, "mean_ep_length": 200.0, "mean_collisions": 13.5, "mean_intent_rate": 0.69, "policy_loss": 0.001581, "value_loss": 119.396119, "entropy": 0.316194, "total_loss": 59.696479, "clip_frac": 0.103271, "sps": 436.8, "elapsed_s": 2476.6} |
| {"update": 529, "steps_done": 1083392, "mean_step_reward": 0.6049, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 112.912, "mean_ep_length": 200.0, "mean_collisions": 19.5, "mean_intent_rate": 0.693, "policy_loss": 0.027536, "value_loss": 154.449902, "entropy": 0.309158, "total_loss": 77.249395, "clip_frac": 0.220947, "sps": 438.6, "elapsed_s": 2481.3} |
| {"update": 530, "steps_done": 1085440, "mean_step_reward": 0.5523, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 116.325, "mean_ep_length": 200.0, "mean_collisions": 18.55, "mean_intent_rate": 0.682, "policy_loss": 0.008221, "value_loss": 161.210886, "entropy": 0.305361, "total_loss": 80.61061, "clip_frac": 0.18811, "sps": 443.4, "elapsed_s": 2485.9} |
| {"update": 531, "steps_done": 1087488, "mean_step_reward": 0.9697, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 192.791, "mean_ep_length": 200.0, "mean_collisions": 11.5, "mean_intent_rate": 0.692, "policy_loss": 0.009616, "value_loss": 166.519382, "entropy": 0.277903, "total_loss": 83.266528, "clip_frac": 0.133911, "sps": 439.0, "elapsed_s": 2490.6} |
| {"update": 532, "steps_done": 1089536, "mean_step_reward": 0.8259, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 158.411, "mean_ep_length": 188.0, "mean_collisions": 12.36, "mean_intent_rate": 0.676, "policy_loss": -0.001025, "value_loss": 137.465673, "entropy": 0.312044, "total_loss": 68.728692, "clip_frac": 0.084473, "sps": 438.0, "elapsed_s": 2495.3} |
| {"update": 533, "steps_done": 1091584, "mean_step_reward": 0.9839, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 192.003, "mean_ep_length": 200.0, "mean_collisions": 10.5, "mean_intent_rate": 0.673, "policy_loss": 0.016221, "value_loss": 122.544443, "entropy": 0.291869, "total_loss": 61.285524, "clip_frac": 0.16626, "sps": 430.7, "elapsed_s": 2500.0} |
| {"update": 534, "steps_done": 1093632, "mean_step_reward": 0.4479, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 92.584, "mean_ep_length": 200.0, "mean_collisions": 25.0, "mean_intent_rate": 0.75, "policy_loss": 0.010043, "value_loss": 224.861905, "entropy": 0.299407, "total_loss": 112.438001, "clip_frac": 0.105591, "sps": 440.9, "elapsed_s": 2504.7} |
| {"update": 535, "steps_done": 1095680, "mean_step_reward": 0.7665, "min_step_reward": -10.5826, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 156.374, "mean_ep_length": 200.0, "mean_collisions": 17.9, "mean_intent_rate": 0.738, "policy_loss": 0.00991, "value_loss": 145.482536, "entropy": 0.290762, "total_loss": 72.748271, "clip_frac": 0.155762, "sps": 440.1, "elapsed_s": 2509.3} |
| {"update": 536, "steps_done": 1097728, "mean_step_reward": 0.3528, "min_step_reward": -10.5631, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 72.192, "mean_ep_length": 200.0, "mean_collisions": 23.82, "mean_intent_rate": 0.698, "policy_loss": 0.003785, "value_loss": 213.771353, "entropy": 0.321725, "total_loss": 106.886245, "clip_frac": 0.075928, "sps": 436.8, "elapsed_s": 2514.0} |
| {"update": 537, "steps_done": 1099776, "mean_step_reward": 0.7987, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 158.992, "mean_ep_length": 200.0, "mean_collisions": 16.0, "mean_intent_rate": 0.712, "policy_loss": 0.011178, "value_loss": 143.753399, "entropy": 0.305331, "total_loss": 71.884825, "clip_frac": 0.145996, "sps": 434.9, "elapsed_s": 2518.7} |
| {"update": 538, "steps_done": 1101824, "mean_step_reward": 0.9151, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 194.289, "mean_ep_length": 200.0, "mean_collisions": 12.4, "mean_intent_rate": 0.71, "policy_loss": 0.017012, "value_loss": 125.210239, "entropy": 0.327128, "total_loss": 62.61886, "clip_frac": 0.204956, "sps": 439.6, "elapsed_s": 2523.4} |
| {"update": 539, "steps_done": 1103872, "mean_step_reward": 0.7991, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 150.095, "mean_ep_length": 200.0, "mean_collisions": 16.0, "mean_intent_rate": 0.697, "policy_loss": 0.007387, "value_loss": 149.186594, "entropy": 0.322342, "total_loss": 74.597461, "clip_frac": 0.10083, "sps": 439.0, "elapsed_s": 2528.0} |
| {"update": 540, "steps_done": 1105920, "mean_step_reward": 0.7921, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 157.54, "mean_ep_length": 200.0, "mean_collisions": 13.73, "mean_intent_rate": 0.672, "policy_loss": 0.005232, "value_loss": 120.480311, "entropy": 0.316138, "total_loss": 60.242225, "clip_frac": 0.113159, "sps": 442.4, "elapsed_s": 2532.7} |
| {"update": 541, "steps_done": 1107968, "mean_step_reward": 1.0898, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 220.652, "mean_ep_length": 200.0, "mean_collisions": 9.6, "mean_intent_rate": 0.706, "policy_loss": 0.007838, "value_loss": 123.587909, "entropy": 0.307343, "total_loss": 61.798719, "clip_frac": 0.147217, "sps": 439.4, "elapsed_s": 2537.3} |
| {"update": 542, "steps_done": 1110016, "mean_step_reward": 0.8711, "min_step_reward": -10.5554, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 176.282, "mean_ep_length": 200.0, "mean_collisions": 12.1, "mean_intent_rate": 0.674, "policy_loss": 0.004129, "value_loss": 119.633562, "entropy": 0.305585, "total_loss": 59.817854, "clip_frac": 0.128052, "sps": 425.5, "elapsed_s": 2542.1} |
| {"update": 543, "steps_done": 1112064, "mean_step_reward": 0.7638, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 142.189, "mean_ep_length": 200.0, "mean_collisions": 17.8, "mean_intent_rate": 0.713, "policy_loss": 0.027366, "value_loss": 201.170413, "entropy": 0.297819, "total_loss": 100.609594, "clip_frac": 0.250488, "sps": 438.0, "elapsed_s": 2546.8} |
| {"update": 544, "steps_done": 1114112, "mean_step_reward": 0.771, "min_step_reward": -10.5868, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 162.308, "mean_ep_length": 200.0, "mean_collisions": 16.27, "mean_intent_rate": 0.72, "policy_loss": -0.000402, "value_loss": 219.406975, "entropy": 0.29567, "total_loss": 109.700128, "clip_frac": 0.078247, "sps": 440.6, "elapsed_s": 2551.5} |
| {"update": 545, "steps_done": 1116160, "mean_step_reward": 0.8057, "min_step_reward": -10.584, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 177.03, "mean_ep_length": 200.0, "mean_collisions": 12.1, "mean_intent_rate": 0.676, "policy_loss": 0.005048, "value_loss": 173.032071, "entropy": 0.332845, "total_loss": 86.517754, "clip_frac": 0.084717, "sps": 441.0, "elapsed_s": 2556.1} |
| {"update": 546, "steps_done": 1118208, "mean_step_reward": 0.8833, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 153.36, "mean_ep_length": 200.0, "mean_collisions": 13.3, "mean_intent_rate": 0.655, "policy_loss": 0.002711, "value_loss": 187.1863, "entropy": 0.313722, "total_loss": 93.592724, "clip_frac": 0.094971, "sps": 440.8, "elapsed_s": 2560.8} |
| {"update": 547, "steps_done": 1120256, "mean_step_reward": 1.1764, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 243.926, "mean_ep_length": 200.0, "mean_collisions": 6.2, "mean_intent_rate": 0.689, "policy_loss": 0.007687, "value_loss": 151.021764, "entropy": 0.319171, "total_loss": 75.515377, "clip_frac": 0.123047, "sps": 442.6, "elapsed_s": 2565.4} |
| {"update": 548, "steps_done": 1122304, "mean_step_reward": 0.9127, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 175.576, "mean_ep_length": 194.2, "mean_collisions": 11.36, "mean_intent_rate": 0.678, "policy_loss": 0.002401, "value_loss": 144.338949, "entropy": 0.331453, "total_loss": 72.168561, "clip_frac": 0.10791, "sps": 435.6, "elapsed_s": 2570.1} |
| {"update": 549, "steps_done": 1124352, "mean_step_reward": 0.8486, "min_step_reward": -10.5336, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 165.722, "mean_ep_length": 200.0, "mean_collisions": 14.3, "mean_intent_rate": 0.695, "policy_loss": 0.001956, "value_loss": 144.955916, "entropy": 0.339113, "total_loss": 72.476522, "clip_frac": 0.131714, "sps": 434.4, "elapsed_s": 2574.8} |
| {"update": 550, "steps_done": 1126400, "mean_step_reward": 0.699, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 143.466, "mean_ep_length": 200.0, "mean_collisions": 14.8, "mean_intent_rate": 0.666, "policy_loss": 0.012376, "value_loss": 149.783683, "entropy": 0.343691, "total_loss": 74.90078, "clip_frac": 0.171387, "sps": 435.8, "elapsed_s": 2579.5} |
| {"update": 551, "steps_done": 1128448, "mean_step_reward": 0.8401, "min_step_reward": -10.5374, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 170.465, "mean_ep_length": 200.0, "mean_collisions": 12.91, "mean_intent_rate": 0.679, "policy_loss": 0.004722, "value_loss": 127.807053, "entropy": 0.303939, "total_loss": 63.90521, "clip_frac": 0.068848, "sps": 431.2, "elapsed_s": 2584.3} |
| {"update": 552, "steps_done": 1130496, "mean_step_reward": 0.801, "min_step_reward": -10.544, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 160.988, "mean_ep_length": 200.0, "mean_collisions": 11.6, "mean_intent_rate": 0.641, "policy_loss": 0.007776, "value_loss": 141.02769, "entropy": 0.32705, "total_loss": 70.518351, "clip_frac": 0.131226, "sps": 439.2, "elapsed_s": 2588.9} |
| {"update": 553, "steps_done": 1132544, "mean_step_reward": 0.9571, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 191.228, "mean_ep_length": 200.0, "mean_collisions": 12.7, "mean_intent_rate": 0.71, "policy_loss": 0.014902, "value_loss": 146.484711, "entropy": 0.32511, "total_loss": 73.254006, "clip_frac": 0.216675, "sps": 438.9, "elapsed_s": 2593.6} |
| {"update": 554, "steps_done": 1134592, "mean_step_reward": 1.1032, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 224.465, "mean_ep_length": 200.0, "mean_collisions": 9.0, "mean_intent_rate": 0.703, "policy_loss": 0.015303, "value_loss": 104.80339, "entropy": 0.308437, "total_loss": 52.413913, "clip_frac": 0.164917, "sps": 440.5, "elapsed_s": 2598.3} |
| {"update": 555, "steps_done": 1136640, "mean_step_reward": 0.8082, "min_step_reward": -10.5389, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 160.169, "mean_ep_length": 200.0, "mean_collisions": 14.82, "mean_intent_rate": 0.692, "policy_loss": 0.004129, "value_loss": 182.610503, "entropy": 0.329361, "total_loss": 91.306085, "clip_frac": 0.11792, "sps": 443.8, "elapsed_s": 2602.9} |
| {"update": 556, "steps_done": 1138688, "mean_step_reward": 1.244, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 248.111, "mean_ep_length": 200.0, "mean_collisions": 6.7, "mean_intent_rate": 0.703, "policy_loss": 0.004226, "value_loss": 125.120653, "entropy": 0.302801, "total_loss": 62.561525, "clip_frac": 0.078247, "sps": 440.1, "elapsed_s": 2607.5} |
| {"update": 557, "steps_done": 1140736, "mean_step_reward": 0.9513, "min_step_reward": -10.5547, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 189.007, "mean_ep_length": 200.0, "mean_collisions": 10.2, "mean_intent_rate": 0.663, "policy_loss": 0.005399, "value_loss": 132.65027, "entropy": 0.276531, "total_loss": 66.327769, "clip_frac": 0.08252, "sps": 442.2, "elapsed_s": 2612.2} |
| {"update": 558, "steps_done": 1142784, "mean_step_reward": 0.9769, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 199.074, "mean_ep_length": 200.0, "mean_collisions": 8.8, "mean_intent_rate": 0.656, "policy_loss": 0.000267, "value_loss": 163.50427, "entropy": 0.304838, "total_loss": 81.749353, "clip_frac": 0.083374, "sps": 443.1, "elapsed_s": 2616.8} |
| {"update": 559, "steps_done": 1144832, "mean_step_reward": 1.1462, "min_step_reward": -10.5644, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 230.888, "mean_ep_length": 200.0, "mean_collisions": 7.5, "mean_intent_rate": 0.688, "policy_loss": 0.015645, "value_loss": 115.513969, "entropy": 0.280783, "total_loss": 57.769822, "clip_frac": 0.169678, "sps": 438.6, "elapsed_s": 2621.4} |
| {"update": 560, "steps_done": 1146880, "mean_step_reward": 1.1795, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 228.903, "mean_ep_length": 200.0, "mean_collisions": 8.82, "mean_intent_rate": 0.707, "policy_loss": 0.001753, "value_loss": 135.668969, "entropy": 0.264436, "total_loss": 67.833593, "clip_frac": 0.054321, "sps": 432.1, "elapsed_s": 2626.2} |
| {"update": 561, "steps_done": 1148928, "mean_step_reward": 0.5279, "min_step_reward": -10.5582, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 103.817, "mean_ep_length": 200.0, "mean_collisions": 20.2, "mean_intent_rate": 0.69, "policy_loss": -0.000252, "value_loss": 177.781204, "entropy": 0.299581, "total_loss": 88.887354, "clip_frac": 0.122192, "sps": 436.1, "elapsed_s": 2630.9} |
| {"update": 562, "steps_done": 1150976, "mean_step_reward": 0.9126, "min_step_reward": -10.585, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 184.652, "mean_ep_length": 200.0, "mean_collisions": 14.1, "mean_intent_rate": 0.721, "policy_loss": 0.005181, "value_loss": 173.823658, "entropy": 0.294434, "total_loss": 86.914065, "clip_frac": 0.077271, "sps": 443.1, "elapsed_s": 2635.5} |
| {"update": 563, "steps_done": 1153024, "mean_step_reward": 0.7309, "min_step_reward": -10.5868, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 159.119, "mean_ep_length": 200.0, "mean_collisions": 15.0, "mean_intent_rate": 0.695, "policy_loss": 0.002802, "value_loss": 103.903954, "entropy": 0.315488, "total_loss": 51.951624, "clip_frac": 0.147339, "sps": 440.4, "elapsed_s": 2640.2} |
| {"update": 564, "steps_done": 1155072, "mean_step_reward": 0.6385, "min_step_reward": -10.5753, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 118.597, "mean_ep_length": 200.0, "mean_collisions": 19.82, "mean_intent_rate": 0.708, "policy_loss": 0.007271, "value_loss": 167.196182, "entropy": 0.302408, "total_loss": 83.602338, "clip_frac": 0.10791, "sps": 442.1, "elapsed_s": 2644.8} |
| {"update": 565, "steps_done": 1157120, "mean_step_reward": 0.7194, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 144.283, "mean_ep_length": 200.0, "mean_collisions": 18.6, "mean_intent_rate": 0.73, "policy_loss": -0.000856, "value_loss": 152.024338, "entropy": 0.309526, "total_loss": 76.008217, "clip_frac": 0.153564, "sps": 439.3, "elapsed_s": 2649.5} |
| {"update": 566, "steps_done": 1159168, "mean_step_reward": 0.4843, "min_step_reward": -10.519, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 99.172, "mean_ep_length": 200.0, "mean_collisions": 20.1, "mean_intent_rate": 0.68, "policy_loss": 0.003651, "value_loss": 172.41409, "entropy": 0.32844, "total_loss": 86.207412, "clip_frac": 0.093872, "sps": 440.9, "elapsed_s": 2654.1} |
| {"update": 567, "steps_done": 1161216, "mean_step_reward": 0.612, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 117.153, "mean_ep_length": 200.0, "mean_collisions": 17.6, "mean_intent_rate": 0.669, "policy_loss": 0.001628, "value_loss": 132.345744, "entropy": 0.328602, "total_loss": 66.171214, "clip_frac": 0.11084, "sps": 439.8, "elapsed_s": 2658.8} |
| {"update": 568, "steps_done": 1163264, "mean_step_reward": 0.7099, "min_step_reward": -10.5888, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 140.868, "mean_ep_length": 200.0, "mean_collisions": 16.55, "mean_intent_rate": 0.69, "policy_loss": 0.026206, "value_loss": 145.727523, "entropy": 0.307121, "total_loss": 72.886897, "clip_frac": 0.199219, "sps": 440.1, "elapsed_s": 2663.4} |
| {"update": 569, "steps_done": 1165312, "mean_step_reward": 1.0101, "min_step_reward": -10.5781, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 207.216, "mean_ep_length": 200.0, "mean_collisions": 8.3, "mean_intent_rate": 0.661, "policy_loss": 0.019597, "value_loss": 116.673707, "entropy": 0.326962, "total_loss": 58.353181, "clip_frac": 0.177002, "sps": 429.2, "elapsed_s": 2668.2} |
| {"update": 570, "steps_done": 1167360, "mean_step_reward": 0.9854, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 196.36, "mean_ep_length": 200.0, "mean_collisions": 9.9, "mean_intent_rate": 0.672, "policy_loss": 0.005037, "value_loss": 113.978232, "entropy": 0.313907, "total_loss": 56.991014, "clip_frac": 0.108765, "sps": 441.4, "elapsed_s": 2672.8} |
| {"update": 571, "steps_done": 1169408, "mean_step_reward": 1.0514, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 206.867, "mean_ep_length": 200.0, "mean_collisions": 10.5, "mean_intent_rate": 0.7, "policy_loss": 0.012279, "value_loss": 112.511895, "entropy": 0.312309, "total_loss": 56.265103, "clip_frac": 0.188354, "sps": 436.4, "elapsed_s": 2677.5} |
| {"update": 572, "steps_done": 1171456, "mean_step_reward": 0.6279, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 133.768, "mean_ep_length": 200.0, "mean_collisions": 16.73, "mean_intent_rate": 0.682, "policy_loss": 0.012719, "value_loss": 135.751341, "entropy": 0.317198, "total_loss": 67.885218, "clip_frac": 0.166016, "sps": 437.3, "elapsed_s": 2682.2} |
| {"update": 573, "steps_done": 1173504, "mean_step_reward": 0.8087, "min_step_reward": -10.5417, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 154.267, "mean_ep_length": 200.0, "mean_collisions": 17.1, "mean_intent_rate": 0.721, "policy_loss": 0.025199, "value_loss": 177.328709, "entropy": 0.311421, "total_loss": 88.68644, "clip_frac": 0.138184, "sps": 438.3, "elapsed_s": 2686.9} |
| {"update": 574, "steps_done": 1175552, "mean_step_reward": 0.7976, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 170.339, "mean_ep_length": 200.0, "mean_collisions": 15.3, "mean_intent_rate": 0.718, "policy_loss": 0.004516, "value_loss": 141.744851, "entropy": 0.350379, "total_loss": 70.873438, "clip_frac": 0.221191, "sps": 439.6, "elapsed_s": 2691.5} |
| {"update": 575, "steps_done": 1177600, "mean_step_reward": 1.0242, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 194.177, "mean_ep_length": 200.0, "mean_collisions": 8.7, "mean_intent_rate": 0.646, "policy_loss": 0.005297, "value_loss": 111.365848, "entropy": 0.343536, "total_loss": 55.684785, "clip_frac": 0.106201, "sps": 435.3, "elapsed_s": 2696.2} |
| {"update": 576, "steps_done": 1179648, "mean_step_reward": 0.9746, "min_step_reward": -10.5001, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 193.185, "mean_ep_length": 193.1, "mean_collisions": 6.27, "mean_intent_rate": 0.619, "policy_loss": 0.010652, "value_loss": 121.580959, "entropy": 0.335859, "total_loss": 60.797773, "clip_frac": 0.120483, "sps": 440.7, "elapsed_s": 2700.9} |
| {"update": 577, "steps_done": 1181696, "mean_step_reward": 1.3333, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 267.747, "mean_ep_length": 200.0, "mean_collisions": 4.4, "mean_intent_rate": 0.697, "policy_loss": 0.007519, "value_loss": 137.160685, "entropy": 0.302298, "total_loss": 68.584839, "clip_frac": 0.101807, "sps": 447.8, "elapsed_s": 2705.5} |
| {"update": 578, "steps_done": 1183744, "mean_step_reward": 0.9405, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 190.971, "mean_ep_length": 200.0, "mean_collisions": 11.8, "mean_intent_rate": 0.694, "policy_loss": 0.004169, "value_loss": 166.634609, "entropy": 0.308262, "total_loss": 83.31839, "clip_frac": 0.090332, "sps": 430.1, "elapsed_s": 2710.2} |
| {"update": 579, "steps_done": 1185792, "mean_step_reward": 0.5137, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 104.74, "mean_ep_length": 200.0, "mean_collisions": 21.18, "mean_intent_rate": 0.708, "policy_loss": 0.008901, "value_loss": 175.757334, "entropy": 0.316242, "total_loss": 87.884405, "clip_frac": 0.123901, "sps": 444.2, "elapsed_s": 2714.8} |
| {"update": 580, "steps_done": 1187840, "mean_step_reward": 0.8418, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 166.726, "mean_ep_length": 200.0, "mean_collisions": 14.8, "mean_intent_rate": 0.705, "policy_loss": 0.005547, "value_loss": 152.391828, "entropy": 0.294031, "total_loss": 76.198521, "clip_frac": 0.11438, "sps": 435.1, "elapsed_s": 2719.5} |
| {"update": 581, "steps_done": 1189888, "mean_step_reward": 0.5943, "min_step_reward": -10.5293, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 128.187, "mean_ep_length": 200.0, "mean_collisions": 19.2, "mean_intent_rate": 0.713, "policy_loss": 0.004032, "value_loss": 175.260641, "entropy": 0.324881, "total_loss": 87.631103, "clip_frac": 0.092773, "sps": 447.3, "elapsed_s": 2724.1} |
| {"update": 582, "steps_done": 1191936, "mean_step_reward": 0.6232, "min_step_reward": -10.585, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 119.324, "mean_ep_length": 200.0, "mean_collisions": 18.8, "mean_intent_rate": 0.693, "policy_loss": 0.003185, "value_loss": 142.876165, "entropy": 0.321771, "total_loss": 71.43805, "clip_frac": 0.072754, "sps": 442.2, "elapsed_s": 2728.8} |
| {"update": 583, "steps_done": 1193984, "mean_step_reward": 0.9106, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 168.445, "mean_ep_length": 189.5, "mean_collisions": 11.91, "mean_intent_rate": 0.687, "policy_loss": 0.009497, "value_loss": 106.163325, "entropy": 0.309035, "total_loss": 53.088069, "clip_frac": 0.123657, "sps": 439.9, "elapsed_s": 2733.4} |
| {"update": 584, "steps_done": 1196032, "mean_step_reward": 0.9461, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 185.298, "mean_ep_length": 200.0, "mean_collisions": 13.8, "mean_intent_rate": 0.718, "policy_loss": 0.000922, "value_loss": 143.24487, "entropy": 0.290834, "total_loss": 71.620448, "clip_frac": 0.10791, "sps": 440.6, "elapsed_s": 2738.1} |
| {"update": 585, "steps_done": 1198080, "mean_step_reward": 0.3403, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 77.991, "mean_ep_length": 200.0, "mean_collisions": 23.36, "mean_intent_rate": 0.7, "policy_loss": 0.004401, "value_loss": 189.175308, "entropy": 0.336051, "total_loss": 94.588695, "clip_frac": 0.095459, "sps": 438.4, "elapsed_s": 2742.7} |
| {"update": 586, "steps_done": 1200128, "mean_step_reward": 0.6487, "min_step_reward": -10.5349, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 126.416, "mean_ep_length": 182.1, "mean_collisions": 13.09, "mean_intent_rate": 0.593, "policy_loss": 0.013028, "value_loss": 135.343965, "entropy": 0.332892, "total_loss": 67.681682, "clip_frac": 0.133545, "sps": 433.6, "elapsed_s": 2747.5} |
| {"update": 587, "steps_done": 1202176, "mean_step_reward": 1.1569, "min_step_reward": -10.584, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 219.802, "mean_ep_length": 200.0, "mean_collisions": 10.2, "mean_intent_rate": 0.716, "policy_loss": 0.009252, "value_loss": 114.187334, "entropy": 0.321846, "total_loss": 57.0997, "clip_frac": 0.189209, "sps": 448.7, "elapsed_s": 2752.0} |
| {"update": 588, "steps_done": 1204224, "mean_step_reward": 0.7185, "min_step_reward": -10.5474, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 156.052, "mean_ep_length": 200.0, "mean_collisions": 14.2, "mean_intent_rate": 0.675, "policy_loss": 0.008662, "value_loss": 118.139588, "entropy": 0.331595, "total_loss": 59.07514, "clip_frac": 0.145996, "sps": 441.1, "elapsed_s": 2756.7} |
| {"update": 589, "steps_done": 1206272, "mean_step_reward": 0.8898, "min_step_reward": -10.5804, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 169.654, "mean_ep_length": 200.0, "mean_collisions": 14.36, "mean_intent_rate": 0.701, "policy_loss": 0.004597, "value_loss": 153.87688, "entropy": 0.313174, "total_loss": 76.939906, "clip_frac": 0.140137, "sps": 441.9, "elapsed_s": 2761.3} |
| {"update": 590, "steps_done": 1208320, "mean_step_reward": 0.9206, "min_step_reward": -10.5863, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 182.789, "mean_ep_length": 200.0, "mean_collisions": 12.5, "mean_intent_rate": 0.693, "policy_loss": 0.004157, "value_loss": 179.614281, "entropy": 0.306518, "total_loss": 89.808232, "clip_frac": 0.091797, "sps": 442.3, "elapsed_s": 2765.9} |
| {"update": 591, "steps_done": 1210368, "mean_step_reward": 0.9292, "min_step_reward": -10.5626, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 186.859, "mean_ep_length": 200.0, "mean_collisions": 11.5, "mean_intent_rate": 0.68, "policy_loss": 0.017097, "value_loss": 176.492885, "entropy": 0.301709, "total_loss": 88.260524, "clip_frac": 0.203613, "sps": 444.4, "elapsed_s": 2770.5} |
| {"update": 592, "steps_done": 1212416, "mean_step_reward": 1.1587, "min_step_reward": -10.5402, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 235.83, "mean_ep_length": 200.0, "mean_collisions": 7.7, "mean_intent_rate": 0.699, "policy_loss": 0.024494, "value_loss": 118.105622, "entropy": 0.313875, "total_loss": 59.074167, "clip_frac": 0.20166, "sps": 435.9, "elapsed_s": 2775.2} |
| {"update": 593, "steps_done": 1214464, "mean_step_reward": 0.4188, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 88.968, "mean_ep_length": 200.0, "mean_collisions": 25.82, "mean_intent_rate": 0.759, "policy_loss": 0.004407, "value_loss": 237.537071, "entropy": 0.291021, "total_loss": 118.770033, "clip_frac": 0.059692, "sps": 439.3, "elapsed_s": 2779.9} |
| {"update": 594, "steps_done": 1216512, "mean_step_reward": 0.5938, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 113.038, "mean_ep_length": 200.0, "mean_collisions": 18.8, "mean_intent_rate": 0.68, "policy_loss": 0.000669, "value_loss": 153.256486, "entropy": 0.311559, "total_loss": 76.625796, "clip_frac": 0.09375, "sps": 437.3, "elapsed_s": 2784.6} |
| {"update": 595, "steps_done": 1218560, "mean_step_reward": 0.6827, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 143.792, "mean_ep_length": 200.0, "mean_collisions": 17.2, "mean_intent_rate": 0.706, "policy_loss": 0.001808, "value_loss": 131.622055, "entropy": 0.30908, "total_loss": 65.809745, "clip_frac": 0.116699, "sps": 429.1, "elapsed_s": 2789.4} |
| {"update": 596, "steps_done": 1220608, "mean_step_reward": 0.6505, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 124.515, "mean_ep_length": 200.0, "mean_collisions": 17.6, "mean_intent_rate": 0.681, "policy_loss": 0.010321, "value_loss": 164.119801, "entropy": 0.301933, "total_loss": 82.067203, "clip_frac": 0.138794, "sps": 439.2, "elapsed_s": 2794.0} |
| {"update": 597, "steps_done": 1222656, "mean_step_reward": 1.1569, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 227.513, "mean_ep_length": 200.0, "mean_collisions": 7.91, "mean_intent_rate": 0.69, "policy_loss": 0.004406, "value_loss": 111.080155, "entropy": 0.302724, "total_loss": 55.541456, "clip_frac": 0.126709, "sps": 441.5, "elapsed_s": 2798.7} |
| {"update": 598, "steps_done": 1224704, "mean_step_reward": 1.0233, "min_step_reward": -10.5403, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 206.795, "mean_ep_length": 200.0, "mean_collisions": 10.6, "mean_intent_rate": 0.698, "policy_loss": 0.008676, "value_loss": 139.264986, "entropy": 0.294292, "total_loss": 69.638227, "clip_frac": 0.138184, "sps": 435.9, "elapsed_s": 2803.4} |
| {"update": 599, "steps_done": 1226752, "mean_step_reward": 0.9355, "min_step_reward": -10.5422, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 178.16, "mean_ep_length": 200.0, "mean_collisions": 11.3, "mean_intent_rate": 0.664, "policy_loss": 0.002138, "value_loss": 134.321548, "entropy": 0.308081, "total_loss": 67.159831, "clip_frac": 0.105957, "sps": 440.7, "elapsed_s": 2808.0} |
| {"update": 600, "steps_done": 1228800, "mean_step_reward": 1.0408, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 212.404, "mean_ep_length": 200.0, "mean_collisions": 9.8, "mean_intent_rate": 0.696, "policy_loss": 0.004646, "value_loss": 131.615487, "entropy": 0.297271, "total_loss": 65.809416, "clip_frac": 0.09729, "sps": 442.7, "elapsed_s": 2812.6} |
| {"update": 601, "steps_done": 1230848, "mean_step_reward": 1.1258, "min_step_reward": -10.5654, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 226.295, "mean_ep_length": 200.0, "mean_collisions": 9.0, "mean_intent_rate": 0.706, "policy_loss": 0.001486, "value_loss": 125.029296, "entropy": 0.285503, "total_loss": 62.513279, "clip_frac": 0.091309, "sps": 444.2, "elapsed_s": 2817.3} |
| {"update": 602, "steps_done": 1232896, "mean_step_reward": 0.792, "min_step_reward": -10.5879, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 157.978, "mean_ep_length": 200.0, "mean_collisions": 15.7, "mean_intent_rate": 0.703, "policy_loss": 0.000289, "value_loss": 156.218584, "entropy": 0.290501, "total_loss": 78.106676, "clip_frac": 0.076416, "sps": 442.3, "elapsed_s": 2821.9} |
| {"update": 603, "steps_done": 1234944, "mean_step_reward": 1.1114, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 220.643, "mean_ep_length": 200.0, "mean_collisions": 9.1, "mean_intent_rate": 0.698, "policy_loss": 0.012524, "value_loss": 131.808707, "entropy": 0.290798, "total_loss": 65.913969, "clip_frac": 0.180176, "sps": 434.5, "elapsed_s": 2826.6} |
| {"update": 604, "steps_done": 1236992, "mean_step_reward": 0.9688, "min_step_reward": -10.5497, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 196.821, "mean_ep_length": 200.0, "mean_collisions": 7.3, "mean_intent_rate": 0.628, "policy_loss": 0.008173, "value_loss": 128.821002, "entropy": 0.280152, "total_loss": 64.415873, "clip_frac": 0.137329, "sps": 429.8, "elapsed_s": 2831.4} |
| {"update": 605, "steps_done": 1239040, "mean_step_reward": 0.8618, "min_step_reward": -10.5585, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 191.548, "mean_ep_length": 200.0, "mean_collisions": 11.8, "mean_intent_rate": 0.694, "policy_loss": 0.006793, "value_loss": 138.895848, "entropy": 0.291083, "total_loss": 69.451805, "clip_frac": 0.131592, "sps": 437.9, "elapsed_s": 2836.0} |
| {"update": 606, "steps_done": 1241088, "mean_step_reward": 0.9841, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 178.166, "mean_ep_length": 200.0, "mean_collisions": 10.0, "mean_intent_rate": 0.643, "policy_loss": 0.00779, "value_loss": 142.342073, "entropy": 0.299087, "total_loss": 71.175836, "clip_frac": 0.073486, "sps": 439.4, "elapsed_s": 2840.7} |
| {"update": 607, "steps_done": 1243136, "mean_step_reward": 1.1561, "min_step_reward": -10.5329, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 237.39, "mean_ep_length": 200.0, "mean_collisions": 5.6, "mean_intent_rate": 0.668, "policy_loss": 0.017798, "value_loss": 145.91955, "entropy": 0.314422, "total_loss": 72.974428, "clip_frac": 0.201416, "sps": 446.6, "elapsed_s": 2845.3} |
| {"update": 608, "steps_done": 1245184, "mean_step_reward": 1.1521, "min_step_reward": -10.5765, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 216.57, "mean_ep_length": 200.0, "mean_collisions": 8.1, "mean_intent_rate": 0.673, "policy_loss": 0.024975, "value_loss": 139.642843, "entropy": 0.302246, "total_loss": 69.843374, "clip_frac": 0.172485, "sps": 443.6, "elapsed_s": 2849.9} |
| {"update": 609, "steps_done": 1247232, "mean_step_reward": 0.9668, "min_step_reward": -10.5353, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 202.243, "mean_ep_length": 200.0, "mean_collisions": 10.1, "mean_intent_rate": 0.684, "policy_loss": 0.009465, "value_loss": 129.313632, "entropy": 0.29744, "total_loss": 64.663307, "clip_frac": 0.102661, "sps": 450.5, "elapsed_s": 2854.5} |
| {"update": 610, "steps_done": 1249280, "mean_step_reward": 1.156, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 227.481, "mean_ep_length": 200.0, "mean_collisions": 7.91, "mean_intent_rate": 0.69, "policy_loss": 0.007227, "value_loss": 140.550468, "entropy": 0.285549, "total_loss": 70.279606, "clip_frac": 0.133545, "sps": 442.8, "elapsed_s": 2859.1} |
| {"update": 611, "steps_done": 1251328, "mean_step_reward": 1.1244, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 228.071, "mean_ep_length": 200.0, "mean_collisions": 7.8, "mean_intent_rate": 0.688, "policy_loss": 0.00389, "value_loss": 136.76578, "entropy": 0.274734, "total_loss": 68.384033, "clip_frac": 0.084229, "sps": 437.7, "elapsed_s": 2863.8} |
| {"update": 612, "steps_done": 1253376, "mean_step_reward": 0.8929, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 172.515, "mean_ep_length": 200.0, "mean_collisions": 10.8, "mean_intent_rate": 0.646, "policy_loss": 0.004617, "value_loss": 184.305021, "entropy": 0.294157, "total_loss": 92.154187, "clip_frac": 0.083374, "sps": 445.4, "elapsed_s": 2868.4} |
| {"update": 613, "steps_done": 1255424, "mean_step_reward": 1.062, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 204.523, "mean_ep_length": 200.0, "mean_collisions": 7.0, "mean_intent_rate": 0.635, "policy_loss": 0.002716, "value_loss": 163.895377, "entropy": 0.288578, "total_loss": 81.947519, "clip_frac": 0.060547, "sps": 432.7, "elapsed_s": 2873.1} |
| {"update": 614, "steps_done": 1257472, "mean_step_reward": 1.0676, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 222.457, "mean_ep_length": 200.0, "mean_collisions": 8.18, "mean_intent_rate": 0.684, "policy_loss": 0.002378, "value_loss": 175.008379, "entropy": 0.285022, "total_loss": 87.503717, "clip_frac": 0.086792, "sps": 435.1, "elapsed_s": 2877.8} |
| {"update": 615, "steps_done": 1259520, "mean_step_reward": 0.8681, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 181.035, "mean_ep_length": 200.0, "mean_collisions": 10.7, "mean_intent_rate": 0.659, "policy_loss": 0.022556, "value_loss": 181.971367, "entropy": 0.288687, "total_loss": 91.005352, "clip_frac": 0.208984, "sps": 445.2, "elapsed_s": 2882.4} |
| {"update": 616, "steps_done": 1261568, "mean_step_reward": 0.922, "min_step_reward": -10.5678, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 175.589, "mean_ep_length": 200.0, "mean_collisions": 14.2, "mean_intent_rate": 0.708, "policy_loss": 0.039094, "value_loss": 180.851856, "entropy": 0.28155, "total_loss": 90.462207, "clip_frac": 0.286621, "sps": 437.5, "elapsed_s": 2887.1} |
| {"update": 617, "steps_done": 1263616, "mean_step_reward": 0.6424, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 136.133, "mean_ep_length": 200.0, "mean_collisions": 17.9, "mean_intent_rate": 0.705, "policy_loss": 0.00753, "value_loss": 206.654895, "entropy": 0.267499, "total_loss": 103.332304, "clip_frac": 0.124756, "sps": 441.8, "elapsed_s": 2891.7} |
| {"update": 618, "steps_done": 1265664, "mean_step_reward": 1.0159, "min_step_reward": -10.5554, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 200.353, "mean_ep_length": 200.0, "mean_collisions": 10.91, "mean_intent_rate": 0.696, "policy_loss": 0.013265, "value_loss": 163.577824, "entropy": 0.280431, "total_loss": 81.799373, "clip_frac": 0.178223, "sps": 440.3, "elapsed_s": 2896.4} |
| {"update": 619, "steps_done": 1267712, "mean_step_reward": 1.0712, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 207.988, "mean_ep_length": 200.0, "mean_collisions": 10.0, "mean_intent_rate": 0.69, "policy_loss": 0.01801, "value_loss": 134.801285, "entropy": 0.274959, "total_loss": 67.415902, "clip_frac": 0.170898, "sps": 437.9, "elapsed_s": 2901.1} |
| {"update": 620, "steps_done": 1269760, "mean_step_reward": 0.3238, "min_step_reward": -10.5801, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 62.695, "mean_ep_length": 200.0, "mean_collisions": 22.5, "mean_intent_rate": 0.66, "policy_loss": 0.004573, "value_loss": 176.353983, "entropy": 0.326258, "total_loss": 88.178302, "clip_frac": 0.134155, "sps": 442.0, "elapsed_s": 2905.7} |
| {"update": 621, "steps_done": 1271808, "mean_step_reward": 0.5253, "min_step_reward": -10.5854, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 112.906, "mean_ep_length": 200.0, "mean_collisions": 21.1, "mean_intent_rate": 0.721, "policy_loss": -0.000955, "value_loss": 168.684942, "entropy": 0.297265, "total_loss": 84.338543, "clip_frac": 0.144531, "sps": 448.0, "elapsed_s": 2910.3} |
| {"update": 622, "steps_done": 1273856, "mean_step_reward": 0.3476, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 70.875, "mean_ep_length": 200.0, "mean_collisions": 25.18, "mean_intent_rate": 0.718, "policy_loss": 0.014848, "value_loss": 221.021199, "entropy": 0.28618, "total_loss": 110.522586, "clip_frac": 0.149292, "sps": 431.4, "elapsed_s": 2915.0} |
| {"update": 623, "steps_done": 1275904, "mean_step_reward": 0.4102, "min_step_reward": -10.5778, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 84.203, "mean_ep_length": 200.0, "mean_collisions": 25.1, "mean_intent_rate": 0.738, "policy_loss": 0.01197, "value_loss": 159.228698, "entropy": 0.291094, "total_loss": 79.623407, "clip_frac": 0.149902, "sps": 439.2, "elapsed_s": 2919.7} |
| {"update": 624, "steps_done": 1277952, "mean_step_reward": 0.4737, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 92.979, "mean_ep_length": 200.0, "mean_collisions": 23.5, "mean_intent_rate": 0.726, "policy_loss": 0.018058, "value_loss": 131.948883, "entropy": 0.313279, "total_loss": 65.989366, "clip_frac": 0.179688, "sps": 435.5, "elapsed_s": 2924.4} |
| {"update": 625, "steps_done": 1280000, "mean_step_reward": 0.2743, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 50.79, "mean_ep_length": 200.0, "mean_collisions": 26.8, "mean_intent_rate": 0.711, "policy_loss": 0.004271, "value_loss": 258.096631, "entropy": 0.307618, "total_loss": 129.04951, "clip_frac": 0.072388, "sps": 436.8, "elapsed_s": 2929.1} |
| {"update": 626, "steps_done": 1282048, "mean_step_reward": 0.0518, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 18.176, "mean_ep_length": 200.0, "mean_collisions": 30.09, "mean_intent_rate": 0.711, "policy_loss": 0.053666, "value_loss": 293.332564, "entropy": 0.313145, "total_loss": 146.716817, "clip_frac": 0.372803, "sps": 442.4, "elapsed_s": 2933.7} |
| {"update": 627, "steps_done": 1284096, "mean_step_reward": 0.61, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 117.246, "mean_ep_length": 200.0, "mean_collisions": 18.3, "mean_intent_rate": 0.68, "policy_loss": 0.007297, "value_loss": 111.453926, "entropy": 0.338361, "total_loss": 55.730876, "clip_frac": 0.079834, "sps": 437.2, "elapsed_s": 2938.4} |
| {"update": 628, "steps_done": 1286144, "mean_step_reward": 0.7353, "min_step_reward": -10.5481, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 149.458, "mean_ep_length": 200.0, "mean_collisions": 15.1, "mean_intent_rate": 0.68, "policy_loss": 0.002607, "value_loss": 78.537597, "entropy": 0.342206, "total_loss": 39.267984, "clip_frac": 0.096558, "sps": 442.6, "elapsed_s": 2943.0} |
| {"update": 629, "steps_done": 1288192, "mean_step_reward": 0.9614, "min_step_reward": -10.5826, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 199.772, "mean_ep_length": 200.0, "mean_collisions": 12.2, "mean_intent_rate": 0.714, "policy_loss": 0.004557, "value_loss": 93.428838, "entropy": 0.336718, "total_loss": 46.715609, "clip_frac": 0.145386, "sps": 439.3, "elapsed_s": 2947.7} |
| {"update": 630, "steps_done": 1290240, "mean_step_reward": 0.9224, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 177.934, "mean_ep_length": 195.7, "mean_collisions": 12.82, "mean_intent_rate": 0.7, "policy_loss": 0.003471, "value_loss": 100.07309, "entropy": 0.329283, "total_loss": 50.036723, "clip_frac": 0.083252, "sps": 437.1, "elapsed_s": 2952.4} |
| {"update": 631, "steps_done": 1292288, "mean_step_reward": 1.2193, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 242.196, "mean_ep_length": 200.0, "mean_collisions": 6.9, "mean_intent_rate": 0.699, "policy_loss": 0.017905, "value_loss": 90.633186, "entropy": 0.300024, "total_loss": 45.331498, "clip_frac": 0.188965, "sps": 431.5, "elapsed_s": 2957.1} |
| {"update": 632, "steps_done": 1294336, "mean_step_reward": 0.8754, "min_step_reward": -10.5384, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 166.502, "mean_ep_length": 200.0, "mean_collisions": 13.5, "mean_intent_rate": 0.682, "policy_loss": 0.003464, "value_loss": 134.270196, "entropy": 0.292354, "total_loss": 67.135639, "clip_frac": 0.08313, "sps": 436.4, "elapsed_s": 2961.8} |
| {"update": 633, "steps_done": 1296384, "mean_step_reward": 0.9641, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 196.69, "mean_ep_length": 200.0, "mean_collisions": 10.1, "mean_intent_rate": 0.676, "policy_loss": 0.009361, "value_loss": 114.096738, "entropy": 0.305053, "total_loss": 57.054678, "clip_frac": 0.083618, "sps": 436.9, "elapsed_s": 2966.5} |
| {"update": 634, "steps_done": 1298432, "mean_step_reward": 1.0268, "min_step_reward": -10.5596, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 211.247, "mean_ep_length": 200.0, "mean_collisions": 11.64, "mean_intent_rate": 0.726, "policy_loss": 0.008884, "value_loss": 132.025393, "entropy": 0.282304, "total_loss": 66.018757, "clip_frac": 0.09436, "sps": 440.8, "elapsed_s": 2971.1} |
| {"update": 635, "steps_done": 1300480, "mean_step_reward": 0.8173, "min_step_reward": -10.5039, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 154.796, "mean_ep_length": 200.0, "mean_collisions": 13.4, "mean_intent_rate": 0.66, "policy_loss": 0.014864, "value_loss": 123.641586, "entropy": 0.283119, "total_loss": 61.832826, "clip_frac": 0.182617, "sps": 440.1, "elapsed_s": 2975.8} |
| {"update": 636, "steps_done": 1302528, "mean_step_reward": 0.9166, "min_step_reward": -10.5573, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 184.296, "mean_ep_length": 200.0, "mean_collisions": 14.3, "mean_intent_rate": 0.724, "policy_loss": 0.008636, "value_loss": 131.415244, "entropy": 0.274727, "total_loss": 65.713511, "clip_frac": 0.14917, "sps": 435.8, "elapsed_s": 2980.5} |
| {"update": 637, "steps_done": 1304576, "mean_step_reward": 0.8272, "min_step_reward": -10.5851, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 161.746, "mean_ep_length": 200.0, "mean_collisions": 12.8, "mean_intent_rate": 0.662, "policy_loss": 0.012855, "value_loss": 146.454459, "entropy": 0.303484, "total_loss": 73.23705, "clip_frac": 0.154175, "sps": 438.5, "elapsed_s": 2985.2} |
| {"update": 638, "steps_done": 1306624, "mean_step_reward": 1.0222, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 209.757, "mean_ep_length": 200.0, "mean_collisions": 11.45, "mean_intent_rate": 0.72, "policy_loss": 0.013807, "value_loss": 157.504269, "entropy": 0.287905, "total_loss": 78.763063, "clip_frac": 0.148804, "sps": 439.0, "elapsed_s": 2989.8} |
| {"update": 639, "steps_done": 1308672, "mean_step_reward": 0.8138, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 159.657, "mean_ep_length": 200.0, "mean_collisions": 14.9, "mean_intent_rate": 0.694, "policy_loss": 0.00854, "value_loss": 133.769291, "entropy": 0.300849, "total_loss": 66.890177, "clip_frac": 0.160767, "sps": 437.6, "elapsed_s": 2994.5} |
| {"update": 640, "steps_done": 1310720, "mean_step_reward": 0.811, "min_step_reward": -10.5838, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 170.879, "mean_ep_length": 200.0, "mean_collisions": 16.9, "mean_intent_rate": 0.747, "policy_loss": 0.006823, "value_loss": 189.192907, "entropy": 0.275519, "total_loss": 94.600522, "clip_frac": 0.119873, "sps": 432.3, "elapsed_s": 2999.2} |
| {"update": 641, "steps_done": 1312768, "mean_step_reward": 1.1897, "min_step_reward": -10.584, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 234.074, "mean_ep_length": 200.0, "mean_collisions": 8.9, "mean_intent_rate": 0.718, "policy_loss": 0.01598, "value_loss": 144.531565, "entropy": 0.295275, "total_loss": 72.27881, "clip_frac": 0.161377, "sps": 440.2, "elapsed_s": 3003.9} |
| {"update": 642, "steps_done": 1314816, "mean_step_reward": 0.8807, "min_step_reward": -10.5868, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 172.282, "mean_ep_length": 200.0, "mean_collisions": 13.09, "mean_intent_rate": 0.685, "policy_loss": 0.011163, "value_loss": 134.657749, "entropy": 0.305447, "total_loss": 67.336983, "clip_frac": 0.159912, "sps": 437.0, "elapsed_s": 3008.6} |
| {"update": 643, "steps_done": 1316864, "mean_step_reward": 0.9252, "min_step_reward": -10.5348, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 192.827, "mean_ep_length": 200.0, "mean_collisions": 13.1, "mean_intent_rate": 0.719, "policy_loss": 0.008145, "value_loss": 153.432976, "entropy": 0.291198, "total_loss": 76.721721, "clip_frac": 0.115234, "sps": 439.9, "elapsed_s": 3013.2} |
| {"update": 644, "steps_done": 1318912, "mean_step_reward": 0.9504, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 173.366, "mean_ep_length": 189.5, "mean_collisions": 11.09, "mean_intent_rate": 0.675, "policy_loss": 0.004196, "value_loss": 128.378734, "entropy": 0.297042, "total_loss": 64.190592, "clip_frac": 0.089233, "sps": 436.9, "elapsed_s": 3017.9} |
| {"update": 645, "steps_done": 1320960, "mean_step_reward": 1.0809, "min_step_reward": -10.5262, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 216.213, "mean_ep_length": 200.0, "mean_collisions": 8.3, "mean_intent_rate": 0.677, "policy_loss": 0.003403, "value_loss": 122.054663, "entropy": 0.274109, "total_loss": 61.027993, "clip_frac": 0.127808, "sps": 441.0, "elapsed_s": 3022.6} |
| {"update": 646, "steps_done": 1323008, "mean_step_reward": 0.9322, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 176.941, "mean_ep_length": 200.0, "mean_collisions": 13.7, "mean_intent_rate": 0.702, "policy_loss": 0.003958, "value_loss": 146.243985, "entropy": 0.279382, "total_loss": 73.123158, "clip_frac": 0.098511, "sps": 435.8, "elapsed_s": 3027.3} |
| {"update": 647, "steps_done": 1325056, "mean_step_reward": 1.0046, "min_step_reward": -10.5746, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 202.532, "mean_ep_length": 200.0, "mean_collisions": 8.0, "mean_intent_rate": 0.65, "policy_loss": 0.004652, "value_loss": 114.145441, "entropy": 0.284058, "total_loss": 57.074532, "clip_frac": 0.075806, "sps": 440.5, "elapsed_s": 3031.9} |
| {"update": 648, "steps_done": 1327104, "mean_step_reward": 1.0643, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 219.921, "mean_ep_length": 200.0, "mean_collisions": 9.45, "mean_intent_rate": 0.703, "policy_loss": 0.005436, "value_loss": 162.731448, "entropy": 0.268726, "total_loss": 81.368473, "clip_frac": 0.125854, "sps": 429.1, "elapsed_s": 3036.7} |
| {"update": 649, "steps_done": 1329152, "mean_step_reward": 0.9201, "min_step_reward": -10.5587, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 183.657, "mean_ep_length": 200.0, "mean_collisions": 13.8, "mean_intent_rate": 0.715, "policy_loss": 0.008401, "value_loss": 184.504783, "entropy": 0.26494, "total_loss": 92.258143, "clip_frac": 0.094727, "sps": 440.5, "elapsed_s": 3041.3} |
| {"update": 650, "steps_done": 1331200, "mean_step_reward": 0.9241, "min_step_reward": -10.5791, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 180.158, "mean_ep_length": 200.0, "mean_collisions": 14.2, "mean_intent_rate": 0.716, "policy_loss": 0.022879, "value_loss": 152.508606, "entropy": 0.28839, "total_loss": 76.274298, "clip_frac": 0.228638, "sps": 435.5, "elapsed_s": 3046.1} |
| {"update": 651, "steps_done": 1333248, "mean_step_reward": 0.6224, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 149.515, "mean_ep_length": 200.0, "mean_collisions": 15.8, "mean_intent_rate": 0.693, "policy_loss": 0.007012, "value_loss": 167.117568, "entropy": 0.290475, "total_loss": 83.562891, "clip_frac": 0.122192, "sps": 432.9, "elapsed_s": 3050.8} |
| {"update": 652, "steps_done": 1335296, "mean_step_reward": 0.7189, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 126.568, "mean_ep_length": 200.0, "mean_collisions": 18.64, "mean_intent_rate": 0.703, "policy_loss": -0.002654, "value_loss": 160.80384, "entropy": 0.282607, "total_loss": 80.39644, "clip_frac": 0.055054, "sps": 438.1, "elapsed_s": 3055.5} |
| {"update": 653, "steps_done": 1337344, "mean_step_reward": 0.9402, "min_step_reward": -10.584, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 187.979, "mean_ep_length": 200.0, "mean_collisions": 12.2, "mean_intent_rate": 0.696, "policy_loss": 0.012561, "value_loss": 129.912704, "entropy": 0.296024, "total_loss": 64.965952, "clip_frac": 0.129761, "sps": 441.3, "elapsed_s": 3060.1} |
| {"update": 654, "steps_done": 1339392, "mean_step_reward": 0.8803, "min_step_reward": -10.5607, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 188.156, "mean_ep_length": 200.0, "mean_collisions": 13.0, "mean_intent_rate": 0.709, "policy_loss": 0.014401, "value_loss": 150.7711, "entropy": 0.289191, "total_loss": 75.397059, "clip_frac": 0.143433, "sps": 442.4, "elapsed_s": 3064.7} |
| {"update": 655, "steps_done": 1341440, "mean_step_reward": 0.2616, "min_step_reward": -10.553, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 49.255, "mean_ep_length": 194.7, "mean_collisions": 27.91, "mean_intent_rate": 0.741, "policy_loss": 0.007841, "value_loss": 244.27128, "entropy": 0.309004, "total_loss": 122.140392, "clip_frac": 0.18042, "sps": 437.6, "elapsed_s": 3069.4} |
| {"update": 656, "steps_done": 1343488, "mean_step_reward": 0.7395, "min_step_reward": -10.5509, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 143.173, "mean_ep_length": 200.0, "mean_collisions": 15.5, "mean_intent_rate": 0.677, "policy_loss": 0.000264, "value_loss": 124.744486, "entropy": 0.307161, "total_loss": 62.369436, "clip_frac": 0.123169, "sps": 441.1, "elapsed_s": 3074.1} |
| {"update": 657, "steps_done": 1345536, "mean_step_reward": 0.8283, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 165.158, "mean_ep_length": 200.0, "mean_collisions": 15.3, "mean_intent_rate": 0.711, "policy_loss": 0.014068, "value_loss": 128.264362, "entropy": 0.321042, "total_loss": 64.143038, "clip_frac": 0.182495, "sps": 423.4, "elapsed_s": 3078.9} |
| {"update": 658, "steps_done": 1347584, "mean_step_reward": 0.703, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 144.302, "mean_ep_length": 200.0, "mean_collisions": 16.2, "mean_intent_rate": 0.69, "policy_loss": 0.009867, "value_loss": 149.272171, "entropy": 0.31044, "total_loss": 74.642848, "clip_frac": 0.137329, "sps": 439.0, "elapsed_s": 3083.6} |
| {"update": 659, "steps_done": 1349632, "mean_step_reward": 0.4529, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 94.368, "mean_ep_length": 200.0, "mean_collisions": 21.0, "mean_intent_rate": 0.686, "policy_loss": 0.003893, "value_loss": 190.993732, "entropy": 0.333441, "total_loss": 95.497425, "clip_frac": 0.103027, "sps": 438.1, "elapsed_s": 3088.3} |
| {"update": 660, "steps_done": 1351680, "mean_step_reward": 0.9405, "min_step_reward": -10.5668, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 185.915, "mean_ep_length": 200.0, "mean_collisions": 12.5, "mean_intent_rate": 0.698, "policy_loss": 0.003598, "value_loss": 114.048804, "entropy": 0.29575, "total_loss": 57.025043, "clip_frac": 0.104736, "sps": 439.7, "elapsed_s": 3092.9} |
| {"update": 661, "steps_done": 1353728, "mean_step_reward": 0.2755, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 58.7, "mean_ep_length": 200.0, "mean_collisions": 25.4, "mean_intent_rate": 0.701, "policy_loss": -0.000571, "value_loss": 175.516064, "entropy": 0.326102, "total_loss": 87.7542, "clip_frac": 0.141357, "sps": 437.6, "elapsed_s": 3097.6} |
| {"update": 662, "steps_done": 1355776, "mean_step_reward": 0.9913, "min_step_reward": -10.5855, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 180.835, "mean_ep_length": 189.8, "mean_collisions": 11.64, "mean_intent_rate": 0.701, "policy_loss": 0.017324, "value_loss": 118.287603, "entropy": 0.335797, "total_loss": 59.157768, "clip_frac": 0.163208, "sps": 441.5, "elapsed_s": 3102.2} |
| {"update": 663, "steps_done": 1357824, "mean_step_reward": 0.5271, "min_step_reward": -10.5801, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 101.938, "mean_ep_length": 200.0, "mean_collisions": 19.0, "mean_intent_rate": 0.666, "policy_loss": 0.016023, "value_loss": 151.26222, "entropy": 0.326187, "total_loss": 75.643871, "clip_frac": 0.170654, "sps": 440.9, "elapsed_s": 3106.9} |
| {"update": 664, "steps_done": 1359872, "mean_step_reward": 1.1308, "min_step_reward": -10.5318, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 220.809, "mean_ep_length": 200.0, "mean_collisions": 8.5, "mean_intent_rate": 0.689, "policy_loss": 0.014579, "value_loss": 107.599705, "entropy": 0.297388, "total_loss": 53.811457, "clip_frac": 0.16333, "sps": 440.9, "elapsed_s": 3111.5} |
| {"update": 665, "steps_done": 1361920, "mean_step_reward": 0.6051, "min_step_reward": -10.5435, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 133.118, "mean_ep_length": 199.1, "mean_collisions": 18.09, "mean_intent_rate": 0.705, "policy_loss": 0.004465, "value_loss": 146.291221, "entropy": 0.312139, "total_loss": 73.146955, "clip_frac": 0.150513, "sps": 433.4, "elapsed_s": 3116.2} |
| {"update": 666, "steps_done": 1363968, "mean_step_reward": 0.5917, "min_step_reward": -10.58, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 118.059, "mean_ep_length": 197.8, "mean_collisions": 18.0, "mean_intent_rate": 0.682, "policy_loss": 0.011142, "value_loss": 153.328328, "entropy": 0.345953, "total_loss": 76.671846, "clip_frac": 0.179565, "sps": 425.5, "elapsed_s": 3121.1} |
| {"update": 667, "steps_done": 1366016, "mean_step_reward": 0.9904, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 193.232, "mean_ep_length": 200.0, "mean_collisions": 11.9, "mean_intent_rate": 0.697, "policy_loss": 0.028803, "value_loss": 164.360932, "entropy": 0.315186, "total_loss": 82.206117, "clip_frac": 0.230103, "sps": 438.3, "elapsed_s": 3125.7} |
| {"update": 668, "steps_done": 1368064, "mean_step_reward": 1.1714, "min_step_reward": -10.5762, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 239.803, "mean_ep_length": 200.0, "mean_collisions": 7.7, "mean_intent_rate": 0.706, "policy_loss": 0.0029, "value_loss": 117.841653, "entropy": 0.295446, "total_loss": 58.920772, "clip_frac": 0.087036, "sps": 440.0, "elapsed_s": 3130.4} |
| {"update": 669, "steps_done": 1370112, "mean_step_reward": 1.049, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 206.925, "mean_ep_length": 200.0, "mean_collisions": 12.18, "mean_intent_rate": 0.725, "policy_loss": 0.014757, "value_loss": 161.485703, "entropy": 0.28895, "total_loss": 80.75472, "clip_frac": 0.130493, "sps": 434.6, "elapsed_s": 3135.1} |
| {"update": 670, "steps_done": 1372160, "mean_step_reward": 0.6655, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 134.661, "mean_ep_length": 200.0, "mean_collisions": 19.0, "mean_intent_rate": 0.722, "policy_loss": -0.001293, "value_loss": 171.9718, "entropy": 0.292895, "total_loss": 85.981678, "clip_frac": 0.140381, "sps": 440.3, "elapsed_s": 3139.8} |
| {"update": 671, "steps_done": 1374208, "mean_step_reward": 0.7895, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 153.831, "mean_ep_length": 200.0, "mean_collisions": 13.5, "mean_intent_rate": 0.661, "policy_loss": 0.00307, "value_loss": 133.934653, "entropy": 0.29709, "total_loss": 66.967425, "clip_frac": 0.122681, "sps": 445.6, "elapsed_s": 3144.4} |
| {"update": 672, "steps_done": 1376256, "mean_step_reward": 0.6245, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 134.916, "mean_ep_length": 200.0, "mean_collisions": 19.7, "mean_intent_rate": 0.731, "policy_loss": 0.008718, "value_loss": 165.666293, "entropy": 0.286087, "total_loss": 82.839002, "clip_frac": 0.093994, "sps": 437.0, "elapsed_s": 3149.0} |
| {"update": 673, "steps_done": 1378304, "mean_step_reward": 0.7354, "min_step_reward": -10.5701, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 142.964, "mean_ep_length": 200.0, "mean_collisions": 15.55, "mean_intent_rate": 0.677, "policy_loss": 0.002513, "value_loss": 161.406519, "entropy": 0.317693, "total_loss": 80.702595, "clip_frac": 0.169678, "sps": 443.8, "elapsed_s": 3153.7} |
| {"update": 674, "steps_done": 1380352, "mean_step_reward": 0.8376, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 161.692, "mean_ep_length": 200.0, "mean_collisions": 16.7, "mean_intent_rate": 0.728, "policy_loss": 0.004138, "value_loss": 141.013753, "entropy": 0.29656, "total_loss": 70.508048, "clip_frac": 0.096802, "sps": 440.2, "elapsed_s": 3158.3} |
| {"update": 675, "steps_done": 1382400, "mean_step_reward": 0.9711, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 179.995, "mean_ep_length": 184.5, "mean_collisions": 10.18, "mean_intent_rate": 0.698, "policy_loss": 0.009982, "value_loss": 101.096276, "entropy": 0.309161, "total_loss": 50.555028, "clip_frac": 0.11145, "sps": 431.6, "elapsed_s": 3163.1} |
| {"update": 676, "steps_done": 1384448, "mean_step_reward": 0.6147, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 139.412, "mean_ep_length": 200.0, "mean_collisions": 17.2, "mean_intent_rate": 0.698, "policy_loss": 0.006935, "value_loss": 130.627495, "entropy": 0.300744, "total_loss": 65.317676, "clip_frac": 0.095215, "sps": 439.5, "elapsed_s": 3167.7} |
| {"update": 677, "steps_done": 1386496, "mean_step_reward": 0.1996, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 26.735, "mean_ep_length": 200.0, "mean_collisions": 28.5, "mean_intent_rate": 0.699, "policy_loss": 0.000401, "value_loss": 179.007368, "entropy": 0.313187, "total_loss": 89.500953, "clip_frac": 0.085205, "sps": 438.5, "elapsed_s": 3172.4} |
| {"update": 678, "steps_done": 1388544, "mean_step_reward": 0.3309, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 68.039, "mean_ep_length": 200.0, "mean_collisions": 24.73, "mean_intent_rate": 0.706, "policy_loss": 0.018426, "value_loss": 143.442979, "entropy": 0.340895, "total_loss": 71.736506, "clip_frac": 0.212646, "sps": 440.0, "elapsed_s": 3177.0} |
| {"update": 679, "steps_done": 1390592, "mean_step_reward": 0.3969, "min_step_reward": -10.5883, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 74.804, "mean_ep_length": 200.0, "mean_collisions": 23.9, "mean_intent_rate": 0.703, "policy_loss": 0.002703, "value_loss": 199.285383, "entropy": 0.305826, "total_loss": 99.642337, "clip_frac": 0.081299, "sps": 437.9, "elapsed_s": 3181.7} |
| {"update": 680, "steps_done": 1392640, "mean_step_reward": 0.3187, "min_step_reward": -10.5612, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 65.546, "mean_ep_length": 200.0, "mean_collisions": 24.4, "mean_intent_rate": 0.696, "policy_loss": 0.003281, "value_loss": 186.782493, "entropy": 0.305621, "total_loss": 93.391472, "clip_frac": 0.09082, "sps": 441.1, "elapsed_s": 3186.4} |
| {"update": 681, "steps_done": 1394688, "mean_step_reward": -0.2431, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": -34.291, "mean_ep_length": 200.0, "mean_collisions": 36.0, "mean_intent_rate": 0.724, "policy_loss": 0.010664, "value_loss": 261.960555, "entropy": 0.311197, "total_loss": 130.98783, "clip_frac": 0.109497, "sps": 436.6, "elapsed_s": 3191.1} |
| {"update": 682, "steps_done": 1396736, "mean_step_reward": 0.2069, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 31.904, "mean_ep_length": 200.0, "mean_collisions": 29.27, "mean_intent_rate": 0.722, "policy_loss": 0.006249, "value_loss": 249.411366, "entropy": 0.316958, "total_loss": 124.708764, "clip_frac": 0.063965, "sps": 441.0, "elapsed_s": 3195.7} |
| {"update": 683, "steps_done": 1398784, "mean_step_reward": 0.5177, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 99.101, "mean_ep_length": 200.0, "mean_collisions": 22.4, "mean_intent_rate": 0.719, "policy_loss": 0.002749, "value_loss": 217.491697, "entropy": 0.31582, "total_loss": 108.74544, "clip_frac": 0.103516, "sps": 439.5, "elapsed_s": 3200.4} |
| {"update": 684, "steps_done": 1400832, "mean_step_reward": 0.1989, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 36.062, "mean_ep_length": 200.0, "mean_collisions": 27.1, "mean_intent_rate": 0.693, "policy_loss": 0.00204, "value_loss": 211.188386, "entropy": 0.329406, "total_loss": 105.59294, "clip_frac": 0.075928, "sps": 431.8, "elapsed_s": 3205.1} |
| {"update": 685, "steps_done": 1402880, "mean_step_reward": 0.5406, "min_step_reward": -10.5744, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 109.268, "mean_ep_length": 200.0, "mean_collisions": 22.2, "mean_intent_rate": 0.732, "policy_loss": 0.004224, "value_loss": 160.519265, "entropy": 0.316784, "total_loss": 80.260689, "clip_frac": 0.067017, "sps": 437.1, "elapsed_s": 3209.8} |
| {"update": 686, "steps_done": 1404928, "mean_step_reward": 0.3402, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 74.312, "mean_ep_length": 200.0, "mean_collisions": 25.36, "mean_intent_rate": 0.727, "policy_loss": 0.01043, "value_loss": 170.653486, "entropy": 0.3408, "total_loss": 85.333765, "clip_frac": 0.176636, "sps": 435.0, "elapsed_s": 3214.5} |
| {"update": 687, "steps_done": 1406976, "mean_step_reward": 0.8919, "min_step_reward": -10.5357, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 175.065, "mean_ep_length": 200.0, "mean_collisions": 13.8, "mean_intent_rate": 0.701, "policy_loss": 0.003825, "value_loss": 127.475389, "entropy": 0.337792, "total_loss": 63.738142, "clip_frac": 0.103149, "sps": 432.0, "elapsed_s": 3219.2} |
| {"update": 688, "steps_done": 1409024, "mean_step_reward": 1.0723, "min_step_reward": -10.5594, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 213.608, "mean_ep_length": 200.0, "mean_collisions": 10.4, "mean_intent_rate": 0.706, "policy_loss": 0.003585, "value_loss": 135.504073, "entropy": 0.321913, "total_loss": 67.752402, "clip_frac": 0.110352, "sps": 437.5, "elapsed_s": 3223.9} |
| {"update": 689, "steps_done": 1411072, "mean_step_reward": 1.1984, "min_step_reward": -10.5809, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 238.066, "mean_ep_length": 200.0, "mean_collisions": 7.4, "mean_intent_rate": 0.698, "policy_loss": 0.008821, "value_loss": 144.457299, "entropy": 0.309721, "total_loss": 72.234372, "clip_frac": 0.118286, "sps": 447.4, "elapsed_s": 3228.5} |
| {"update": 690, "steps_done": 1413120, "mean_step_reward": 1.1076, "min_step_reward": -10.5295, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 225.374, "mean_ep_length": 200.0, "mean_collisions": 8.45, "mean_intent_rate": 0.695, "policy_loss": 0.022467, "value_loss": 138.545645, "entropy": 0.306311, "total_loss": 69.292225, "clip_frac": 0.196411, "sps": 437.8, "elapsed_s": 3233.2} |
| {"update": 691, "steps_done": 1415168, "mean_step_reward": 1.1635, "min_step_reward": -10.5792, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 231.719, "mean_ep_length": 200.0, "mean_collisions": 6.3, "mean_intent_rate": 0.669, "policy_loss": 0.008216, "value_loss": 97.719324, "entropy": 0.299495, "total_loss": 48.864883, "clip_frac": 0.099487, "sps": 437.4, "elapsed_s": 3237.9} |
| {"update": 692, "steps_done": 1417216, "mean_step_reward": 0.9989, "min_step_reward": -10.5327, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 198.869, "mean_ep_length": 200.0, "mean_collisions": 8.9, "mean_intent_rate": 0.659, "policy_loss": 0.007912, "value_loss": 147.812575, "entropy": 0.293887, "total_loss": 73.911261, "clip_frac": 0.111572, "sps": 441.3, "elapsed_s": 3242.5} |
| {"update": 693, "steps_done": 1419264, "mean_step_reward": 1.166, "min_step_reward": -10.5409, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 238.833, "mean_ep_length": 200.0, "mean_collisions": 5.2, "mean_intent_rate": 0.663, "policy_loss": 0.021079, "value_loss": 133.332134, "entropy": 0.283665, "total_loss": 66.684309, "clip_frac": 0.179443, "sps": 427.4, "elapsed_s": 3247.3} |
| {"update": 694, "steps_done": 1421312, "mean_step_reward": 1.1825, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 233.093, "mean_ep_length": 200.0, "mean_collisions": 6.55, "mean_intent_rate": 0.676, "policy_loss": 0.004461, "value_loss": 138.834098, "entropy": 0.282703, "total_loss": 69.418683, "clip_frac": 0.074707, "sps": 442.7, "elapsed_s": 3251.9} |
| {"update": 695, "steps_done": 1423360, "mean_step_reward": 1.1801, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 232.938, "mean_ep_length": 200.0, "mean_collisions": 5.5, "mean_intent_rate": 0.66, "policy_loss": 0.002437, "value_loss": 117.953324, "entropy": 0.277143, "total_loss": 58.976327, "clip_frac": 0.066772, "sps": 437.4, "elapsed_s": 3256.6} |
| {"update": 696, "steps_done": 1425408, "mean_step_reward": 0.9757, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 225.739, "mean_ep_length": 200.0, "mean_collisions": 8.9, "mean_intent_rate": 0.705, "policy_loss": 0.007404, "value_loss": 158.667158, "entropy": 0.283501, "total_loss": 79.338148, "clip_frac": 0.165039, "sps": 440.7, "elapsed_s": 3261.3} |
| {"update": 697, "steps_done": 1427456, "mean_step_reward": 0.9764, "min_step_reward": -10.5594, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 166.526, "mean_ep_length": 200.0, "mean_collisions": 14.9, "mean_intent_rate": 0.705, "policy_loss": 0.002136, "value_loss": 135.767109, "entropy": 0.276759, "total_loss": 67.882924, "clip_frac": 0.088379, "sps": 440.8, "elapsed_s": 3265.9} |
| {"update": 698, "steps_done": 1429504, "mean_step_reward": 0.6313, "min_step_reward": -10.5631, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 144.785, "mean_ep_length": 200.0, "mean_collisions": 16.8, "mean_intent_rate": 0.7, "policy_loss": 0.015661, "value_loss": 198.264579, "entropy": 0.287046, "total_loss": 99.14508, "clip_frac": 0.176392, "sps": 435.8, "elapsed_s": 3270.6} |
| {"update": 699, "steps_done": 1431552, "mean_step_reward": 0.6557, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 119.61, "mean_ep_length": 200.0, "mean_collisions": 20.09, "mean_intent_rate": 0.715, "policy_loss": 0.005464, "value_loss": 156.050694, "entropy": 0.295074, "total_loss": 78.027861, "clip_frac": 0.114502, "sps": 439.6, "elapsed_s": 3275.3} |
| {"update": 700, "steps_done": 1433600, "mean_step_reward": 0.6566, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 131.807, "mean_ep_length": 200.0, "mean_collisions": 18.4, "mean_intent_rate": 0.706, "policy_loss": 0.00739, "value_loss": 153.722335, "entropy": 0.305766, "total_loss": 76.865499, "clip_frac": 0.098633, "sps": 448.7, "elapsed_s": 3279.8} |
| {"update": 701, "steps_done": 1435648, "mean_step_reward": 0.8321, "min_step_reward": -10.585, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 165.288, "mean_ep_length": 200.0, "mean_collisions": 14.5, "mean_intent_rate": 0.696, "policy_loss": 0.014684, "value_loss": 114.243373, "entropy": 0.297843, "total_loss": 57.133392, "clip_frac": 0.124146, "sps": 438.7, "elapsed_s": 3284.5} |
| {"update": 702, "steps_done": 1437696, "mean_step_reward": 0.4521, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 78.587, "mean_ep_length": 200.0, "mean_collisions": 23.4, "mean_intent_rate": 0.702, "policy_loss": 0.006828, "value_loss": 174.711211, "entropy": 0.30605, "total_loss": 87.359373, "clip_frac": 0.140869, "sps": 427.8, "elapsed_s": 3289.3} |
| {"update": 703, "steps_done": 1439744, "mean_step_reward": 0.7176, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 150.303, "mean_ep_length": 200.0, "mean_collisions": 15.64, "mean_intent_rate": 0.69, "policy_loss": 0.001715, "value_loss": 147.399428, "entropy": 0.304511, "total_loss": 73.698383, "clip_frac": 0.130371, "sps": 434.9, "elapsed_s": 3294.0} |
| {"update": 704, "steps_done": 1441792, "mean_step_reward": 0.6861, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 146.091, "mean_ep_length": 200.0, "mean_collisions": 17.6, "mean_intent_rate": 0.716, "policy_loss": 0.000496, "value_loss": 150.679773, "entropy": 0.296563, "total_loss": 75.337417, "clip_frac": 0.103149, "sps": 435.5, "elapsed_s": 3298.7} |
| {"update": 705, "steps_done": 1443840, "mean_step_reward": 0.8388, "min_step_reward": -10.5329, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 157.475, "mean_ep_length": 200.0, "mean_collisions": 14.8, "mean_intent_rate": 0.689, "policy_loss": 0.001403, "value_loss": 93.86186, "entropy": 0.29733, "total_loss": 46.929359, "clip_frac": 0.074463, "sps": 434.3, "elapsed_s": 3303.4} |
| {"update": 706, "steps_done": 1445888, "mean_step_reward": 0.2499, "min_step_reward": -10.5576, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 64.883, "mean_ep_length": 200.0, "mean_collisions": 23.4, "mean_intent_rate": 0.677, "policy_loss": 0.012348, "value_loss": 185.790915, "entropy": 0.323643, "total_loss": 92.904569, "clip_frac": 0.159302, "sps": 440.5, "elapsed_s": 3308.1} |
| {"update": 707, "steps_done": 1447936, "mean_step_reward": 0.8049, "min_step_reward": -10.5801, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 146.643, "mean_ep_length": 199.7, "mean_collisions": 16.55, "mean_intent_rate": 0.7, "policy_loss": 0.006805, "value_loss": 110.310914, "entropy": 0.288902, "total_loss": 55.159373, "clip_frac": 0.09021, "sps": 448.4, "elapsed_s": 3312.6} |
| {"update": 708, "steps_done": 1449984, "mean_step_reward": 0.9006, "min_step_reward": -10.5834, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 185.312, "mean_ep_length": 200.0, "mean_collisions": 12.1, "mean_intent_rate": 0.69, "policy_loss": 0.022775, "value_loss": 104.693319, "entropy": 0.291362, "total_loss": 52.36652, "clip_frac": 0.211548, "sps": 438.7, "elapsed_s": 3317.3} |
| {"update": 709, "steps_done": 1452032, "mean_step_reward": 0.4919, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 84.711, "mean_ep_length": 200.0, "mean_collisions": 22.8, "mean_intent_rate": 0.702, "policy_loss": 0.006145, "value_loss": 154.275874, "entropy": 0.281071, "total_loss": 77.141272, "clip_frac": 0.119873, "sps": 444.0, "elapsed_s": 3321.9} |
| {"update": 710, "steps_done": 1454080, "mean_step_reward": 0.9726, "min_step_reward": -10.5619, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 192.816, "mean_ep_length": 200.0, "mean_collisions": 11.1, "mean_intent_rate": 0.684, "policy_loss": -0.001997, "value_loss": 122.992067, "entropy": 0.299187, "total_loss": 61.491045, "clip_frac": 0.088989, "sps": 438.6, "elapsed_s": 3326.6} |
| {"update": 711, "steps_done": 1456128, "mean_step_reward": 0.8199, "min_step_reward": -10.5656, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 175.927, "mean_ep_length": 200.0, "mean_collisions": 10.91, "mean_intent_rate": 0.655, "policy_loss": 0.012381, "value_loss": 125.123002, "entropy": 0.293593, "total_loss": 62.570945, "clip_frac": 0.14624, "sps": 430.4, "elapsed_s": 3331.3} |
| {"update": 712, "steps_done": 1458176, "mean_step_reward": 1.1051, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 220.651, "mean_ep_length": 200.0, "mean_collisions": 7.0, "mean_intent_rate": 0.662, "policy_loss": 0.003665, "value_loss": 105.900173, "entropy": 0.290306, "total_loss": 52.950848, "clip_frac": 0.10791, "sps": 446.8, "elapsed_s": 3335.9} |
| {"update": 713, "steps_done": 1460224, "mean_step_reward": 1.2788, "min_step_reward": -10.5083, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 252.283, "mean_ep_length": 200.0, "mean_collisions": 3.9, "mean_intent_rate": 0.663, "policy_loss": 0.029477, "value_loss": 96.768053, "entropy": 0.284631, "total_loss": 48.410656, "clip_frac": 0.207275, "sps": 435.1, "elapsed_s": 3340.6} |
| {"update": 714, "steps_done": 1462272, "mean_step_reward": 1.4042, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 290.468, "mean_ep_length": 200.0, "mean_collisions": 1.6, "mean_intent_rate": 0.688, "policy_loss": 0.007539, "value_loss": 113.318899, "entropy": 0.297343, "total_loss": 56.664015, "clip_frac": 0.108887, "sps": 439.0, "elapsed_s": 3345.3} |
| {"update": 715, "steps_done": 1464320, "mean_step_reward": 1.3369, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 258.989, "mean_ep_length": 200.0, "mean_collisions": 4.09, "mean_intent_rate": 0.678, "policy_loss": 0.006188, "value_loss": 139.451281, "entropy": 0.292509, "total_loss": 69.728903, "clip_frac": 0.118286, "sps": 439.7, "elapsed_s": 3350.0} |
| {"update": 716, "steps_done": 1466368, "mean_step_reward": 1.2916, "min_step_reward": -10.5374, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 256.898, "mean_ep_length": 200.0, "mean_collisions": 3.2, "mean_intent_rate": 0.657, "policy_loss": 0.010781, "value_loss": 138.09547, "entropy": 0.282155, "total_loss": 69.055695, "clip_frac": 0.155151, "sps": 440.8, "elapsed_s": 3354.6} |
| {"update": 717, "steps_done": 1468416, "mean_step_reward": 1.2227, "min_step_reward": -10.5352, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 247.03, "mean_ep_length": 200.0, "mean_collisions": 6.0, "mean_intent_rate": 0.69, "policy_loss": 0.021736, "value_loss": 142.35006, "entropy": 0.263967, "total_loss": 71.194127, "clip_frac": 0.17395, "sps": 437.7, "elapsed_s": 3359.3} |
| {"update": 718, "steps_done": 1470464, "mean_step_reward": 1.195, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 238.972, "mean_ep_length": 200.0, "mean_collisions": 5.4, "mean_intent_rate": 0.666, "policy_loss": 0.003209, "value_loss": 164.524177, "entropy": 0.254841, "total_loss": 82.262749, "clip_frac": 0.082642, "sps": 452.6, "elapsed_s": 3363.8} |
| {"update": 719, "steps_done": 1472512, "mean_step_reward": 0.9736, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 198.051, "mean_ep_length": 200.0, "mean_collisions": 9.18, "mean_intent_rate": 0.664, "policy_loss": 0.017689, "value_loss": 168.448284, "entropy": 0.283673, "total_loss": 84.238995, "clip_frac": 0.150513, "sps": 435.1, "elapsed_s": 3368.5} |
| {"update": 720, "steps_done": 1474560, "mean_step_reward": 1.1162, "min_step_reward": -10.5564, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 231.94, "mean_ep_length": 200.0, "mean_collisions": 6.7, "mean_intent_rate": 0.677, "policy_loss": 0.002266, "value_loss": 145.340711, "entropy": 0.256861, "total_loss": 72.670052, "clip_frac": 0.102295, "sps": 446.2, "elapsed_s": 3373.1} |
| {"update": 721, "steps_done": 1476608, "mean_step_reward": 0.8429, "min_step_reward": -10.5609, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 156.145, "mean_ep_length": 200.0, "mean_collisions": 15.7, "mean_intent_rate": 0.701, "policy_loss": 0.010792, "value_loss": 196.358591, "entropy": 0.257684, "total_loss": 98.187512, "clip_frac": 0.12146, "sps": 442.9, "elapsed_s": 3377.7} |
| {"update": 722, "steps_done": 1478656, "mean_step_reward": 0.8938, "min_step_reward": -10.5315, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 179.871, "mean_ep_length": 200.0, "mean_collisions": 10.1, "mean_intent_rate": 0.647, "policy_loss": 0.027031, "value_loss": 133.723079, "entropy": 0.279449, "total_loss": 66.885776, "clip_frac": 0.207153, "sps": 438.5, "elapsed_s": 3382.4} |
| {"update": 723, "steps_done": 1480704, "mean_step_reward": 1.028, "min_step_reward": -10.5452, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 215.567, "mean_ep_length": 200.0, "mean_collisions": 8.6, "mean_intent_rate": 0.679, "policy_loss": 0.016502, "value_loss": 130.621253, "entropy": 0.281419, "total_loss": 65.324315, "clip_frac": 0.169678, "sps": 435.1, "elapsed_s": 3387.1} |
| {"update": 724, "steps_done": 1482752, "mean_step_reward": 1.0895, "min_step_reward": -10.5596, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 207.739, "mean_ep_length": 200.0, "mean_collisions": 11.18, "mean_intent_rate": 0.712, "policy_loss": 0.009017, "value_loss": 119.68441, "entropy": 0.255268, "total_loss": 59.848669, "clip_frac": 0.097168, "sps": 439.5, "elapsed_s": 3391.8} |
| {"update": 725, "steps_done": 1484800, "mean_step_reward": 0.8015, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 163.565, "mean_ep_length": 200.0, "mean_collisions": 15.1, "mean_intent_rate": 0.704, "policy_loss": 0.010475, "value_loss": 162.610069, "entropy": 0.258282, "total_loss": 81.312926, "clip_frac": 0.135132, "sps": 444.7, "elapsed_s": 3396.4} |
| {"update": 726, "steps_done": 1486848, "mean_step_reward": 0.9642, "min_step_reward": -10.5895, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 189.203, "mean_ep_length": 200.0, "mean_collisions": 11.6, "mean_intent_rate": 0.688, "policy_loss": 0.009801, "value_loss": 126.771298, "entropy": 0.268478, "total_loss": 63.392765, "clip_frac": 0.122314, "sps": 441.2, "elapsed_s": 3401.0} |
| {"update": 727, "steps_done": 1488896, "mean_step_reward": 0.9596, "min_step_reward": -10.5404, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 194.3, "mean_ep_length": 200.0, "mean_collisions": 11.8, "mean_intent_rate": 0.698, "policy_loss": -0.001008, "value_loss": 151.178656, "entropy": 0.2631, "total_loss": 75.585689, "clip_frac": 0.04895, "sps": 439.8, "elapsed_s": 3405.7} |
| {"update": 728, "steps_done": 1490944, "mean_step_reward": 0.9159, "min_step_reward": -10.5342, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 181.519, "mean_ep_length": 200.0, "mean_collisions": 13.45, "mean_intent_rate": 0.706, "policy_loss": 0.015664, "value_loss": 161.788761, "entropy": 0.279815, "total_loss": 80.907245, "clip_frac": 0.240112, "sps": 426.7, "elapsed_s": 3410.5} |
| {"update": 729, "steps_done": 1492992, "mean_step_reward": 0.9239, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 185.644, "mean_ep_length": 200.0, "mean_collisions": 12.7, "mean_intent_rate": 0.701, "policy_loss": 0.020501, "value_loss": 130.294638, "entropy": 0.293004, "total_loss": 65.164891, "clip_frac": 0.176025, "sps": 435.0, "elapsed_s": 3415.2} |
| {"update": 730, "steps_done": 1495040, "mean_step_reward": 1.1772, "min_step_reward": -10.5497, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 235.868, "mean_ep_length": 200.0, "mean_collisions": 8.3, "mean_intent_rate": 0.71, "policy_loss": 0.026135, "value_loss": 137.50963, "entropy": 0.265717, "total_loss": 68.778293, "clip_frac": 0.157837, "sps": 438.1, "elapsed_s": 3419.9} |
| {"update": 731, "steps_done": 1497088, "mean_step_reward": 1.1008, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 222.814, "mean_ep_length": 200.0, "mean_collisions": 6.8, "mean_intent_rate": 0.665, "policy_loss": 0.004428, "value_loss": 126.80487, "entropy": 0.272533, "total_loss": 63.404137, "clip_frac": 0.07251, "sps": 441.4, "elapsed_s": 3424.5} |
| {"update": 732, "steps_done": 1499136, "mean_step_reward": 1.1069, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 225.178, "mean_ep_length": 200.0, "mean_collisions": 7.73, "mean_intent_rate": 0.682, "policy_loss": 0.004925, "value_loss": 172.438368, "entropy": 0.272105, "total_loss": 86.221388, "clip_frac": 0.105103, "sps": 446.0, "elapsed_s": 3429.1} |
| {"update": 733, "steps_done": 1501184, "mean_step_reward": 0.9828, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 185.438, "mean_ep_length": 200.0, "mean_collisions": 9.3, "mean_intent_rate": 0.644, "policy_loss": 0.010154, "value_loss": 122.119742, "entropy": 0.269304, "total_loss": 61.067332, "clip_frac": 0.098511, "sps": 438.0, "elapsed_s": 3433.8} |
| {"update": 734, "steps_done": 1503232, "mean_step_reward": 1.3508, "min_step_reward": -10.5539, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 272.935, "mean_ep_length": 200.0, "mean_collisions": 4.4, "mean_intent_rate": 0.706, "policy_loss": 0.016916, "value_loss": 148.092186, "entropy": 0.245809, "total_loss": 74.060549, "clip_frac": 0.133789, "sps": 444.3, "elapsed_s": 3438.4} |
| {"update": 735, "steps_done": 1505280, "mean_step_reward": 0.7034, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 149.111, "mean_ep_length": 200.0, "mean_collisions": 16.7, "mean_intent_rate": 0.705, "policy_loss": 0.014164, "value_loss": 205.431793, "entropy": 0.258715, "total_loss": 102.727474, "clip_frac": 0.141602, "sps": 442.9, "elapsed_s": 3443.0} |
| {"update": 736, "steps_done": 1507328, "mean_step_reward": 1.155, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 223.412, "mean_ep_length": 200.0, "mean_collisions": 10.27, "mean_intent_rate": 0.721, "policy_loss": 0.006198, "value_loss": 165.652072, "entropy": 0.266704, "total_loss": 82.829566, "clip_frac": 0.096558, "sps": 443.5, "elapsed_s": 3447.6} |
| {"update": 737, "steps_done": 1509376, "mean_step_reward": 1.2549, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 249.301, "mean_ep_length": 200.0, "mean_collisions": 4.7, "mean_intent_rate": 0.672, "policy_loss": 0.02377, "value_loss": 125.875981, "entropy": 0.273296, "total_loss": 62.959027, "clip_frac": 0.188965, "sps": 433.7, "elapsed_s": 3452.4} |
| {"update": 738, "steps_done": 1511424, "mean_step_reward": 1.3083, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 260.564, "mean_ep_length": 200.0, "mean_collisions": 5.9, "mean_intent_rate": 0.711, "policy_loss": 0.004397, "value_loss": 140.517732, "entropy": 0.267391, "total_loss": 70.260589, "clip_frac": 0.09729, "sps": 444.6, "elapsed_s": 3457.0} |
| {"update": 739, "steps_done": 1513472, "mean_step_reward": 1.1534, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 231.598, "mean_ep_length": 200.0, "mean_collisions": 4.9, "mean_intent_rate": 0.647, "policy_loss": 0.006059, "value_loss": 153.131746, "entropy": 0.268572, "total_loss": 76.569247, "clip_frac": 0.087158, "sps": 444.2, "elapsed_s": 3461.6} |
| {"update": 740, "steps_done": 1515520, "mean_step_reward": 1.0744, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 216.931, "mean_ep_length": 200.0, "mean_collisions": 9.27, "mean_intent_rate": 0.695, "policy_loss": 0.01628, "value_loss": 164.108322, "entropy": 0.277441, "total_loss": 82.067667, "clip_frac": 0.166504, "sps": 436.3, "elapsed_s": 3466.3} |
| {"update": 741, "steps_done": 1517568, "mean_step_reward": 1.3354, "min_step_reward": -10.5756, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 278.778, "mean_ep_length": 200.0, "mean_collisions": 3.6, "mean_intent_rate": 0.702, "policy_loss": 0.010039, "value_loss": 175.401466, "entropy": 0.257051, "total_loss": 87.708201, "clip_frac": 0.126953, "sps": 441.5, "elapsed_s": 3470.9} |
| {"update": 742, "steps_done": 1519616, "mean_step_reward": 1.2658, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 239.739, "mean_ep_length": 200.0, "mean_collisions": 5.7, "mean_intent_rate": 0.672, "policy_loss": 0.017942, "value_loss": 155.01021, "entropy": 0.268094, "total_loss": 77.520365, "clip_frac": 0.150391, "sps": 436.4, "elapsed_s": 3475.6} |
| {"update": 743, "steps_done": 1521664, "mean_step_reward": 0.9148, "min_step_reward": -10.5193, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 190.159, "mean_ep_length": 200.0, "mean_collisions": 8.8, "mean_intent_rate": 0.643, "policy_loss": 0.015143, "value_loss": 148.685959, "entropy": 0.275842, "total_loss": 74.355364, "clip_frac": 0.162964, "sps": 441.7, "elapsed_s": 3480.2} |
| {"update": 744, "steps_done": 1523712, "mean_step_reward": 1.1494, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 225.905, "mean_ep_length": 200.0, "mean_collisions": 4.36, "mean_intent_rate": 0.627, "policy_loss": 0.015333, "value_loss": 176.625616, "entropy": 0.28161, "total_loss": 88.325325, "clip_frac": 0.144043, "sps": 446.8, "elapsed_s": 3484.8} |
| {"update": 745, "steps_done": 1525760, "mean_step_reward": 1.3223, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 262.895, "mean_ep_length": 200.0, "mean_collisions": 2.9, "mean_intent_rate": 0.665, "policy_loss": 0.015905, "value_loss": 150.320101, "entropy": 0.270127, "total_loss": 75.173254, "clip_frac": 0.158203, "sps": 440.0, "elapsed_s": 3489.5} |
| {"update": 746, "steps_done": 1527808, "mean_step_reward": 1.2164, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 243.871, "mean_ep_length": 200.0, "mean_collisions": 4.0, "mean_intent_rate": 0.651, "policy_loss": 0.011955, "value_loss": 147.652092, "entropy": 0.276806, "total_loss": 73.835233, "clip_frac": 0.148193, "sps": 427.1, "elapsed_s": 3494.3} |
| {"update": 747, "steps_done": 1529856, "mean_step_reward": 1.3518, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 271.349, "mean_ep_length": 200.0, "mean_collisions": 2.8, "mean_intent_rate": 0.675, "policy_loss": 0.015059, "value_loss": 160.931993, "entropy": 0.255817, "total_loss": 80.478497, "clip_frac": 0.092529, "sps": 438.8, "elapsed_s": 3498.9} |
| {"update": 748, "steps_done": 1531904, "mean_step_reward": 1.2039, "min_step_reward": -10.5621, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 251.655, "mean_ep_length": 200.0, "mean_collisions": 5.6, "mean_intent_rate": 0.692, "policy_loss": 0.020887, "value_loss": 154.269815, "entropy": 0.272063, "total_loss": 77.153075, "clip_frac": 0.184448, "sps": 438.8, "elapsed_s": 3503.6} |
| {"update": 749, "steps_done": 1533952, "mean_step_reward": 1.3012, "min_step_reward": -10.5515, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 250.753, "mean_ep_length": 200.0, "mean_collisions": 4.45, "mean_intent_rate": 0.67, "policy_loss": 0.029111, "value_loss": 164.329777, "entropy": 0.282521, "total_loss": 82.191174, "clip_frac": 0.220093, "sps": 436.2, "elapsed_s": 3508.3} |
| {"update": 750, "steps_done": 1536000, "mean_step_reward": 1.3593, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 267.221, "mean_ep_length": 200.0, "mean_collisions": 1.9, "mean_intent_rate": 0.653, "policy_loss": 0.011124, "value_loss": 162.735732, "entropy": 0.291872, "total_loss": 81.376072, "clip_frac": 0.123169, "sps": 438.0, "elapsed_s": 3513.0} |
| {"update": 751, "steps_done": 1538048, "mean_step_reward": 1.2643, "min_step_reward": -7.5835, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 256.574, "mean_ep_length": 200.0, "mean_collisions": 0.4, "mean_intent_rate": 0.612, "policy_loss": 0.018881, "value_loss": 163.684619, "entropy": 0.299803, "total_loss": 81.858192, "clip_frac": 0.168335, "sps": 438.3, "elapsed_s": 3517.7} |
| {"update": 752, "steps_done": 1540096, "mean_step_reward": 1.1359, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 220.668, "mean_ep_length": 200.0, "mean_collisions": 1.0, "mean_intent_rate": 0.564, "policy_loss": 0.013845, "value_loss": 157.57659, "entropy": 0.306233, "total_loss": 78.799078, "clip_frac": 0.15686, "sps": 441.3, "elapsed_s": 3522.3} |
| {"update": 753, "steps_done": 1542144, "mean_step_reward": 1.2193, "min_step_reward": -7.5863, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 252.948, "mean_ep_length": 200.0, "mean_collisions": 0.73, "mean_intent_rate": 0.611, "policy_loss": 0.011496, "value_loss": 159.24068, "entropy": 0.314799, "total_loss": 79.628688, "clip_frac": 0.122681, "sps": 435.0, "elapsed_s": 3527.0} |
| {"update": 754, "steps_done": 1544192, "mean_step_reward": 1.1187, "min_step_reward": -7.5882, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 223.708, "mean_ep_length": 200.0, "mean_collisions": 0.8, "mean_intent_rate": 0.564, "policy_loss": 0.002706, "value_loss": 139.842965, "entropy": 0.302612, "total_loss": 69.921164, "clip_frac": 0.065186, "sps": 438.4, "elapsed_s": 3531.7} |
| {"update": 755, "steps_done": 1546240, "mean_step_reward": 0.9686, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 191.8, "mean_ep_length": 200.0, "mean_collisions": 1.2, "mean_intent_rate": 0.516, "policy_loss": 0.011409, "value_loss": 127.01156, "entropy": 0.304885, "total_loss": 63.51414, "clip_frac": 0.149048, "sps": 433.0, "elapsed_s": 3536.4} |
| {"update": 756, "steps_done": 1548288, "mean_step_reward": 1.1031, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 225.872, "mean_ep_length": 200.0, "mean_collisions": 1.9, "mean_intent_rate": 0.586, "policy_loss": 0.000182, "value_loss": 142.666378, "entropy": 0.306604, "total_loss": 71.330305, "clip_frac": 0.0625, "sps": 445.2, "elapsed_s": 3541.0} |
| {"update": 757, "steps_done": 1550336, "mean_step_reward": 1.1786, "min_step_reward": -7.5869, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 229.011, "mean_ep_length": 200.0, "mean_collisions": 0.64, "mean_intent_rate": 0.569, "policy_loss": -0.00099, "value_loss": 137.19048, "entropy": 0.311853, "total_loss": 68.59113, "clip_frac": 0.074341, "sps": 436.5, "elapsed_s": 3545.7} |
| {"update": 758, "steps_done": 1552384, "mean_step_reward": 1.1198, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 230.293, "mean_ep_length": 200.0, "mean_collisions": 1.4, "mean_intent_rate": 0.584, "policy_loss": 0.009119, "value_loss": 127.604027, "entropy": 0.342573, "total_loss": 63.807707, "clip_frac": 0.146729, "sps": 436.8, "elapsed_s": 3550.4} |
| {"update": 759, "steps_done": 1554432, "mean_step_reward": 0.9358, "min_step_reward": -7.5863, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 182.14, "mean_ep_length": 200.0, "mean_collisions": 1.4, "mean_intent_rate": 0.502, "policy_loss": 0.010935, "value_loss": 108.578869, "entropy": 0.340345, "total_loss": 54.296966, "clip_frac": 0.150024, "sps": 440.1, "elapsed_s": 3555.1} |
| {"update": 760, "steps_done": 1556480, "mean_step_reward": 1.0013, "min_step_reward": -7.5845, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 202.233, "mean_ep_length": 200.0, "mean_collisions": 1.4, "mean_intent_rate": 0.538, "policy_loss": 0.013277, "value_loss": 127.102171, "entropy": 0.290515, "total_loss": 63.561457, "clip_frac": 0.188721, "sps": 436.9, "elapsed_s": 3559.7} |
| {"update": 761, "steps_done": 1558528, "mean_step_reward": 1.0516, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 206.874, "mean_ep_length": 200.0, "mean_collisions": 1.09, "mean_intent_rate": 0.54, "policy_loss": 0.008031, "value_loss": 132.617789, "entropy": 0.297744, "total_loss": 66.313949, "clip_frac": 0.127441, "sps": 438.8, "elapsed_s": 3564.4} |
| {"update": 762, "steps_done": 1560576, "mean_step_reward": 1.1535, "min_step_reward": -7.583, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 226.569, "mean_ep_length": 200.0, "mean_collisions": 1.0, "mean_intent_rate": 0.57, "policy_loss": 0.009498, "value_loss": 126.7246, "entropy": 0.308142, "total_loss": 63.368716, "clip_frac": 0.130737, "sps": 439.2, "elapsed_s": 3569.1} |
| {"update": 763, "steps_done": 1562624, "mean_step_reward": 0.9725, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 214.329, "mean_ep_length": 200.0, "mean_collisions": 1.2, "mean_intent_rate": 0.554, "policy_loss": 0.018266, "value_loss": 135.796138, "entropy": 0.299244, "total_loss": 67.913342, "clip_frac": 0.1604, "sps": 438.0, "elapsed_s": 3573.8} |
| {"update": 764, "steps_done": 1564672, "mean_step_reward": 1.0526, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 191.391, "mean_ep_length": 200.0, "mean_collisions": 1.2, "mean_intent_rate": 0.516, "policy_loss": 0.006316, "value_loss": 131.679412, "entropy": 0.303565, "total_loss": 65.842986, "clip_frac": 0.107178, "sps": 432.9, "elapsed_s": 3578.5} |
| {"update": 765, "steps_done": 1566720, "mean_step_reward": 0.871, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 182.052, "mean_ep_length": 200.0, "mean_collisions": 1.09, "mean_intent_rate": 0.497, "policy_loss": 0.008675, "value_loss": 115.306808, "entropy": 0.307119, "total_loss": 57.659008, "clip_frac": 0.099121, "sps": 442.9, "elapsed_s": 3583.1} |
| {"update": 766, "steps_done": 1568768, "mean_step_reward": 1.0049, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 199.512, "mean_ep_length": 200.0, "mean_collisions": 0.8, "mean_intent_rate": 0.522, "policy_loss": 0.002239, "value_loss": 105.558007, "entropy": 0.318969, "total_loss": 52.778053, "clip_frac": 0.118286, "sps": 441.4, "elapsed_s": 3587.8} |
| {"update": 767, "steps_done": 1570816, "mean_step_reward": 1.2298, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 248.943, "mean_ep_length": 200.0, "mean_collisions": 1.4, "mean_intent_rate": 0.614, "policy_loss": 0.000743, "value_loss": 108.042031, "entropy": 0.303777, "total_loss": 54.018721, "clip_frac": 0.119263, "sps": 441.4, "elapsed_s": 3592.4} |
| {"update": 768, "steps_done": 1572864, "mean_step_reward": 1.1462, "min_step_reward": -7.5847, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 225.396, "mean_ep_length": 200.0, "mean_collisions": 1.1, "mean_intent_rate": 0.569, "policy_loss": 0.006235, "value_loss": 121.188712, "entropy": 0.278666, "total_loss": 60.597805, "clip_frac": 0.091675, "sps": 439.8, "elapsed_s": 3597.1} |
| {"update": 769, "steps_done": 1574912, "mean_step_reward": 1.2391, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 246.644, "mean_ep_length": 200.0, "mean_collisions": 0.91, "mean_intent_rate": 0.603, "policy_loss": 0.008268, "value_loss": 150.193151, "entropy": 0.279065, "total_loss": 75.102054, "clip_frac": 0.133789, "sps": 436.7, "elapsed_s": 3601.7} |
| {"update": 770, "steps_done": 1576960, "mean_step_reward": 1.0449, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 210.237, "mean_ep_length": 200.0, "mean_collisions": 1.5, "mean_intent_rate": 0.551, "policy_loss": 0.008349, "value_loss": 113.037415, "entropy": 0.297023, "total_loss": 56.524086, "clip_frac": 0.151001, "sps": 444.7, "elapsed_s": 3606.3} |
| {"update": 771, "steps_done": 1579008, "mean_step_reward": 0.9795, "min_step_reward": -7.5843, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 196.247, "mean_ep_length": 200.0, "mean_collisions": 1.0, "mean_intent_rate": 0.52, "policy_loss": 0.0118, "value_loss": 107.833584, "entropy": 0.307441, "total_loss": 53.925517, "clip_frac": 0.162109, "sps": 439.8, "elapsed_s": 3611.0} |
| {"update": 772, "steps_done": 1581056, "mean_step_reward": 0.9885, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 199.247, "mean_ep_length": 200.0, "mean_collisions": 1.9, "mean_intent_rate": 0.541, "policy_loss": 0.018943, "value_loss": 117.890755, "entropy": 0.297541, "total_loss": 58.961344, "clip_frac": 0.18457, "sps": 443.6, "elapsed_s": 3615.6} |
| {"update": 773, "steps_done": 1583104, "mean_step_reward": 1.2845, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 252.0, "mean_ep_length": 200.0, "mean_collisions": 1.8, "mean_intent_rate": 0.628, "policy_loss": 0.007598, "value_loss": 134.03285, "entropy": 0.28521, "total_loss": 67.02117, "clip_frac": 0.099731, "sps": 424.5, "elapsed_s": 3620.4} |
| {"update": 774, "steps_done": 1585152, "mean_step_reward": 1.0458, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 208.027, "mean_ep_length": 200.0, "mean_collisions": 1.0, "mean_intent_rate": 0.541, "policy_loss": 0.012347, "value_loss": 135.946479, "entropy": 0.279857, "total_loss": 67.982788, "clip_frac": 0.13855, "sps": 444.7, "elapsed_s": 3625.1} |
| {"update": 775, "steps_done": 1587200, "mean_step_reward": 0.9796, "min_step_reward": -7.5851, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 198.557, "mean_ep_length": 200.0, "mean_collisions": 0.8, "mean_intent_rate": 0.523, "policy_loss": 0.008534, "value_loss": 125.524199, "entropy": 0.288563, "total_loss": 62.767747, "clip_frac": 0.146973, "sps": 441.3, "elapsed_s": 3629.7} |
| {"update": 776, "steps_done": 1589248, "mean_step_reward": 0.9816, "min_step_reward": -7.5871, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 198.783, "mean_ep_length": 200.0, "mean_collisions": 2.0, "mean_intent_rate": 0.54, "policy_loss": 0.040318, "value_loss": 112.596242, "entropy": 0.296399, "total_loss": 56.335475, "clip_frac": 0.248901, "sps": 442.3, "elapsed_s": 3634.3} |
| {"update": 777, "steps_done": 1591296, "mean_step_reward": 1.0854, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 208.44, "mean_ep_length": 200.0, "mean_collisions": 4.5, "mean_intent_rate": 0.599, "policy_loss": 0.010801, "value_loss": 134.286384, "entropy": 0.314269, "total_loss": 67.150851, "clip_frac": 0.119507, "sps": 441.9, "elapsed_s": 3639.0} |
| {"update": 778, "steps_done": 1593344, "mean_step_reward": 1.1609, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 234.891, "mean_ep_length": 200.0, "mean_collisions": 3.64, "mean_intent_rate": 0.628, "policy_loss": 0.002572, "value_loss": 138.832477, "entropy": 0.300173, "total_loss": 69.415808, "clip_frac": 0.053467, "sps": 437.1, "elapsed_s": 3643.7} |
| {"update": 779, "steps_done": 1595392, "mean_step_reward": 1.2495, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 259.248, "mean_ep_length": 200.0, "mean_collisions": 1.9, "mean_intent_rate": 0.64, "policy_loss": 0.002879, "value_loss": 141.664195, "entropy": 0.307063, "total_loss": 70.831906, "clip_frac": 0.064453, "sps": 435.5, "elapsed_s": 3648.4} |
| {"update": 780, "steps_done": 1597440, "mean_step_reward": 1.1207, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 216.935, "mean_ep_length": 200.0, "mean_collisions": 3.2, "mean_intent_rate": 0.591, "policy_loss": 0.005143, "value_loss": 133.723357, "entropy": 0.308085, "total_loss": 66.86374, "clip_frac": 0.127441, "sps": 437.0, "elapsed_s": 3653.0} |
| {"update": 781, "steps_done": 1599488, "mean_step_reward": 1.2296, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 241.186, "mean_ep_length": 200.0, "mean_collisions": 2.3, "mean_intent_rate": 0.616, "policy_loss": 0.008006, "value_loss": 128.580625, "entropy": 0.300225, "total_loss": 64.295317, "clip_frac": 0.137207, "sps": 432.2, "elapsed_s": 3657.8} |
| {"update": 782, "steps_done": 1601536, "mean_step_reward": 1.0065, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 208.191, "mean_ep_length": 200.0, "mean_collisions": 4.64, "mean_intent_rate": 0.602, "policy_loss": 0.021359, "value_loss": 150.643947, "entropy": 0.314101, "total_loss": 75.340192, "clip_frac": 0.214966, "sps": 438.2, "elapsed_s": 3662.5} |
| {"update": 783, "steps_done": 1603584, "mean_step_reward": 1.2532, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 248.07, "mean_ep_length": 200.0, "mean_collisions": 2.7, "mean_intent_rate": 0.636, "policy_loss": 0.012499, "value_loss": 147.9583, "entropy": 0.314868, "total_loss": 73.9885, "clip_frac": 0.130859, "sps": 434.7, "elapsed_s": 3667.2} |
| {"update": 784, "steps_done": 1605632, "mean_step_reward": 1.1937, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 239.566, "mean_ep_length": 200.0, "mean_collisions": 1.9, "mean_intent_rate": 0.608, "policy_loss": 0.006505, "value_loss": 141.05657, "entropy": 0.312002, "total_loss": 70.531669, "clip_frac": 0.109619, "sps": 442.1, "elapsed_s": 3671.8} |
| {"update": 785, "steps_done": 1607680, "mean_step_reward": 1.3378, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 262.991, "mean_ep_length": 200.0, "mean_collisions": 1.7, "mean_intent_rate": 0.643, "policy_loss": 0.012442, "value_loss": 134.006338, "entropy": 0.297144, "total_loss": 67.012639, "clip_frac": 0.1427, "sps": 439.3, "elapsed_s": 3676.5} |
| {"update": 786, "steps_done": 1609728, "mean_step_reward": 1.2416, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 253.442, "mean_ep_length": 200.0, "mean_collisions": 2.91, "mean_intent_rate": 0.648, "policy_loss": 0.004311, "value_loss": 162.675385, "entropy": 0.306554, "total_loss": 81.338937, "clip_frac": 0.071655, "sps": 448.9, "elapsed_s": 3681.0} |
| {"update": 787, "steps_done": 1611776, "mean_step_reward": 1.1233, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 222.728, "mean_ep_length": 200.0, "mean_collisions": 3.3, "mean_intent_rate": 0.603, "policy_loss": 0.004096, "value_loss": 137.871126, "entropy": 0.296357, "total_loss": 68.936695, "clip_frac": 0.099609, "sps": 439.5, "elapsed_s": 3685.7} |
| {"update": 788, "steps_done": 1613824, "mean_step_reward": 1.325, "min_step_reward": -7.5894, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 259.087, "mean_ep_length": 200.0, "mean_collisions": 1.8, "mean_intent_rate": 0.638, "policy_loss": 0.003773, "value_loss": 145.834615, "entropy": 0.275287, "total_loss": 72.918328, "clip_frac": 0.073364, "sps": 436.8, "elapsed_s": 3690.4} |
| {"update": 789, "steps_done": 1615872, "mean_step_reward": 1.2882, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 268.26, "mean_ep_length": 200.0, "mean_collisions": 1.5, "mean_intent_rate": 0.648, "policy_loss": 0.005559, "value_loss": 149.008223, "entropy": 0.274563, "total_loss": 74.506924, "clip_frac": 0.07605, "sps": 439.9, "elapsed_s": 3695.0} |
| {"update": 790, "steps_done": 1617920, "mean_step_reward": 1.2491, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 248.486, "mean_ep_length": 200.0, "mean_collisions": 3.27, "mean_intent_rate": 0.645, "policy_loss": 0.011054, "value_loss": 166.808564, "entropy": 0.288219, "total_loss": 83.412455, "clip_frac": 0.123413, "sps": 429.6, "elapsed_s": 3699.8} |
| {"update": 791, "steps_done": 1619968, "mean_step_reward": 1.2723, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 251.937, "mean_ep_length": 200.0, "mean_collisions": 1.6, "mean_intent_rate": 0.623, "policy_loss": 0.022323, "value_loss": 150.138112, "entropy": 0.29957, "total_loss": 75.088383, "clip_frac": 0.219727, "sps": 443.3, "elapsed_s": 3704.4} |
| {"update": 792, "steps_done": 1622016, "mean_step_reward": 1.1353, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 232.493, "mean_ep_length": 200.0, "mean_collisions": 3.6, "mean_intent_rate": 0.624, "policy_loss": 0.012402, "value_loss": 143.362959, "entropy": 0.341966, "total_loss": 71.690463, "clip_frac": 0.11377, "sps": 443.5, "elapsed_s": 3709.0} |
| {"update": 793, "steps_done": 1624064, "mean_step_reward": 1.133, "min_step_reward": -10.535, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 223.389, "mean_ep_length": 200.0, "mean_collisions": 1.8, "mean_intent_rate": 0.578, "policy_loss": 0.012174, "value_loss": 125.77587, "entropy": 0.323429, "total_loss": 62.896874, "clip_frac": 0.174194, "sps": 440.6, "elapsed_s": 3713.7} |
| {"update": 794, "steps_done": 1626112, "mean_step_reward": 1.216, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 242.53, "mean_ep_length": 200.0, "mean_collisions": 2.09, "mean_intent_rate": 0.615, "policy_loss": 0.001674, "value_loss": 129.296974, "entropy": 0.31063, "total_loss": 64.647055, "clip_frac": 0.086426, "sps": 437.1, "elapsed_s": 3718.4} |
| {"update": 795, "steps_done": 1628160, "mean_step_reward": 1.2116, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 243.618, "mean_ep_length": 200.0, "mean_collisions": 1.9, "mean_intent_rate": 0.613, "policy_loss": 0.007481, "value_loss": 129.842082, "entropy": 0.295473, "total_loss": 64.925567, "clip_frac": 0.115723, "sps": 436.8, "elapsed_s": 3723.1} |
| {"update": 796, "steps_done": 1630208, "mean_step_reward": 1.2934, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 257.687, "mean_ep_length": 200.0, "mean_collisions": 3.9, "mean_intent_rate": 0.671, "policy_loss": 0.009469, "value_loss": 162.134979, "entropy": 0.29794, "total_loss": 81.073979, "clip_frac": 0.13916, "sps": 439.8, "elapsed_s": 3727.7} |
| {"update": 797, "steps_done": 1632256, "mean_step_reward": 1.1106, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 224.398, "mean_ep_length": 200.0, "mean_collisions": 4.6, "mean_intent_rate": 0.627, "policy_loss": 0.004316, "value_loss": 151.786522, "entropy": 0.30416, "total_loss": 75.894536, "clip_frac": 0.089844, "sps": 440.4, "elapsed_s": 3732.4} |
| {"update": 798, "steps_done": 1634304, "mean_step_reward": 1.19, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 240.585, "mean_ep_length": 200.0, "mean_collisions": 3.9, "mean_intent_rate": 0.642, "policy_loss": 0.004844, "value_loss": 154.828037, "entropy": 0.282694, "total_loss": 77.416034, "clip_frac": 0.064453, "sps": 438.0, "elapsed_s": 3737.0} |
| {"update": 799, "steps_done": 1636352, "mean_step_reward": 1.2187, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 239.498, "mean_ep_length": 200.0, "mean_collisions": 5.18, "mean_intent_rate": 0.662, "policy_loss": 0.004362, "value_loss": 156.573832, "entropy": 0.27231, "total_loss": 78.288555, "clip_frac": 0.104614, "sps": 430.1, "elapsed_s": 3741.8} |
| {"update": 800, "steps_done": 1638400, "mean_step_reward": 1.2288, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 242.28, "mean_ep_length": 200.0, "mean_collisions": 7.2, "mean_intent_rate": 0.702, "policy_loss": 0.005287, "value_loss": 168.382023, "entropy": 0.2668, "total_loss": 84.193631, "clip_frac": 0.084839, "sps": 434.8, "elapsed_s": 3746.5} |
| {"update": 801, "steps_done": 1640448, "mean_step_reward": 1.347, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 265.113, "mean_ep_length": 200.0, "mean_collisions": 4.5, "mean_intent_rate": 0.693, "policy_loss": 0.013901, "value_loss": 171.526567, "entropy": 0.258007, "total_loss": 85.774605, "clip_frac": 0.143433, "sps": 440.4, "elapsed_s": 3751.2} |
| {"update": 802, "steps_done": 1642496, "mean_step_reward": 1.1449, "min_step_reward": -10.5315, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 237.696, "mean_ep_length": 200.0, "mean_collisions": 6.8, "mean_intent_rate": 0.686, "policy_loss": 0.011779, "value_loss": 151.058583, "entropy": 0.272084, "total_loss": 75.53835, "clip_frac": 0.128418, "sps": 437.7, "elapsed_s": 3755.9} |
| {"update": 803, "steps_done": 1644544, "mean_step_reward": 1.0968, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 217.931, "mean_ep_length": 200.0, "mean_collisions": 8.0, "mean_intent_rate": 0.673, "policy_loss": 0.020797, "value_loss": 168.294006, "entropy": 0.276252, "total_loss": 84.165038, "clip_frac": 0.17395, "sps": 441.8, "elapsed_s": 3760.5} |
| {"update": 804, "steps_done": 1646592, "mean_step_reward": 1.1758, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 241.302, "mean_ep_length": 200.0, "mean_collisions": 7.4, "mean_intent_rate": 0.704, "policy_loss": 0.003882, "value_loss": 151.297918, "entropy": 0.251721, "total_loss": 75.650324, "clip_frac": 0.095825, "sps": 439.3, "elapsed_s": 3765.2} |
| {"update": 805, "steps_done": 1648640, "mean_step_reward": 1.1235, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 215.949, "mean_ep_length": 200.0, "mean_collisions": 8.3, "mean_intent_rate": 0.676, "policy_loss": 0.005083, "value_loss": 152.623948, "entropy": 0.241305, "total_loss": 76.314645, "clip_frac": 0.080933, "sps": 441.8, "elapsed_s": 3769.8} |
| {"update": 806, "steps_done": 1650688, "mean_step_reward": 1.4472, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 292.052, "mean_ep_length": 200.0, "mean_collisions": 2.4, "mean_intent_rate": 0.704, "policy_loss": 0.01491, "value_loss": 150.128535, "entropy": 0.241685, "total_loss": 75.076761, "clip_frac": 0.160034, "sps": 437.4, "elapsed_s": 3774.5} |
| {"update": 807, "steps_done": 1652736, "mean_step_reward": 1.1964, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 241.881, "mean_ep_length": 200.0, "mean_collisions": 6.09, "mean_intent_rate": 0.68, "policy_loss": 0.004689, "value_loss": 166.827447, "entropy": 0.252616, "total_loss": 83.415888, "clip_frac": 0.090698, "sps": 440.1, "elapsed_s": 3779.1} |
| {"update": 808, "steps_done": 1654784, "mean_step_reward": 0.917, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 175.735, "mean_ep_length": 200.0, "mean_collisions": 12.6, "mean_intent_rate": 0.68, "policy_loss": 0.010057, "value_loss": 219.784877, "entropy": 0.265109, "total_loss": 109.899843, "clip_frac": 0.123657, "sps": 433.9, "elapsed_s": 3783.9} |
| {"update": 809, "steps_done": 1656832, "mean_step_reward": 1.2518, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 255.164, "mean_ep_length": 200.0, "mean_collisions": 5.2, "mean_intent_rate": 0.69, "policy_loss": 0.007124, "value_loss": 166.167884, "entropy": 0.277199, "total_loss": 83.088295, "clip_frac": 0.12561, "sps": 437.4, "elapsed_s": 3788.5} |
| {"update": 810, "steps_done": 1658880, "mean_step_reward": 1.3617, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 277.154, "mean_ep_length": 200.0, "mean_collisions": 2.8, "mean_intent_rate": 0.684, "policy_loss": 0.008087, "value_loss": 150.696516, "entropy": 0.29201, "total_loss": 75.353425, "clip_frac": 0.09668, "sps": 438.5, "elapsed_s": 3793.2} |
| {"update": 811, "steps_done": 1660928, "mean_step_reward": 1.1962, "min_step_reward": -10.561, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 235.224, "mean_ep_length": 200.0, "mean_collisions": 7.09, "mean_intent_rate": 0.686, "policy_loss": 0.005997, "value_loss": 187.590889, "entropy": 0.255395, "total_loss": 93.798887, "clip_frac": 0.120361, "sps": 441.5, "elapsed_s": 3797.9} |
| {"update": 812, "steps_done": 1662976, "mean_step_reward": 1.1569, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 228.49, "mean_ep_length": 200.0, "mean_collisions": 4.3, "mean_intent_rate": 0.628, "policy_loss": 0.002709, "value_loss": 145.270674, "entropy": 0.289258, "total_loss": 72.635154, "clip_frac": 0.069092, "sps": 437.5, "elapsed_s": 3802.5} |
| {"update": 813, "steps_done": 1665024, "mean_step_reward": 1.1365, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 233.584, "mean_ep_length": 200.0, "mean_collisions": 6.4, "mean_intent_rate": 0.671, "policy_loss": 0.023754, "value_loss": 168.520527, "entropy": 0.260291, "total_loss": 84.281413, "clip_frac": 0.187988, "sps": 437.7, "elapsed_s": 3807.2} |
| {"update": 814, "steps_done": 1667072, "mean_step_reward": 1.0015, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 188.812, "mean_ep_length": 200.0, "mean_collisions": 11.7, "mean_intent_rate": 0.689, "policy_loss": 0.016143, "value_loss": 184.477405, "entropy": 0.267248, "total_loss": 92.252172, "clip_frac": 0.16333, "sps": 439.8, "elapsed_s": 3811.9} |
| {"update": 815, "steps_done": 1669120, "mean_step_reward": 1.069, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 221.762, "mean_ep_length": 200.0, "mean_collisions": 9.55, "mean_intent_rate": 0.708, "policy_loss": -0.000215, "value_loss": 189.723761, "entropy": 0.252022, "total_loss": 94.859146, "clip_frac": 0.063843, "sps": 437.7, "elapsed_s": 3816.6} |
| {"update": 816, "steps_done": 1671168, "mean_step_reward": 0.8831, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 177.02, "mean_ep_length": 200.0, "mean_collisions": 12.1, "mean_intent_rate": 0.676, "policy_loss": 0.015433, "value_loss": 178.177227, "entropy": 0.286767, "total_loss": 89.101179, "clip_frac": 0.158081, "sps": 441.4, "elapsed_s": 3821.2} |
| {"update": 817, "steps_done": 1673216, "mean_step_reward": 1.0329, "min_step_reward": -10.5874, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 205.083, "mean_ep_length": 200.0, "mean_collisions": 10.0, "mean_intent_rate": 0.686, "policy_loss": 0.006932, "value_loss": 122.096274, "entropy": 0.264141, "total_loss": 61.052428, "clip_frac": 0.075928, "sps": 430.2, "elapsed_s": 3826.0} |
| {"update": 818, "steps_done": 1675264, "mean_step_reward": 0.8731, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 166.466, "mean_ep_length": 200.0, "mean_collisions": 14.8, "mean_intent_rate": 0.704, "policy_loss": 0.011885, "value_loss": 136.423549, "entropy": 0.273704, "total_loss": 68.220923, "clip_frac": 0.12085, "sps": 438.8, "elapsed_s": 3830.6} |
| {"update": 819, "steps_done": 1677312, "mean_step_reward": 0.6201, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 137.818, "mean_ep_length": 200.0, "mean_collisions": 18.36, "mean_intent_rate": 0.714, "policy_loss": 0.008703, "value_loss": 195.414093, "entropy": 0.283176, "total_loss": 97.712918, "clip_frac": 0.147217, "sps": 437.0, "elapsed_s": 3835.3} |
| {"update": 820, "steps_done": 1679360, "mean_step_reward": 0.922, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 181.736, "mean_ep_length": 200.0, "mean_collisions": 14.4, "mean_intent_rate": 0.72, "policy_loss": 0.010525, "value_loss": 141.484098, "entropy": 0.275377, "total_loss": 70.74982, "clip_frac": 0.108765, "sps": 439.0, "elapsed_s": 3840.0} |
| {"update": 821, "steps_done": 1681408, "mean_step_reward": 0.3894, "min_step_reward": -10.5853, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 80.061, "mean_ep_length": 200.0, "mean_collisions": 24.1, "mean_intent_rate": 0.714, "policy_loss": 0.024042, "value_loss": 264.227562, "entropy": 0.287622, "total_loss": 132.134947, "clip_frac": 0.201782, "sps": 438.9, "elapsed_s": 3844.6} |
| {"update": 822, "steps_done": 1683456, "mean_step_reward": 0.6658, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 130.88, "mean_ep_length": 200.0, "mean_collisions": 18.6, "mean_intent_rate": 0.707, "policy_loss": 0.030672, "value_loss": 149.816082, "entropy": 0.287489, "total_loss": 74.935837, "clip_frac": 0.193726, "sps": 440.3, "elapsed_s": 3849.3} |
| {"update": 823, "steps_done": 1685504, "mean_step_reward": 0.675, "min_step_reward": -10.5332, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 139.288, "mean_ep_length": 200.0, "mean_collisions": 17.1, "mean_intent_rate": 0.697, "policy_loss": 0.015173, "value_loss": 122.812923, "entropy": 0.299282, "total_loss": 61.418641, "clip_frac": 0.1604, "sps": 435.1, "elapsed_s": 3854.0} |
| {"update": 824, "steps_done": 1687552, "mean_step_reward": 0.8687, "min_step_reward": -10.5874, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 167.392, "mean_ep_length": 200.0, "mean_collisions": 13.91, "mean_intent_rate": 0.69, "policy_loss": 0.009544, "value_loss": 119.439506, "entropy": 0.274991, "total_loss": 59.726548, "clip_frac": 0.122314, "sps": 438.0, "elapsed_s": 3858.7} |
| {"update": 825, "steps_done": 1689600, "mean_step_reward": 0.7138, "min_step_reward": -10.5565, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 152.673, "mean_ep_length": 200.0, "mean_collisions": 15.1, "mean_intent_rate": 0.686, "policy_loss": 0.008183, "value_loss": 154.447459, "entropy": 0.274892, "total_loss": 77.229164, "clip_frac": 0.13208, "sps": 441.8, "elapsed_s": 3863.3} |
| {"update": 826, "steps_done": 1691648, "mean_step_reward": 0.5059, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 93.074, "mean_ep_length": 200.0, "mean_collisions": 21.6, "mean_intent_rate": 0.697, "policy_loss": 0.007055, "value_loss": 180.278434, "entropy": 0.293898, "total_loss": 90.143332, "clip_frac": 0.116943, "sps": 427.7, "elapsed_s": 3868.1} |
| {"update": 827, "steps_done": 1693696, "mean_step_reward": 0.5162, "min_step_reward": -10.5633, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 128.661, "mean_ep_length": 200.0, "mean_collisions": 19.2, "mean_intent_rate": 0.714, "policy_loss": 0.001258, "value_loss": 194.960438, "entropy": 0.312137, "total_loss": 97.478355, "clip_frac": 0.120239, "sps": 439.0, "elapsed_s": 3872.8} |
| {"update": 828, "steps_done": 1695744, "mean_step_reward": 0.6964, "min_step_reward": -10.5609, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 107.12, "mean_ep_length": 190.6, "mean_collisions": 19.73, "mean_intent_rate": 0.712, "policy_loss": 0.009199, "value_loss": 137.764968, "entropy": 0.308027, "total_loss": 68.888602, "clip_frac": 0.151978, "sps": 435.1, "elapsed_s": 3877.5} |
| {"update": 829, "steps_done": 1697792, "mean_step_reward": 0.4461, "min_step_reward": -10.5845, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 92.622, "mean_ep_length": 200.0, "mean_collisions": 22.8, "mean_intent_rate": 0.715, "policy_loss": 0.004228, "value_loss": 183.037186, "entropy": 0.302896, "total_loss": 91.519792, "clip_frac": 0.084595, "sps": 440.9, "elapsed_s": 3882.1} |
| {"update": 830, "steps_done": 1699840, "mean_step_reward": 0.4152, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 82.216, "mean_ep_length": 200.0, "mean_collisions": 22.91, "mean_intent_rate": 0.698, "policy_loss": 0.004339, "value_loss": 190.008046, "entropy": 0.311497, "total_loss": 95.005247, "clip_frac": 0.097534, "sps": 437.2, "elapsed_s": 3886.8} |
| {"update": 831, "steps_done": 1701888, "mean_step_reward": 1.0425, "min_step_reward": -10.5547, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 205.984, "mean_ep_length": 200.0, "mean_collisions": 11.9, "mean_intent_rate": 0.722, "policy_loss": 0.019312, "value_loss": 155.285276, "entropy": 0.321695, "total_loss": 77.658732, "clip_frac": 0.176147, "sps": 436.3, "elapsed_s": 3891.5} |
| {"update": 832, "steps_done": 1703936, "mean_step_reward": 0.8622, "min_step_reward": -10.537, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 184.338, "mean_ep_length": 200.0, "mean_collisions": 13.4, "mean_intent_rate": 0.71, "policy_loss": 0.008021, "value_loss": 164.486232, "entropy": 0.283675, "total_loss": 82.248299, "clip_frac": 0.112549, "sps": 434.4, "elapsed_s": 3896.2} |
| {"update": 833, "steps_done": 1705984, "mean_step_reward": 0.9031, "min_step_reward": -10.5824, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 169.482, "mean_ep_length": 200.0, "mean_collisions": 15.7, "mean_intent_rate": 0.723, "policy_loss": 0.006251, "value_loss": 155.525928, "entropy": 0.264528, "total_loss": 77.766569, "clip_frac": 0.080933, "sps": 439.2, "elapsed_s": 3900.9} |
| {"update": 834, "steps_done": 1708032, "mean_step_reward": 0.9108, "min_step_reward": -10.5084, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 185.754, "mean_ep_length": 200.0, "mean_collisions": 13.91, "mean_intent_rate": 0.72, "policy_loss": 0.005856, "value_loss": 123.731808, "entropy": 0.276609, "total_loss": 61.868994, "clip_frac": 0.114014, "sps": 437.1, "elapsed_s": 3905.6} |
| {"update": 835, "steps_done": 1710080, "mean_step_reward": 0.6958, "min_step_reward": -10.5608, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 136.048, "mean_ep_length": 200.0, "mean_collisions": 16.0, "mean_intent_rate": 0.673, "policy_loss": 0.006735, "value_loss": 163.349275, "entropy": 0.279414, "total_loss": 81.678578, "clip_frac": 0.102905, "sps": 433.7, "elapsed_s": 3910.3} |
| {"update": 836, "steps_done": 1712128, "mean_step_reward": 0.9467, "min_step_reward": -10.5454, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 190.861, "mean_ep_length": 200.0, "mean_collisions": 13.0, "mean_intent_rate": 0.713, "policy_loss": 0.014293, "value_loss": 183.742551, "entropy": 0.251819, "total_loss": 91.883049, "clip_frac": 0.126099, "sps": 442.6, "elapsed_s": 3914.9} |
| {"update": 837, "steps_done": 1714176, "mean_step_reward": 1.1202, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 219.063, "mean_ep_length": 200.0, "mean_collisions": 10.0, "mean_intent_rate": 0.71, "policy_loss": 0.004763, "value_loss": 165.562056, "entropy": 0.258051, "total_loss": 82.783211, "clip_frac": 0.099609, "sps": 438.7, "elapsed_s": 3919.6} |
| {"update": 838, "steps_done": 1716224, "mean_step_reward": 1.3135, "min_step_reward": -10.5816, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 263.119, "mean_ep_length": 200.0, "mean_collisions": 6.82, "mean_intent_rate": 0.729, "policy_loss": 0.006261, "value_loss": 156.464366, "entropy": 0.236231, "total_loss": 78.236082, "clip_frac": 0.139282, "sps": 437.3, "elapsed_s": 3924.3} |
| {"update": 839, "steps_done": 1718272, "mean_step_reward": 1.2002, "min_step_reward": -10.5272, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 238.471, "mean_ep_length": 200.0, "mean_collisions": 7.0, "mean_intent_rate": 0.693, "policy_loss": 0.005032, "value_loss": 118.230398, "entropy": 0.270969, "total_loss": 59.117521, "clip_frac": 0.085938, "sps": 438.1, "elapsed_s": 3929.0} |
| {"update": 840, "steps_done": 1720320, "mean_step_reward": 1.0599, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 212.006, "mean_ep_length": 200.0, "mean_collisions": 8.2, "mean_intent_rate": 0.668, "policy_loss": 0.007964, "value_loss": 129.214338, "entropy": 0.255754, "total_loss": 64.612575, "clip_frac": 0.099731, "sps": 440.2, "elapsed_s": 3933.6} |
| {"update": 841, "steps_done": 1722368, "mean_step_reward": 0.9692, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 198.622, "mean_ep_length": 200.0, "mean_collisions": 10.0, "mean_intent_rate": 0.675, "policy_loss": 0.019922, "value_loss": 151.999921, "entropy": 0.273936, "total_loss": 76.017143, "clip_frac": 0.161621, "sps": 440.7, "elapsed_s": 3938.3} |
| {"update": 842, "steps_done": 1724416, "mean_step_reward": 1.0527, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 206.845, "mean_ep_length": 200.0, "mean_collisions": 6.73, "mean_intent_rate": 0.635, "policy_loss": -0.000288, "value_loss": 160.680374, "entropy": 0.256391, "total_loss": 80.337334, "clip_frac": 0.066772, "sps": 439.6, "elapsed_s": 3942.9} |
| {"update": 843, "steps_done": 1726464, "mean_step_reward": 1.1997, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 238.622, "mean_ep_length": 200.0, "mean_collisions": 5.6, "mean_intent_rate": 0.67, "policy_loss": 0.004946, "value_loss": 140.615659, "entropy": 0.247444, "total_loss": 70.310302, "clip_frac": 0.080811, "sps": 430.3, "elapsed_s": 3947.7} |
| {"update": 844, "steps_done": 1728512, "mean_step_reward": 1.11, "min_step_reward": -10.5813, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 219.171, "mean_ep_length": 200.0, "mean_collisions": 8.5, "mean_intent_rate": 0.685, "policy_loss": 0.009444, "value_loss": 136.145817, "entropy": 0.252512, "total_loss": 68.079828, "clip_frac": 0.12561, "sps": 436.2, "elapsed_s": 3952.4} |
| {"update": 845, "steps_done": 1730560, "mean_step_reward": 1.0396, "min_step_reward": -10.5381, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 210.076, "mean_ep_length": 200.0, "mean_collisions": 9.1, "mean_intent_rate": 0.68, "policy_loss": 0.005583, "value_loss": 129.222446, "entropy": 0.267722, "total_loss": 64.614129, "clip_frac": 0.133423, "sps": 439.3, "elapsed_s": 3957.0} |
| {"update": 846, "steps_done": 1732608, "mean_step_reward": 0.8374, "min_step_reward": -10.5768, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 173.856, "mean_ep_length": 200.0, "mean_collisions": 13.0, "mean_intent_rate": 0.685, "policy_loss": 0.001679, "value_loss": 175.070136, "entropy": 0.289425, "total_loss": 87.533851, "clip_frac": 0.099121, "sps": 435.7, "elapsed_s": 3961.7} |
| {"update": 847, "steps_done": 1734656, "mean_step_reward": 1.1572, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 230.365, "mean_ep_length": 200.0, "mean_collisions": 5.9, "mean_intent_rate": 0.662, "policy_loss": 0.009929, "value_loss": 125.446037, "entropy": 0.256828, "total_loss": 62.730379, "clip_frac": 0.1073, "sps": 442.5, "elapsed_s": 3966.4} |
| {"update": 848, "steps_done": 1736704, "mean_step_reward": 0.7013, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 149.081, "mean_ep_length": 200.0, "mean_collisions": 14.9, "mean_intent_rate": 0.678, "policy_loss": 0.009949, "value_loss": 174.91684, "entropy": 0.274817, "total_loss": 87.46562, "clip_frac": 0.157104, "sps": 436.2, "elapsed_s": 3971.1} |
| {"update": 849, "steps_done": 1738752, "mean_step_reward": 0.9259, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 200.544, "mean_ep_length": 200.0, "mean_collisions": 10.8, "mean_intent_rate": 0.692, "policy_loss": 0.017606, "value_loss": 167.224349, "entropy": 0.277731, "total_loss": 83.627004, "clip_frac": 0.190674, "sps": 440.7, "elapsed_s": 3975.7} |
| {"update": 850, "steps_done": 1740800, "mean_step_reward": 0.8163, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 126.954, "mean_ep_length": 200.0, "mean_collisions": 18.7, "mean_intent_rate": 0.704, "policy_loss": 0.000137, "value_loss": 165.163591, "entropy": 0.274272, "total_loss": 82.57919, "clip_frac": 0.074951, "sps": 445.2, "elapsed_s": 3980.3} |
| {"update": 851, "steps_done": 1742848, "mean_step_reward": 0.9517, "min_step_reward": -10.5489, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 201.355, "mean_ep_length": 200.0, "mean_collisions": 10.18, "mean_intent_rate": 0.685, "policy_loss": 0.007785, "value_loss": 146.933699, "entropy": 0.268123, "total_loss": 73.471953, "clip_frac": 0.112671, "sps": 444.8, "elapsed_s": 3984.9} |
| {"update": 852, "steps_done": 1744896, "mean_step_reward": 0.9769, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 191.703, "mean_ep_length": 200.0, "mean_collisions": 11.4, "mean_intent_rate": 0.689, "policy_loss": 0.007839, "value_loss": 145.012471, "entropy": 0.28822, "total_loss": 72.511192, "clip_frac": 0.119019, "sps": 426.3, "elapsed_s": 3989.7} |
| {"update": 853, "steps_done": 1746944, "mean_step_reward": 1.1156, "min_step_reward": -10.5635, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 243.473, "mean_ep_length": 200.0, "mean_collisions": 7.9, "mean_intent_rate": 0.714, "policy_loss": 0.003515, "value_loss": 144.278975, "entropy": 0.256399, "total_loss": 72.140439, "clip_frac": 0.09082, "sps": 435.4, "elapsed_s": 3994.4} |
| {"update": 854, "steps_done": 1748992, "mean_step_reward": 0.5823, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 115.464, "mean_ep_length": 200.0, "mean_collisions": 14.6, "mean_intent_rate": 0.616, "policy_loss": 0.002324, "value_loss": 182.643318, "entropy": 0.301368, "total_loss": 91.320969, "clip_frac": 0.078613, "sps": 448.4, "elapsed_s": 3999.0} |
| {"update": 855, "steps_done": 1751040, "mean_step_reward": 1.0472, "min_step_reward": -10.5569, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 196.106, "mean_ep_length": 200.0, "mean_collisions": 7.82, "mean_intent_rate": 0.637, "policy_loss": 0.009303, "value_loss": 118.491447, "entropy": 0.272204, "total_loss": 59.252305, "clip_frac": 0.111816, "sps": 441.2, "elapsed_s": 4003.6} |
| {"update": 856, "steps_done": 1753088, "mean_step_reward": 1.0306, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 201.037, "mean_ep_length": 200.0, "mean_collisions": 9.4, "mean_intent_rate": 0.67, "policy_loss": 0.01678, "value_loss": 127.770846, "entropy": 0.253916, "total_loss": 63.899664, "clip_frac": 0.144775, "sps": 435.5, "elapsed_s": 4008.3} |
| {"update": 857, "steps_done": 1755136, "mean_step_reward": 0.9543, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 190.009, "mean_ep_length": 200.0, "mean_collisions": 11.8, "mean_intent_rate": 0.691, "policy_loss": 0.004792, "value_loss": 148.942921, "entropy": 0.282998, "total_loss": 74.473422, "clip_frac": 0.103027, "sps": 433.9, "elapsed_s": 4013.1} |
| {"update": 858, "steps_done": 1757184, "mean_step_reward": 0.9661, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 193.974, "mean_ep_length": 200.0, "mean_collisions": 11.8, "mean_intent_rate": 0.7, "policy_loss": 0.005424, "value_loss": 154.519562, "entropy": 0.286108, "total_loss": 77.262344, "clip_frac": 0.067505, "sps": 444.8, "elapsed_s": 4017.7} |
| {"update": 859, "steps_done": 1759232, "mean_step_reward": 0.8636, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 176.543, "mean_ep_length": 200.0, "mean_collisions": 8.91, "mean_intent_rate": 0.621, "policy_loss": 0.004231, "value_loss": 131.328619, "entropy": 0.272632, "total_loss": 65.665815, "clip_frac": 0.065552, "sps": 451.7, "elapsed_s": 4022.2} |
| {"update": 860, "steps_done": 1761280, "mean_step_reward": 1.0256, "min_step_reward": -10.5509, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 199.373, "mean_ep_length": 200.0, "mean_collisions": 9.7, "mean_intent_rate": 0.673, "policy_loss": 0.002353, "value_loss": 143.965462, "entropy": 0.297279, "total_loss": 71.982111, "clip_frac": 0.038574, "sps": 449.8, "elapsed_s": 4026.8} |
| {"update": 861, "steps_done": 1763328, "mean_step_reward": 0.9195, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 166.667, "mean_ep_length": 182.6, "mean_collisions": 9.64, "mean_intent_rate": 0.622, "policy_loss": 0.005611, "value_loss": 129.324338, "entropy": 0.295494, "total_loss": 64.664825, "clip_frac": 0.102051, "sps": 430.0, "elapsed_s": 4031.5} |
| {"update": 862, "steps_done": 1765376, "mean_step_reward": 1.0471, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 214.822, "mean_ep_length": 200.0, "mean_collisions": 8.2, "mean_intent_rate": 0.673, "policy_loss": 0.008044, "value_loss": 121.877325, "entropy": 0.290864, "total_loss": 60.943797, "clip_frac": 0.1073, "sps": 438.6, "elapsed_s": 4036.2} |
| {"update": 863, "steps_done": 1767424, "mean_step_reward": 0.8605, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 176.243, "mean_ep_length": 200.0, "mean_collisions": 11.73, "mean_intent_rate": 0.667, "policy_loss": 0.003447, "value_loss": 143.25009, "entropy": 0.312111, "total_loss": 71.625371, "clip_frac": 0.093994, "sps": 443.8, "elapsed_s": 4040.8} |
| {"update": 864, "steps_done": 1769472, "mean_step_reward": 0.7711, "min_step_reward": -10.5602, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 149.636, "mean_ep_length": 200.0, "mean_collisions": 16.2, "mean_intent_rate": 0.697, "policy_loss": 0.015151, "value_loss": 160.4616, "entropy": 0.29272, "total_loss": 80.243024, "clip_frac": 0.142334, "sps": 437.6, "elapsed_s": 4045.5} |
| {"update": 865, "steps_done": 1771520, "mean_step_reward": 1.0042, "min_step_reward": -10.5353, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 199.537, "mean_ep_length": 200.0, "mean_collisions": 12.5, "mean_intent_rate": 0.719, "policy_loss": 0.005785, "value_loss": 154.622637, "entropy": 0.282743, "total_loss": 77.314276, "clip_frac": 0.092163, "sps": 441.7, "elapsed_s": 4050.1} |
| {"update": 866, "steps_done": 1773568, "mean_step_reward": 0.9145, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 203.061, "mean_ep_length": 200.0, "mean_collisions": 10.0, "mean_intent_rate": 0.683, "policy_loss": 0.005615, "value_loss": 155.220822, "entropy": 0.32175, "total_loss": 77.612809, "clip_frac": 0.123413, "sps": 443.1, "elapsed_s": 4054.7} |
| {"update": 867, "steps_done": 1775616, "mean_step_reward": 1.1023, "min_step_reward": -10.5356, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 203.132, "mean_ep_length": 200.0, "mean_collisions": 11.45, "mean_intent_rate": 0.709, "policy_loss": 0.006814, "value_loss": 153.585198, "entropy": 0.290557, "total_loss": 76.796507, "clip_frac": 0.099365, "sps": 449.6, "elapsed_s": 4059.3} |
| {"update": 868, "steps_done": 1777664, "mean_step_reward": 1.1862, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 236.404, "mean_ep_length": 200.0, "mean_collisions": 7.2, "mean_intent_rate": 0.693, "policy_loss": 0.006139, "value_loss": 147.756708, "entropy": 0.28538, "total_loss": 73.88164, "clip_frac": 0.091675, "sps": 449.3, "elapsed_s": 4063.9} |
| {"update": 869, "steps_done": 1779712, "mean_step_reward": 1.096, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 221.533, "mean_ep_length": 200.0, "mean_collisions": 9.5, "mean_intent_rate": 0.707, "policy_loss": 0.006439, "value_loss": 164.784392, "entropy": 0.299373, "total_loss": 82.395642, "clip_frac": 0.108398, "sps": 448.9, "elapsed_s": 4068.4} |
| {"update": 870, "steps_done": 1781760, "mean_step_reward": 0.8109, "min_step_reward": -10.5377, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 184.265, "mean_ep_length": 200.0, "mean_collisions": 13.6, "mean_intent_rate": 0.712, "policy_loss": 0.009816, "value_loss": 205.474826, "entropy": 0.306017, "total_loss": 102.744169, "clip_frac": 0.123657, "sps": 438.4, "elapsed_s": 4073.1} |
| {"update": 871, "steps_done": 1783808, "mean_step_reward": 1.1477, "min_step_reward": -10.5195, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 209.08, "mean_ep_length": 200.0, "mean_collisions": 12.7, "mean_intent_rate": 0.74, "policy_loss": 0.005162, "value_loss": 150.598764, "entropy": 0.283862, "total_loss": 75.301706, "clip_frac": 0.093994, "sps": 446.0, "elapsed_s": 4077.7} |
| {"update": 872, "steps_done": 1785856, "mean_step_reward": 1.0721, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 212.376, "mean_ep_length": 200.0, "mean_collisions": 10.45, "mean_intent_rate": 0.708, "policy_loss": 0.003844, "value_loss": 154.647389, "entropy": 0.273547, "total_loss": 77.324803, "clip_frac": 0.083374, "sps": 439.0, "elapsed_s": 4082.4} |
| {"update": 873, "steps_done": 1787904, "mean_step_reward": 1.047, "min_step_reward": -10.5453, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 212.558, "mean_ep_length": 200.0, "mean_collisions": 12.0, "mean_intent_rate": 0.733, "policy_loss": 0.004908, "value_loss": 174.769787, "entropy": 0.275759, "total_loss": 87.387043, "clip_frac": 0.099121, "sps": 441.6, "elapsed_s": 4087.0} |
| {"update": 874, "steps_done": 1789952, "mean_step_reward": 0.9482, "min_step_reward": -10.5309, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 186.666, "mean_ep_length": 200.0, "mean_collisions": 11.6, "mean_intent_rate": 0.682, "policy_loss": 0.004948, "value_loss": 138.139792, "entropy": 0.303748, "total_loss": 69.071806, "clip_frac": 0.111572, "sps": 441.1, "elapsed_s": 4091.6} |
| {"update": 875, "steps_done": 1792000, "mean_step_reward": 0.8354, "min_step_reward": -10.5376, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 167.8, "mean_ep_length": 200.0, "mean_collisions": 15.7, "mean_intent_rate": 0.72, "policy_loss": -0.002179, "value_loss": 175.640071, "entropy": 0.312846, "total_loss": 87.814728, "clip_frac": 0.077271, "sps": 439.6, "elapsed_s": 4096.3} |
| {"update": 876, "steps_done": 1794048, "mean_step_reward": 0.8445, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 168.135, "mean_ep_length": 200.0, "mean_collisions": 13.36, "mean_intent_rate": 0.683, "policy_loss": 0.002472, "value_loss": 156.902164, "entropy": 0.342223, "total_loss": 78.450131, "clip_frac": 0.079468, "sps": 437.2, "elapsed_s": 4101.0} |
| {"update": 877, "steps_done": 1796096, "mean_step_reward": 0.9483, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 182.788, "mean_ep_length": 200.0, "mean_collisions": 9.9, "mean_intent_rate": 0.649, "policy_loss": 0.008217, "value_loss": 144.329841, "entropy": 0.320728, "total_loss": 72.16993, "clip_frac": 0.104248, "sps": 439.5, "elapsed_s": 4105.6} |
| {"update": 878, "steps_done": 1798144, "mean_step_reward": 0.8875, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 198.205, "mean_ep_length": 200.0, "mean_collisions": 11.1, "mean_intent_rate": 0.695, "policy_loss": 0.018921, "value_loss": 157.713179, "entropy": 0.333057, "total_loss": 78.87218, "clip_frac": 0.144165, "sps": 436.6, "elapsed_s": 4110.3} |
| {"update": 879, "steps_done": 1800192, "mean_step_reward": 0.8043, "min_step_reward": -10.5349, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 142.6, "mean_ep_length": 200.0, "mean_collisions": 15.9, "mean_intent_rate": 0.682, "policy_loss": 0.01296, "value_loss": 167.49472, "entropy": 0.347922, "total_loss": 83.756841, "clip_frac": 0.197021, "sps": 429.5, "elapsed_s": 4115.1} |
| {"update": 880, "steps_done": 1802240, "mean_step_reward": 1.2572, "min_step_reward": -10.5488, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 249.201, "mean_ep_length": 200.0, "mean_collisions": 4.27, "mean_intent_rate": 0.665, "policy_loss": 0.004361, "value_loss": 129.305147, "entropy": 0.320537, "total_loss": 64.653729, "clip_frac": 0.054077, "sps": 439.5, "elapsed_s": 4119.8} |
| {"update": 881, "steps_done": 1804288, "mean_step_reward": 1.241, "min_step_reward": -10.5668, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 249.56, "mean_ep_length": 200.0, "mean_collisions": 7.2, "mean_intent_rate": 0.714, "policy_loss": 0.007828, "value_loss": 126.573702, "entropy": 0.3124, "total_loss": 63.291555, "clip_frac": 0.124512, "sps": 442.2, "elapsed_s": 4124.4} |
| {"update": 882, "steps_done": 1806336, "mean_step_reward": 1.0727, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 212.104, "mean_ep_length": 195.8, "mean_collisions": 6.1, "mean_intent_rate": 0.645, "policy_loss": 0.016374, "value_loss": 130.861778, "entropy": 0.311779, "total_loss": 65.444145, "clip_frac": 0.158447, "sps": 441.0, "elapsed_s": 4129.0} |
| {"update": 883, "steps_done": 1808384, "mean_step_reward": 1.2562, "min_step_reward": -10.5501, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 247.246, "mean_ep_length": 200.0, "mean_collisions": 3.55, "mean_intent_rate": 0.65, "policy_loss": 0.00861, "value_loss": 151.080009, "entropy": 0.30356, "total_loss": 75.545579, "clip_frac": 0.074341, "sps": 436.5, "elapsed_s": 4133.7} |
| {"update": 884, "steps_done": 1810432, "mean_step_reward": 1.3469, "min_step_reward": -7.5855, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 271.917, "mean_ep_length": 200.0, "mean_collisions": 1.7, "mean_intent_rate": 0.657, "policy_loss": 0.007656, "value_loss": 137.167691, "entropy": 0.300055, "total_loss": 68.588501, "clip_frac": 0.091675, "sps": 440.0, "elapsed_s": 4138.4} |
| {"update": 885, "steps_done": 1812480, "mean_step_reward": 1.2253, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 240.813, "mean_ep_length": 200.0, "mean_collisions": 3.4, "mean_intent_rate": 0.636, "policy_loss": 0.006445, "value_loss": 155.229738, "entropy": 0.320875, "total_loss": 77.618105, "clip_frac": 0.066284, "sps": 436.8, "elapsed_s": 4143.1} |
| {"update": 886, "steps_done": 1814528, "mean_step_reward": 1.1333, "min_step_reward": -7.5901, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 228.262, "mean_ep_length": 200.0, "mean_collisions": 3.8, "mean_intent_rate": 0.621, "policy_loss": 0.012629, "value_loss": 157.692164, "entropy": 0.315218, "total_loss": 78.855559, "clip_frac": 0.152466, "sps": 437.2, "elapsed_s": 4147.8} |
| {"update": 887, "steps_done": 1816576, "mean_step_reward": 1.3057, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 261.13, "mean_ep_length": 200.0, "mean_collisions": 3.36, "mean_intent_rate": 0.669, "policy_loss": 0.005925, "value_loss": 167.775878, "entropy": 0.315163, "total_loss": 83.890713, "clip_frac": 0.08606, "sps": 438.6, "elapsed_s": 4152.4} |
| {"update": 888, "steps_done": 1818624, "mean_step_reward": 1.2203, "min_step_reward": -7.5899, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 247.341, "mean_ep_length": 200.0, "mean_collisions": 4.4, "mean_intent_rate": 0.663, "policy_loss": 0.008074, "value_loss": 161.255771, "entropy": 0.329477, "total_loss": 80.632664, "clip_frac": 0.122925, "sps": 433.2, "elapsed_s": 4157.2} |
| {"update": 889, "steps_done": 1820672, "mean_step_reward": 1.2556, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 247.123, "mean_ep_length": 200.0, "mean_collisions": 2.5, "mean_intent_rate": 0.63, "policy_loss": 0.00377, "value_loss": 159.115352, "entropy": 0.254872, "total_loss": 79.558898, "clip_frac": 0.110352, "sps": 442.9, "elapsed_s": 4161.8} |
| {"update": 890, "steps_done": 1822720, "mean_step_reward": 1.2795, "min_step_reward": -7.5877, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 255.838, "mean_ep_length": 200.0, "mean_collisions": 0.6, "mean_intent_rate": 0.613, "policy_loss": 0.005609, "value_loss": 165.125648, "entropy": 0.289114, "total_loss": 82.565542, "clip_frac": 0.076416, "sps": 436.6, "elapsed_s": 4166.5} |
| {"update": 891, "steps_done": 1824768, "mean_step_reward": 1.079, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 215.505, "mean_ep_length": 200.0, "mean_collisions": 3.5, "mean_intent_rate": 0.595, "policy_loss": 0.011764, "value_loss": 150.702313, "entropy": 0.282534, "total_loss": 75.360096, "clip_frac": 0.152466, "sps": 444.5, "elapsed_s": 4171.1} |
| {"update": 892, "steps_done": 1826816, "mean_step_reward": 1.2428, "min_step_reward": -7.5869, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 246.936, "mean_ep_length": 200.0, "mean_collisions": 2.18, "mean_intent_rate": 0.623, "policy_loss": 0.009659, "value_loss": 155.832194, "entropy": 0.306741, "total_loss": 77.922689, "clip_frac": 0.10022, "sps": 440.7, "elapsed_s": 4175.7} |
| {"update": 893, "steps_done": 1828864, "mean_step_reward": 1.2216, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 241.373, "mean_ep_length": 200.0, "mean_collisions": 2.5, "mean_intent_rate": 0.62, "policy_loss": 0.005608, "value_loss": 164.16991, "entropy": 0.296862, "total_loss": 82.087594, "clip_frac": 0.094727, "sps": 436.6, "elapsed_s": 4180.4} |
| {"update": 894, "steps_done": 1830912, "mean_step_reward": 1.2775, "min_step_reward": -7.5823, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 262.745, "mean_ep_length": 200.0, "mean_collisions": 0.9, "mean_intent_rate": 0.63, "policy_loss": 0.007699, "value_loss": 163.34592, "entropy": 0.341348, "total_loss": 81.677245, "clip_frac": 0.146729, "sps": 443.8, "elapsed_s": 4185.0} |
| {"update": 895, "steps_done": 1832960, "mean_step_reward": 1.2656, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 251.194, "mean_ep_length": 200.0, "mean_collisions": 1.0, "mean_intent_rate": 0.611, "policy_loss": 0.009479, "value_loss": 136.346531, "entropy": 0.321797, "total_loss": 68.179527, "clip_frac": 0.125488, "sps": 440.1, "elapsed_s": 4189.7} |
| {"update": 896, "steps_done": 1835008, "mean_step_reward": 1.1084, "min_step_reward": -7.588, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 224.098, "mean_ep_length": 200.0, "mean_collisions": 2.27, "mean_intent_rate": 0.588, "policy_loss": 0.009013, "value_loss": 147.419935, "entropy": 0.327939, "total_loss": 73.715702, "clip_frac": 0.123901, "sps": 440.6, "elapsed_s": 4194.3} |
| {"update": 897, "steps_done": 1837056, "mean_step_reward": 1.2476, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 228.898, "mean_ep_length": 182.6, "mean_collisions": 1.09, "mean_intent_rate": 0.561, "policy_loss": 0.004211, "value_loss": 156.025128, "entropy": 0.290211, "total_loss": 78.013873, "clip_frac": 0.125244, "sps": 430.8, "elapsed_s": 4199.1} |
| {"update": 898, "steps_done": 1839104, "mean_step_reward": 1.2114, "min_step_reward": -7.5853, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 241.508, "mean_ep_length": 200.0, "mean_collisions": 1.0, "mean_intent_rate": 0.595, "policy_loss": 0.012972, "value_loss": 148.793017, "entropy": 0.306071, "total_loss": 74.40642, "clip_frac": 0.135254, "sps": 444.0, "elapsed_s": 4203.7} |
| {"update": 899, "steps_done": 1841152, "mean_step_reward": 1.0674, "min_step_reward": -7.5771, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 222.954, "mean_ep_length": 200.0, "mean_collisions": 0.2, "mean_intent_rate": 0.551, "policy_loss": 0.013222, "value_loss": 138.24334, "entropy": 0.296009, "total_loss": 69.131932, "clip_frac": 0.162598, "sps": 434.8, "elapsed_s": 4208.4} |
| {"update": 900, "steps_done": 1843200, "mean_step_reward": 1.1808, "min_step_reward": -7.5814, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 226.896, "mean_ep_length": 200.0, "mean_collisions": 1.18, "mean_intent_rate": 0.575, "policy_loss": 0.005807, "value_loss": 148.1198, "entropy": 0.303975, "total_loss": 74.062667, "clip_frac": 0.095337, "sps": 436.0, "elapsed_s": 4213.1} |
| {"update": 901, "steps_done": 1845248, "mean_step_reward": 1.271, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 259.578, "mean_ep_length": 200.0, "mean_collisions": 0.4, "mean_intent_rate": 0.615, "policy_loss": 0.004653, "value_loss": 139.275897, "entropy": 0.240666, "total_loss": 69.640195, "clip_frac": 0.095703, "sps": 439.9, "elapsed_s": 4217.8} |
| {"update": 902, "steps_done": 1847296, "mean_step_reward": 1.2357, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 239.642, "mean_ep_length": 200.0, "mean_collisions": 0.2, "mean_intent_rate": 0.579, "policy_loss": 0.007103, "value_loss": 144.094945, "entropy": 0.286701, "total_loss": 72.051709, "clip_frac": 0.118774, "sps": 431.7, "elapsed_s": 4222.5} |
| {"update": 903, "steps_done": 1849344, "mean_step_reward": 1.2133, "min_step_reward": -7.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 238.879, "mean_ep_length": 200.0, "mean_collisions": 0.2, "mean_intent_rate": 0.578, "policy_loss": 0.009654, "value_loss": 154.039754, "entropy": 0.299081, "total_loss": 77.02654, "clip_frac": 0.171631, "sps": 438.7, "elapsed_s": 4227.2} |
| {"update": 904, "steps_done": 1851392, "mean_step_reward": 1.2848, "min_step_reward": -10.5334, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 262.864, "mean_ep_length": 200.0, "mean_collisions": 0.73, "mean_intent_rate": 0.627, "policy_loss": 0.006334, "value_loss": 153.859031, "entropy": 0.271782, "total_loss": 76.933131, "clip_frac": 0.101807, "sps": 435.2, "elapsed_s": 4231.9} |
| {"update": 905, "steps_done": 1853440, "mean_step_reward": 1.2757, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 250.541, "mean_ep_length": 200.0, "mean_collisions": 1.2, "mean_intent_rate": 0.613, "policy_loss": 0.002636, "value_loss": 145.885852, "entropy": 0.285231, "total_loss": 72.942709, "clip_frac": 0.076172, "sps": 439.8, "elapsed_s": 4236.6} |
| {"update": 906, "steps_done": 1855488, "mean_step_reward": 1.1334, "min_step_reward": -7.5515, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 230.844, "mean_ep_length": 200.0, "mean_collisions": 1.0, "mean_intent_rate": 0.577, "policy_loss": 0.005045, "value_loss": 134.047493, "entropy": 0.300155, "total_loss": 67.025789, "clip_frac": 0.096313, "sps": 435.1, "elapsed_s": 4241.3} |
| {"update": 907, "steps_done": 1857536, "mean_step_reward": 1.2776, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 255.622, "mean_ep_length": 200.0, "mean_collisions": 0.7, "mean_intent_rate": 0.613, "policy_loss": 0.01394, "value_loss": 128.961875, "entropy": 0.266285, "total_loss": 64.492215, "clip_frac": 0.170166, "sps": 437.7, "elapsed_s": 4246.0} |
| {"update": 908, "steps_done": 1859584, "mean_step_reward": 1.3133, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 261.07, "mean_ep_length": 200.0, "mean_collisions": 1.64, "mean_intent_rate": 0.639, "policy_loss": 0.02428, "value_loss": 163.342736, "entropy": 0.290319, "total_loss": 81.692745, "clip_frac": 0.163818, "sps": 435.3, "elapsed_s": 4250.7} |
| {"update": 909, "steps_done": 1861632, "mean_step_reward": 1.2351, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 244.348, "mean_ep_length": 200.0, "mean_collisions": 2.1, "mean_intent_rate": 0.618, "policy_loss": 0.011123, "value_loss": 155.570604, "entropy": 0.26485, "total_loss": 77.793776, "clip_frac": 0.107788, "sps": 443.0, "elapsed_s": 4255.3} |
| {"update": 910, "steps_done": 1863680, "mean_step_reward": 1.2555, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 248.034, "mean_ep_length": 200.0, "mean_collisions": 2.7, "mean_intent_rate": 0.635, "policy_loss": 0.008209, "value_loss": 169.106427, "entropy": 0.254806, "total_loss": 84.558874, "clip_frac": 0.106445, "sps": 439.5, "elapsed_s": 4259.9} |
| {"update": 911, "steps_done": 1865728, "mean_step_reward": 1.2778, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 266.731, "mean_ep_length": 200.0, "mean_collisions": 1.0, "mean_intent_rate": 0.638, "policy_loss": 0.006124, "value_loss": 145.556827, "entropy": 0.26873, "total_loss": 72.781849, "clip_frac": 0.087402, "sps": 438.0, "elapsed_s": 4264.6} |
| {"update": 912, "steps_done": 1867776, "mean_step_reward": 1.3827, "min_step_reward": -7.5869, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 262.831, "mean_ep_length": 200.0, "mean_collisions": 2.5, "mean_intent_rate": 0.656, "policy_loss": 0.006395, "value_loss": 148.380485, "entropy": 0.245378, "total_loss": 74.194185, "clip_frac": 0.0979, "sps": 443.8, "elapsed_s": 4269.2} |
| {"update": 913, "steps_done": 1869824, "mean_step_reward": 1.4899, "min_step_reward": -7.5842, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 303.595, "mean_ep_length": 200.0, "mean_collisions": 1.45, "mean_intent_rate": 0.707, "policy_loss": 0.006393, "value_loss": 163.111166, "entropy": 0.218727, "total_loss": 81.559789, "clip_frac": 0.072021, "sps": 441.0, "elapsed_s": 4273.9} |
| {"update": 914, "steps_done": 1871872, "mean_step_reward": 1.416, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 277.808, "mean_ep_length": 200.0, "mean_collisions": 2.5, "mean_intent_rate": 0.681, "policy_loss": 0.008127, "value_loss": 178.275698, "entropy": 0.217175, "total_loss": 89.143803, "clip_frac": 0.097168, "sps": 431.8, "elapsed_s": 4278.6} |
| {"update": 915, "steps_done": 1873920, "mean_step_reward": 1.2848, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 260.801, "mean_ep_length": 200.0, "mean_collisions": 2.8, "mean_intent_rate": 0.657, "policy_loss": 0.006952, "value_loss": 170.885391, "entropy": 0.238461, "total_loss": 85.447263, "clip_frac": 0.119507, "sps": 443.8, "elapsed_s": 4283.2} |
| {"update": 916, "steps_done": 1875968, "mean_step_reward": 1.3175, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 262.552, "mean_ep_length": 200.0, "mean_collisions": 4.6, "mean_intent_rate": 0.692, "policy_loss": 0.01441, "value_loss": 189.628411, "entropy": 0.22759, "total_loss": 94.826341, "clip_frac": 0.150146, "sps": 439.3, "elapsed_s": 4287.9} |
| {"update": 917, "steps_done": 1878016, "mean_step_reward": 0.8704, "min_step_reward": -10.5489, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 181.501, "mean_ep_length": 200.0, "mean_collisions": 13.55, "mean_intent_rate": 0.707, "policy_loss": 0.006309, "value_loss": 240.305321, "entropy": 0.236588, "total_loss": 120.156604, "clip_frac": 0.120972, "sps": 440.9, "elapsed_s": 4292.6} |
| {"update": 918, "steps_done": 1880064, "mean_step_reward": 0.987, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 191.522, "mean_ep_length": 200.0, "mean_collisions": 13.4, "mean_intent_rate": 0.722, "policy_loss": 0.013402, "value_loss": 226.270278, "entropy": 0.237601, "total_loss": 113.146165, "clip_frac": 0.133301, "sps": 441.8, "elapsed_s": 4297.2} |
| {"update": 919, "steps_done": 1882112, "mean_step_reward": 1.023, "min_step_reward": -10.5707, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 210.194, "mean_ep_length": 200.0, "mean_collisions": 13.2, "mean_intent_rate": 0.748, "policy_loss": 0.002475, "value_loss": 186.718996, "entropy": 0.22691, "total_loss": 93.359704, "clip_frac": 0.094971, "sps": 440.7, "elapsed_s": 4301.8} |
| {"update": 920, "steps_done": 1884160, "mean_step_reward": 0.7893, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 157.304, "mean_ep_length": 200.0, "mean_collisions": 15.8, "mean_intent_rate": 0.705, "policy_loss": 0.031924, "value_loss": 173.59965, "entropy": 0.277557, "total_loss": 86.828974, "clip_frac": 0.251587, "sps": 436.6, "elapsed_s": 4306.5} |
| {"update": 921, "steps_done": 1886208, "mean_step_reward": 0.8769, "min_step_reward": -10.5556, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 179.122, "mean_ep_length": 200.0, "mean_collisions": 11.82, "mean_intent_rate": 0.675, "policy_loss": 0.017689, "value_loss": 152.624631, "entropy": 0.268254, "total_loss": 76.327322, "clip_frac": 0.130615, "sps": 441.3, "elapsed_s": 4311.2} |
| {"update": 922, "steps_done": 1888256, "mean_step_reward": 0.7756, "min_step_reward": -10.5192, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 150.738, "mean_ep_length": 200.0, "mean_collisions": 15.5, "mean_intent_rate": 0.69, "policy_loss": 0.007882, "value_loss": 156.58024, "entropy": 0.293328, "total_loss": 78.295069, "clip_frac": 0.139282, "sps": 435.3, "elapsed_s": 4315.9} |
| {"update": 923, "steps_done": 1890304, "mean_step_reward": 0.9194, "min_step_reward": -10.5554, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 189.441, "mean_ep_length": 200.0, "mean_collisions": 13.2, "mean_intent_rate": 0.714, "policy_loss": 0.01567, "value_loss": 165.357461, "entropy": 0.282575, "total_loss": 82.691576, "clip_frac": 0.158081, "sps": 427.2, "elapsed_s": 4320.7} |
| {"update": 924, "steps_done": 1892352, "mean_step_reward": 0.771, "min_step_reward": -10.5184, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 160.104, "mean_ep_length": 200.0, "mean_collisions": 16.4, "mean_intent_rate": 0.719, "policy_loss": 0.010725, "value_loss": 229.121214, "entropy": 0.298169, "total_loss": 114.56835, "clip_frac": 0.172119, "sps": 442.8, "elapsed_s": 4325.3} |
| {"update": 925, "steps_done": 1894400, "mean_step_reward": 0.5317, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 103.45, "mean_ep_length": 200.0, "mean_collisions": 21.64, "mean_intent_rate": 0.715, "policy_loss": 0.003012, "value_loss": 164.42714, "entropy": 0.295999, "total_loss": 82.213621, "clip_frac": 0.110474, "sps": 437.5, "elapsed_s": 4330.0} |
| {"update": 926, "steps_done": 1896448, "mean_step_reward": 0.7382, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 143.662, "mean_ep_length": 200.0, "mean_collisions": 17.3, "mean_intent_rate": 0.708, "policy_loss": 0.008804, "value_loss": 142.585987, "entropy": 0.291099, "total_loss": 71.298886, "clip_frac": 0.08374, "sps": 437.7, "elapsed_s": 4334.7} |
| {"update": 927, "steps_done": 1898496, "mean_step_reward": 0.8534, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 167.989, "mean_ep_length": 200.0, "mean_collisions": 15.3, "mean_intent_rate": 0.714, "policy_loss": 0.017706, "value_loss": 121.520046, "entropy": 0.27372, "total_loss": 60.774992, "clip_frac": 0.197754, "sps": 438.5, "elapsed_s": 4339.3} |
| {"update": 928, "steps_done": 1900544, "mean_step_reward": 0.4999, "min_step_reward": -10.5554, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 102.292, "mean_ep_length": 193.7, "mean_collisions": 20.18, "mean_intent_rate": 0.696, "policy_loss": 0.005441, "value_loss": 200.032919, "entropy": 0.276068, "total_loss": 100.01914, "clip_frac": 0.106201, "sps": 437.8, "elapsed_s": 4344.0} |
| {"update": 929, "steps_done": 1902592, "mean_step_reward": 0.7607, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 149.69, "mean_ep_length": 199.4, "mean_collisions": 16.3, "mean_intent_rate": 0.701, "policy_loss": 0.006288, "value_loss": 137.618157, "entropy": 0.278821, "total_loss": 68.812578, "clip_frac": 0.090454, "sps": 434.8, "elapsed_s": 4348.7} |
| {"update": 930, "steps_done": 1904640, "mean_step_reward": 0.7901, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 162.106, "mean_ep_length": 200.0, "mean_collisions": 15.6, "mean_intent_rate": 0.71, "policy_loss": 0.014652, "value_loss": 168.929006, "entropy": 0.2886, "total_loss": 84.476269, "clip_frac": 0.127808, "sps": 437.5, "elapsed_s": 4353.4} |
| {"update": 931, "steps_done": 1906688, "mean_step_reward": 0.7588, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 146.293, "mean_ep_length": 200.0, "mean_collisions": 18.2, "mean_intent_rate": 0.725, "policy_loss": 0.01067, "value_loss": 167.678143, "entropy": 0.291368, "total_loss": 83.846828, "clip_frac": 0.150757, "sps": 438.5, "elapsed_s": 4358.1} |
| {"update": 932, "steps_done": 1908736, "mean_step_reward": 0.82, "min_step_reward": -10.5465, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 170.301, "mean_ep_length": 200.0, "mean_collisions": 15.18, "mean_intent_rate": 0.715, "policy_loss": 0.024768, "value_loss": 147.138285, "entropy": 0.301918, "total_loss": 73.590891, "clip_frac": 0.167847, "sps": 431.7, "elapsed_s": 4362.8} |
| {"update": 933, "steps_done": 1910784, "mean_step_reward": 0.9365, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 183.656, "mean_ep_length": 200.0, "mean_collisions": 14.1, "mean_intent_rate": 0.719, "policy_loss": 0.00169, "value_loss": 125.499363, "entropy": 0.285184, "total_loss": 62.748519, "clip_frac": 0.074951, "sps": 437.6, "elapsed_s": 4367.5} |
| {"update": 934, "steps_done": 1912832, "mean_step_reward": 1.0421, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 200.891, "mean_ep_length": 196.5, "mean_collisions": 9.8, "mean_intent_rate": 0.685, "policy_loss": 0.002095, "value_loss": 109.430739, "entropy": 0.287108, "total_loss": 54.714593, "clip_frac": 0.068604, "sps": 441.0, "elapsed_s": 4372.1} |
| {"update": 935, "steps_done": 1914880, "mean_step_reward": 0.8229, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 176.524, "mean_ep_length": 197.9, "mean_collisions": 13.09, "mean_intent_rate": 0.697, "policy_loss": 0.011667, "value_loss": 127.537348, "entropy": 0.28535, "total_loss": 63.777487, "clip_frac": 0.142578, "sps": 439.9, "elapsed_s": 4376.8} |
| {"update": 936, "steps_done": 1916928, "mean_step_reward": 0.7637, "min_step_reward": -10.5815, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 147.302, "mean_ep_length": 200.0, "mean_collisions": 16.5, "mean_intent_rate": 0.7, "policy_loss": 0.00745, "value_loss": 135.96162, "entropy": 0.264963, "total_loss": 67.98561, "clip_frac": 0.084839, "sps": 437.2, "elapsed_s": 4381.5} |
| {"update": 937, "steps_done": 1918976, "mean_step_reward": 0.7852, "min_step_reward": -10.5724, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 160.358, "mean_ep_length": 200.0, "mean_collisions": 15.3, "mean_intent_rate": 0.701, "policy_loss": 0.015079, "value_loss": 113.327858, "entropy": 0.280547, "total_loss": 56.676203, "clip_frac": 0.14563, "sps": 442.0, "elapsed_s": 4386.1} |
| {"update": 938, "steps_done": 1921024, "mean_step_reward": 0.952, "min_step_reward": -10.544, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 193.528, "mean_ep_length": 200.0, "mean_collisions": 10.7, "mean_intent_rate": 0.679, "policy_loss": 0.018447, "value_loss": 115.0932, "entropy": 0.261769, "total_loss": 57.56243, "clip_frac": 0.166138, "sps": 451.8, "elapsed_s": 4390.7} |
| {"update": 939, "steps_done": 1923072, "mean_step_reward": 1.0965, "min_step_reward": -10.5502, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 207.798, "mean_ep_length": 200.0, "mean_collisions": 8.0, "mean_intent_rate": 0.657, "policy_loss": 0.011056, "value_loss": 121.160799, "entropy": 0.274948, "total_loss": 60.588706, "clip_frac": 0.12854, "sps": 443.9, "elapsed_s": 4395.3} |
| {"update": 940, "steps_done": 1925120, "mean_step_reward": 1.0549, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 207.492, "mean_ep_length": 200.0, "mean_collisions": 9.7, "mean_intent_rate": 0.687, "policy_loss": 0.017097, "value_loss": 118.029062, "entropy": 0.261468, "total_loss": 59.029013, "clip_frac": 0.167725, "sps": 438.9, "elapsed_s": 4399.9} |
| {"update": 941, "steps_done": 1927168, "mean_step_reward": 1.0253, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 208.211, "mean_ep_length": 200.0, "mean_collisions": 9.8, "mean_intent_rate": 0.69, "policy_loss": 0.005266, "value_loss": 158.107482, "entropy": 0.255357, "total_loss": 79.056453, "clip_frac": 0.109009, "sps": 431.0, "elapsed_s": 4404.7} |
| {"update": 942, "steps_done": 1929216, "mean_step_reward": 0.9857, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 197.306, "mean_ep_length": 200.0, "mean_collisions": 10.7, "mean_intent_rate": 0.685, "policy_loss": 0.002557, "value_loss": 134.929507, "entropy": 0.292563, "total_loss": 67.464384, "clip_frac": 0.091675, "sps": 438.6, "elapsed_s": 4409.4} |
| {"update": 943, "steps_done": 1931264, "mean_step_reward": 1.1446, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 227.499, "mean_ep_length": 200.0, "mean_collisions": 9.0, "mean_intent_rate": 0.707, "policy_loss": 0.008278, "value_loss": 141.585409, "entropy": 0.238917, "total_loss": 70.798592, "clip_frac": 0.107788, "sps": 444.1, "elapsed_s": 4414.0} |
| {"update": 944, "steps_done": 1933312, "mean_step_reward": 1.156, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 229.81, "mean_ep_length": 200.0, "mean_collisions": 10.1, "mean_intent_rate": 0.73, "policy_loss": 0.006006, "value_loss": 126.777179, "entropy": 0.238907, "total_loss": 63.392207, "clip_frac": 0.103149, "sps": 439.7, "elapsed_s": 4418.6} |
| {"update": 945, "steps_done": 1935360, "mean_step_reward": 1.0292, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 199.697, "mean_ep_length": 200.0, "mean_collisions": 11.9, "mean_intent_rate": 0.71, "policy_loss": 0.009995, "value_loss": 166.994398, "entropy": 0.237157, "total_loss": 83.504823, "clip_frac": 0.063599, "sps": 448.7, "elapsed_s": 4423.2} |
| {"update": 946, "steps_done": 1937408, "mean_step_reward": 0.7836, "min_step_reward": -10.5006, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 176.328, "mean_ep_length": 200.0, "mean_collisions": 13.3, "mean_intent_rate": 0.696, "policy_loss": 0.002829, "value_loss": 150.404691, "entropy": 0.274799, "total_loss": 75.202428, "clip_frac": 0.054443, "sps": 436.6, "elapsed_s": 4427.9} |
| {"update": 947, "steps_done": 1939456, "mean_step_reward": 0.8851, "min_step_reward": -10.5229, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 153.051, "mean_ep_length": 186.4, "mean_collisions": 15.55, "mean_intent_rate": 0.729, "policy_loss": 0.014906, "value_loss": 175.65579, "entropy": 0.293914, "total_loss": 87.839862, "clip_frac": 0.202026, "sps": 439.3, "elapsed_s": 4432.5} |
| {"update": 948, "steps_done": 1941504, "mean_step_reward": 1.292, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 254.543, "mean_ep_length": 200.0, "mean_collisions": 6.82, "mean_intent_rate": 0.716, "policy_loss": 0.016105, "value_loss": 150.573352, "entropy": 0.27689, "total_loss": 75.300012, "clip_frac": 0.10791, "sps": 442.4, "elapsed_s": 4437.2} |
| {"update": 949, "steps_done": 1943552, "mean_step_reward": 1.1195, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 225.002, "mean_ep_length": 200.0, "mean_collisions": 9.8, "mean_intent_rate": 0.72, "policy_loss": 0.004866, "value_loss": 152.676785, "entropy": 0.271669, "total_loss": 76.340541, "clip_frac": 0.127441, "sps": 432.7, "elapsed_s": 4441.9} |
| {"update": 950, "steps_done": 1945600, "mean_step_reward": 0.743, "min_step_reward": -10.5864, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 148.556, "mean_ep_length": 200.0, "mean_collisions": 11.9, "mean_intent_rate": 0.626, "policy_loss": 0.016323, "value_loss": 141.073488, "entropy": 0.275988, "total_loss": 70.550307, "clip_frac": 0.131958, "sps": 435.5, "elapsed_s": 4446.6} |
| {"update": 951, "steps_done": 1947648, "mean_step_reward": 1.3059, "min_step_reward": -10.5735, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 250.824, "mean_ep_length": 200.0, "mean_collisions": 4.9, "mean_intent_rate": 0.675, "policy_loss": 0.015455, "value_loss": 115.436126, "entropy": 0.26913, "total_loss": 57.730827, "clip_frac": 0.146606, "sps": 440.3, "elapsed_s": 4451.3} |
| {"update": 952, "steps_done": 1949696, "mean_step_reward": 1.2705, "min_step_reward": -10.5656, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 260.352, "mean_ep_length": 200.0, "mean_collisions": 5.27, "mean_intent_rate": 0.7, "policy_loss": 0.006162, "value_loss": 138.876917, "entropy": 0.275636, "total_loss": 69.441863, "clip_frac": 0.115479, "sps": 444.2, "elapsed_s": 4455.9} |
| {"update": 953, "steps_done": 1951744, "mean_step_reward": 1.0706, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 213.338, "mean_ep_length": 200.0, "mean_collisions": 8.6, "mean_intent_rate": 0.678, "policy_loss": 0.009359, "value_loss": 132.098106, "entropy": 0.287903, "total_loss": 66.055534, "clip_frac": 0.128052, "sps": 440.9, "elapsed_s": 4460.5} |
| {"update": 954, "steps_done": 1953792, "mean_step_reward": 0.8396, "min_step_reward": -10.5722, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 161.952, "mean_ep_length": 200.0, "mean_collisions": 13.0, "mean_intent_rate": 0.665, "policy_loss": 0.004259, "value_loss": 173.138855, "entropy": 0.26091, "total_loss": 86.571078, "clip_frac": 0.103882, "sps": 441.3, "elapsed_s": 4465.2} |
| {"update": 955, "steps_done": 1955840, "mean_step_reward": 1.2645, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 236.073, "mean_ep_length": 185.5, "mean_collisions": 6.73, "mean_intent_rate": 0.71, "policy_loss": 0.007693, "value_loss": 159.893821, "entropy": 0.250593, "total_loss": 79.952098, "clip_frac": 0.10791, "sps": 436.8, "elapsed_s": 4469.9} |
| {"update": 956, "steps_done": 1957888, "mean_step_reward": 1.3221, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 265.302, "mean_ep_length": 200.0, "mean_collisions": 5.5, "mean_intent_rate": 0.711, "policy_loss": 0.008246, "value_loss": 140.030863, "entropy": 0.252011, "total_loss": 70.021157, "clip_frac": 0.096069, "sps": 439.5, "elapsed_s": 4474.5} |
| {"update": 957, "steps_done": 1959936, "mean_step_reward": 0.9097, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 188.708, "mean_ep_length": 200.0, "mean_collisions": 12.5, "mean_intent_rate": 0.702, "policy_loss": 0.010547, "value_loss": 160.912717, "entropy": 0.291073, "total_loss": 80.463994, "clip_frac": 0.137085, "sps": 437.5, "elapsed_s": 4479.2} |
| {"update": 958, "steps_done": 1961984, "mean_step_reward": 0.8162, "min_step_reward": -10.5451, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 159.976, "mean_ep_length": 200.0, "mean_collisions": 13.27, "mean_intent_rate": 0.668, "policy_loss": 0.011053, "value_loss": 173.979735, "entropy": 0.287786, "total_loss": 86.998042, "clip_frac": 0.119263, "sps": 441.0, "elapsed_s": 4483.9} |
| {"update": 959, "steps_done": 1964032, "mean_step_reward": 1.1339, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 224.218, "mean_ep_length": 200.0, "mean_collisions": 8.3, "mean_intent_rate": 0.691, "policy_loss": 0.005069, "value_loss": 138.965354, "entropy": 0.272873, "total_loss": 69.485018, "clip_frac": 0.091431, "sps": 435.1, "elapsed_s": 4488.6} |
| {"update": 960, "steps_done": 1966080, "mean_step_reward": 1.1707, "min_step_reward": -10.5353, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 237.552, "mean_ep_length": 200.0, "mean_collisions": 6.3, "mean_intent_rate": 0.678, "policy_loss": 0.006551, "value_loss": 120.343594, "entropy": 0.278566, "total_loss": 60.175563, "clip_frac": 0.120483, "sps": 437.5, "elapsed_s": 4493.2} |
| {"update": 961, "steps_done": 1968128, "mean_step_reward": 1.3058, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 261.677, "mean_ep_length": 200.0, "mean_collisions": 6.1, "mean_intent_rate": 0.717, "policy_loss": 0.009537, "value_loss": 133.518413, "entropy": 0.262313, "total_loss": 66.766119, "clip_frac": 0.122437, "sps": 436.0, "elapsed_s": 4497.9} |
| {"update": 962, "steps_done": 1970176, "mean_step_reward": 0.9041, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 188.726, "mean_ep_length": 200.0, "mean_collisions": 13.18, "mean_intent_rate": 0.714, "policy_loss": 0.000724, "value_loss": 190.964342, "entropy": 0.284802, "total_loss": 95.480047, "clip_frac": 0.135864, "sps": 437.2, "elapsed_s": 4502.6} |
| {"update": 963, "steps_done": 1972224, "mean_step_reward": 1.1929, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 236.754, "mean_ep_length": 200.0, "mean_collisions": 5.2, "mean_intent_rate": 0.66, "policy_loss": 0.007212, "value_loss": 143.369314, "entropy": 0.286653, "total_loss": 71.689002, "clip_frac": 0.106445, "sps": 440.2, "elapsed_s": 4507.3} |
| {"update": 964, "steps_done": 1974272, "mean_step_reward": 1.1228, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 224.441, "mean_ep_length": 200.0, "mean_collisions": 7.9, "mean_intent_rate": 0.683, "policy_loss": 0.007757, "value_loss": 142.980797, "entropy": 0.270809, "total_loss": 71.495448, "clip_frac": 0.106567, "sps": 443.1, "elapsed_s": 4511.9} |
| {"update": 965, "steps_done": 1976320, "mean_step_reward": 1.1555, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 228.734, "mean_ep_length": 200.0, "mean_collisions": 9.0, "mean_intent_rate": 0.709, "policy_loss": 0.002833, "value_loss": 150.631091, "entropy": 0.265257, "total_loss": 75.315725, "clip_frac": 0.089844, "sps": 439.5, "elapsed_s": 4516.6} |
| {"update": 966, "steps_done": 1978368, "mean_step_reward": 1.3166, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 263.571, "mean_ep_length": 200.0, "mean_collisions": 5.82, "mean_intent_rate": 0.714, "policy_loss": 0.00437, "value_loss": 167.469646, "entropy": 0.258097, "total_loss": 83.736612, "clip_frac": 0.092407, "sps": 444.0, "elapsed_s": 4521.2} |
| {"update": 967, "steps_done": 1980416, "mean_step_reward": 0.9213, "min_step_reward": -10.5811, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 188.458, "mean_ep_length": 200.0, "mean_collisions": 12.1, "mean_intent_rate": 0.694, "policy_loss": 0.001815, "value_loss": 155.929606, "entropy": 0.263739, "total_loss": 77.963981, "clip_frac": 0.083252, "sps": 433.4, "elapsed_s": 4525.9} |
| {"update": 968, "steps_done": 1982464, "mean_step_reward": 1.1294, "min_step_reward": -7.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 214.48, "mean_ep_length": 200.0, "mean_collisions": 9.5, "mean_intent_rate": 0.694, "policy_loss": 0.008447, "value_loss": 139.649209, "entropy": 0.274671, "total_loss": 69.830304, "clip_frac": 0.108765, "sps": 428.7, "elapsed_s": 4530.7} |
| {"update": 969, "steps_done": 1984512, "mean_step_reward": 1.0306, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 214.219, "mean_ep_length": 200.0, "mean_collisions": 11.0, "mean_intent_rate": 0.717, "policy_loss": 0.006175, "value_loss": 144.085573, "entropy": 0.266007, "total_loss": 72.046302, "clip_frac": 0.084106, "sps": 434.5, "elapsed_s": 4535.4} |
| {"update": 970, "steps_done": 1986560, "mean_step_reward": 0.4358, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 92.57, "mean_ep_length": 200.0, "mean_collisions": 21.45, "mean_intent_rate": 0.693, "policy_loss": 0.009933, "value_loss": 283.806307, "entropy": 0.304804, "total_loss": 141.910038, "clip_frac": 0.141602, "sps": 435.4, "elapsed_s": 4540.1} |
| {"update": 971, "steps_done": 1988608, "mean_step_reward": 1.1794, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 236.623, "mean_ep_length": 200.0, "mean_collisions": 9.7, "mean_intent_rate": 0.734, "policy_loss": 0.006111, "value_loss": 145.054918, "entropy": 0.263365, "total_loss": 72.530937, "clip_frac": 0.099854, "sps": 442.1, "elapsed_s": 4544.7} |
| {"update": 972, "steps_done": 1990656, "mean_step_reward": 1.0361, "min_step_reward": -10.5506, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 206.575, "mean_ep_length": 200.0, "mean_collisions": 11.4, "mean_intent_rate": 0.713, "policy_loss": 0.001587, "value_loss": 138.598083, "entropy": 0.286741, "total_loss": 69.297761, "clip_frac": 0.054321, "sps": 444.6, "elapsed_s": 4549.3} |
| {"update": 973, "steps_done": 1992704, "mean_step_reward": 1.06, "min_step_reward": -10.5, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 203.644, "mean_ep_length": 200.0, "mean_collisions": 10.0, "mean_intent_rate": 0.684, "policy_loss": 0.003841, "value_loss": 137.422974, "entropy": 0.282468, "total_loss": 68.712504, "clip_frac": 0.094727, "sps": 437.8, "elapsed_s": 4554.0} |
| {"update": 974, "steps_done": 1994752, "mean_step_reward": 1.2066, "min_step_reward": -10.599, "max_step_reward": 2.5, "n_episodes_finished": 11, "mean_ep_return": 248.404, "mean_ep_length": 200.0, "mean_collisions": 6.18, "mean_intent_rate": 0.696, "policy_loss": 0.021298, "value_loss": 149.264186, "entropy": 0.287592, "total_loss": 74.650515, "clip_frac": 0.141113, "sps": 431.4, "elapsed_s": 4558.8} |
| {"update": 975, "steps_done": 1996800, "mean_step_reward": 0.7028, "min_step_reward": -10.5838, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 137.056, "mean_ep_length": 200.0, "mean_collisions": 18.2, "mean_intent_rate": 0.713, "policy_loss": 0.006305, "value_loss": 179.316981, "entropy": 0.286028, "total_loss": 89.661936, "clip_frac": 0.11499, "sps": 437.2, "elapsed_s": 4563.5} |
| {"update": 976, "steps_done": 1998848, "mean_step_reward": 0.865, "min_step_reward": -10.5554, "max_step_reward": 2.5, "n_episodes_finished": 10, "mean_ep_return": 174.71, "mean_ep_length": 200.0, "mean_collisions": 13.9, "mean_intent_rate": 0.701, "policy_loss": 0.011259, "value_loss": 140.063934, "entropy": 0.252669, "total_loss": 70.0407, "clip_frac": 0.12439, "sps": 427.8, "elapsed_s": 4568.2} |
| |
