diff --git "a/sf_log.txt" "b/sf_log.txt" --- "a/sf_log.txt" +++ "b/sf_log.txt" @@ -1,50 +1,34 @@ -[2023-04-27 22:28:30,298][19320] Saving configuration to /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/config.json... -[2023-04-27 22:28:30,299][19320] Rollout worker 0 uses device cpu -[2023-04-27 22:28:30,300][19320] Rollout worker 1 uses device cpu -[2023-04-27 22:28:30,300][19320] Rollout worker 2 uses device cpu -[2023-04-27 22:28:30,302][19320] Rollout worker 3 uses device cpu -[2023-04-27 22:28:30,302][19320] Rollout worker 4 uses device cpu -[2023-04-27 22:28:30,303][19320] Rollout worker 5 uses device cpu -[2023-04-27 22:28:30,304][19320] Rollout worker 6 uses device cpu -[2023-04-27 22:28:30,304][19320] Rollout worker 7 uses device cpu -[2023-04-27 22:28:30,345][19320] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-27 22:28:30,345][19320] InferenceWorker_p0-w0: min num requests: 2 -[2023-04-27 22:28:30,363][19320] Starting all processes... -[2023-04-27 22:28:30,364][19320] Starting process learner_proc0 -[2023-04-27 22:28:30,489][19320] Starting all processes... -[2023-04-27 22:28:30,494][19320] Starting process inference_proc0-0 -[2023-04-27 22:28:30,494][19320] Starting process rollout_proc0 -[2023-04-27 22:28:30,495][19320] Starting process rollout_proc1 -[2023-04-27 22:28:30,495][19320] Starting process rollout_proc2 -[2023-04-27 22:28:30,496][19320] Starting process rollout_proc3 -[2023-04-27 22:28:30,496][19320] Starting process rollout_proc4 -[2023-04-27 22:28:30,496][19320] Starting process rollout_proc5 -[2023-04-27 22:28:30,497][19320] Starting process rollout_proc6 -[2023-04-27 22:28:30,497][19320] Starting process rollout_proc7 -[2023-04-27 22:28:31,380][26612] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-27 22:28:31,380][26612] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0 -[2023-04-27 22:28:31,403][26612] Num visible devices: 1 -[2023-04-27 22:28:31,424][26630] Worker 3 uses CPU cores [9, 10, 11] -[2023-04-27 22:28:31,427][26612] Starting seed is not provided -[2023-04-27 22:28:31,427][26612] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-27 22:28:31,427][26612] Initializing actor-critic model on device cuda:0 -[2023-04-27 22:28:31,427][26612] RunningMeanStd input shape: (3, 72, 128) -[2023-04-27 22:28:31,428][26612] RunningMeanStd input shape: (1,) -[2023-04-27 22:28:31,430][26629] Worker 2 uses CPU cores [6, 7, 8] -[2023-04-27 22:28:31,434][26628] Worker 1 uses CPU cores [3, 4, 5] -[2023-04-27 22:28:31,439][26612] ConvEncoder: input_channels=3 -[2023-04-27 22:28:31,444][26638] Worker 7 uses CPU cores [21, 22, 23] -[2023-04-27 22:28:31,455][26626] Worker 0 uses CPU cores [0, 1, 2] -[2023-04-27 22:28:31,457][26632] Worker 5 uses CPU cores [15, 16, 17] -[2023-04-27 22:28:31,467][26631] Worker 4 uses CPU cores [12, 13, 14] -[2023-04-27 22:28:31,474][26627] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-27 22:28:31,474][26627] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0 -[2023-04-27 22:28:31,481][26657] Worker 6 uses CPU cores [18, 19, 20] -[2023-04-27 22:28:31,507][26627] Num visible devices: 1 -[2023-04-27 22:28:31,580][26612] Conv encoder output size: 512 -[2023-04-27 22:28:31,581][26612] Policy head output size: 512 -[2023-04-27 22:28:31,603][26612] Created Actor Critic model with architecture: -[2023-04-27 22:28:31,603][26612] ActorCriticSharedWeights( +[2023-04-30 12:43:57,803][678550] Saving configuration to ./train_dir/doom_health_gathering_supreme/config.json... +[2023-04-30 12:43:57,803][678550] Rollout worker 0 uses device cpu +[2023-04-30 12:43:57,803][678550] Rollout worker 1 uses device cpu +[2023-04-30 12:43:57,804][678550] Rollout worker 2 uses device cpu +[2023-04-30 12:43:57,804][678550] Rollout worker 3 uses device cpu +[2023-04-30 12:43:57,804][678550] Rollout worker 4 uses device cpu +[2023-04-30 12:43:57,804][678550] Rollout worker 5 uses device cpu +[2023-04-30 12:43:57,804][678550] Rollout worker 6 uses device cpu +[2023-04-30 12:43:57,804][678550] Rollout worker 7 uses device cpu +[2023-04-30 12:43:57,858][678550] InferenceWorker_p0-w0: min num requests: 2 +[2023-04-30 12:43:57,999][678550] Starting all processes... +[2023-04-30 12:43:57,999][678550] Starting process learner_proc0 +[2023-04-30 12:43:58,808][678550] Starting all processes... +[2023-04-30 12:43:58,813][678550] Starting process inference_proc0-0 +[2023-04-30 12:43:58,813][678550] Starting process rollout_proc0 +[2023-04-30 12:43:58,814][678641] Starting seed is not provided +[2023-04-30 12:43:58,814][678641] Initializing actor-critic model on device cpu +[2023-04-30 12:43:58,814][678641] RunningMeanStd input shape: (3, 72, 128) +[2023-04-30 12:43:58,815][678641] RunningMeanStd input shape: (1,) +[2023-04-30 12:43:58,813][678550] Starting process rollout_proc1 +[2023-04-30 12:43:58,814][678550] Starting process rollout_proc2 +[2023-04-30 12:43:58,822][678641] ConvEncoder: input_channels=3 +[2023-04-30 12:43:58,818][678550] Starting process rollout_proc3 +[2023-04-30 12:43:58,820][678550] Starting process rollout_proc4 +[2023-04-30 12:43:58,821][678550] Starting process rollout_proc5 +[2023-04-30 12:43:58,824][678550] Starting process rollout_proc6 +[2023-04-30 12:43:58,827][678550] Starting process rollout_proc7 +[2023-04-30 12:43:58,928][678641] Conv encoder output size: 512 +[2023-04-30 12:43:58,929][678641] Policy head output size: 512 +[2023-04-30 12:43:58,948][678641] Created Actor Critic model with architecture: +[2023-04-30 12:43:58,948][678641] ActorCriticSharedWeights( (obs_normalizer): ObservationNormalizer( (running_mean_std): RunningMeanStdDictInPlace( (running_mean_std): ModuleDict( @@ -85,591 +69,216 @@ (distribution_linear): Linear(in_features=512, out_features=5, bias=True) ) ) -[2023-04-27 22:28:33,372][26612] Using optimizer -[2023-04-27 22:28:33,373][26612] No checkpoints found -[2023-04-27 22:28:33,373][26612] Did not load from checkpoint, starting from scratch! -[2023-04-27 22:28:33,373][26612] Initialized policy 0 weights for model version 0 -[2023-04-27 22:28:33,376][26612] LearnerWorker_p0 finished initialization! -[2023-04-27 22:28:33,376][26612] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-27 22:28:33,815][19320] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 0. Throughput: 0: nan. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) -[2023-04-27 22:28:34,070][26627] RunningMeanStd input shape: (3, 72, 128) -[2023-04-27 22:28:34,071][26627] RunningMeanStd input shape: (1,) -[2023-04-27 22:28:34,079][26627] ConvEncoder: input_channels=3 -[2023-04-27 22:28:34,158][26627] Conv encoder output size: 512 -[2023-04-27 22:28:34,159][26627] Policy head output size: 512 -[2023-04-27 22:28:34,896][19320] Inference worker 0-0 is ready! -[2023-04-27 22:28:34,897][19320] All inference workers are ready! Signal rollout workers to start! -[2023-04-27 22:28:34,912][26657] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-27 22:28:34,913][26630] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-27 22:28:34,914][26638] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-27 22:28:34,914][26629] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-27 22:28:34,914][26626] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-27 22:28:34,915][26632] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-27 22:28:34,915][26631] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-27 22:28:34,915][26628] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-27 22:28:34,949][26628] VizDoom game.init() threw an exception ViZDoomUnexpectedExitException('Controlled ViZDoom instance exited unexpectedly.'). Terminate process... -[2023-04-27 22:28:34,950][26628] EvtLoop [rollout_proc1_evt_loop, process=rollout_proc1] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=() -Traceback (most recent call last): - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 228, in _game_init - self.game.init() -vizdoom.vizdoom.ViZDoomUnexpectedExitException: Controlled ViZDoom instance exited unexpectedly. - -During handling of the above exception, another exception occurred: - -Traceback (most recent call last): - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal - slot_callable(*args) - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init - env_runner.init(self.timing) - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init - self._reset() - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset - observations, info = e.reset(seed=seed) # new way of doing seeding since Gym 0.26.0 - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/gym/core.py", line 323, in reset - return self.env.reset(**kwargs) - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset - obs, info = self.env.reset(**kwargs) - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset - obs, info = self.env.reset(**kwargs) - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset - return self.env.reset(**kwargs) - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/gym/core.py", line 379, in reset - obs, info = self.env.reset(**kwargs) - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/sample_factory/envs/env_wrappers.py", line 84, in reset - obs, info = self.env.reset(**kwargs) - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/gym/core.py", line 323, in reset - return self.env.reset(**kwargs) - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/sf_examples/vizdoom/doom/wrappers/multiplayer_stats.py", line 51, in reset - return self.env.reset(**kwargs) - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 323, in reset - self._ensure_initialized() - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 274, in _ensure_initialized - self.initialize() - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 269, in initialize - self._game_init() - File "/home/byron/miniconda3/envs/ml-agents/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 244, in _game_init - raise EnvCriticalError() -sample_factory.envs.env_utils.EnvCriticalError -[2023-04-27 22:28:34,951][26628] Unhandled exception in evt loop rollout_proc1_evt_loop -[2023-04-27 22:28:35,135][26629] Decorrelating experience for 0 frames... -[2023-04-27 22:28:35,135][26630] Decorrelating experience for 0 frames... -[2023-04-27 22:28:35,135][26657] Decorrelating experience for 0 frames... -[2023-04-27 22:28:35,141][26631] Decorrelating experience for 0 frames... -[2023-04-27 22:28:35,326][26629] Decorrelating experience for 32 frames... -[2023-04-27 22:28:35,327][26631] Decorrelating experience for 32 frames... -[2023-04-27 22:28:35,361][26630] Decorrelating experience for 32 frames... -[2023-04-27 22:28:35,399][26632] Decorrelating experience for 0 frames... -[2023-04-27 22:28:35,410][26626] Decorrelating experience for 0 frames... -[2023-04-27 22:28:35,566][26631] Decorrelating experience for 64 frames... -[2023-04-27 22:28:35,567][26657] Decorrelating experience for 32 frames... -[2023-04-27 22:28:35,596][26632] Decorrelating experience for 32 frames... -[2023-04-27 22:28:35,596][26629] Decorrelating experience for 64 frames... -[2023-04-27 22:28:35,597][26630] Decorrelating experience for 64 frames... -[2023-04-27 22:28:35,783][26631] Decorrelating experience for 96 frames... -[2023-04-27 22:28:35,784][26657] Decorrelating experience for 64 frames... -[2023-04-27 22:28:35,825][26626] Decorrelating experience for 32 frames... -[2023-04-27 22:28:35,826][26629] Decorrelating experience for 96 frames... -[2023-04-27 22:28:35,860][26630] Decorrelating experience for 96 frames... -[2023-04-27 22:28:36,028][26626] Decorrelating experience for 64 frames... -[2023-04-27 22:28:36,029][26638] Decorrelating experience for 0 frames... -[2023-04-27 22:28:36,236][26638] Decorrelating experience for 32 frames... -[2023-04-27 22:28:36,278][26626] Decorrelating experience for 96 frames... -[2023-04-27 22:28:36,483][26638] Decorrelating experience for 64 frames... -[2023-04-27 22:28:36,526][26657] Decorrelating experience for 96 frames... -[2023-04-27 22:28:36,742][26638] Decorrelating experience for 96 frames... -[2023-04-27 22:28:36,763][26632] Decorrelating experience for 64 frames... -[2023-04-27 22:28:37,009][26632] Decorrelating experience for 96 frames... -[2023-04-27 22:28:38,609][26612] Signal inference workers to stop experience collection... -[2023-04-27 22:28:38,611][26627] InferenceWorker_p0-w0: stopping experience collection -[2023-04-27 22:28:38,815][19320] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 130.4. Samples: 652. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) -[2023-04-27 22:28:38,816][19320] Avg episode reward: [(0, '2.566')] -[2023-04-27 22:28:40,413][26612] Signal inference workers to resume experience collection... -[2023-04-27 22:28:40,414][26627] InferenceWorker_p0-w0: resuming experience collection -[2023-04-27 22:28:43,469][26627] Updated weights for policy 0, policy_version 10 (0.0582) -[2023-04-27 22:28:43,815][19320] Fps is (10 sec: 4505.6, 60 sec: 4505.6, 300 sec: 4505.6). Total num frames: 45056. Throughput: 0: 307.2. Samples: 3072. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) -[2023-04-27 22:28:43,816][19320] Avg episode reward: [(0, '4.236')] -[2023-04-27 22:28:46,007][26627] Updated weights for policy 0, policy_version 20 (0.0008) -[2023-04-27 22:28:47,944][26627] Updated weights for policy 0, policy_version 30 (0.0012) -[2023-04-27 22:28:48,815][19320] Fps is (10 sec: 13926.4, 60 sec: 9284.3, 300 sec: 9284.3). Total num frames: 139264. Throughput: 0: 1720.3. Samples: 25804. Policy #0 lag: (min: 0.0, avg: 0.8, max: 1.0) -[2023-04-27 22:28:48,815][19320] Avg episode reward: [(0, '4.246')] -[2023-04-27 22:28:48,821][26612] Saving new best policy, reward=4.246! -[2023-04-27 22:28:50,306][26627] Updated weights for policy 0, policy_version 40 (0.0008) -[2023-04-27 22:28:50,340][19320] Heartbeat connected on Batcher_0 -[2023-04-27 22:28:50,342][19320] Heartbeat connected on LearnerWorker_p0 -[2023-04-27 22:28:50,349][19320] Heartbeat connected on RolloutWorker_w0 -[2023-04-27 22:28:50,350][19320] Heartbeat connected on InferenceWorker_p0-w0 -[2023-04-27 22:28:50,353][19320] Heartbeat connected on RolloutWorker_w2 -[2023-04-27 22:28:50,356][19320] Heartbeat connected on RolloutWorker_w3 -[2023-04-27 22:28:50,358][19320] Heartbeat connected on RolloutWorker_w4 -[2023-04-27 22:28:50,361][19320] Heartbeat connected on RolloutWorker_w5 -[2023-04-27 22:28:50,365][19320] Heartbeat connected on RolloutWorker_w6 -[2023-04-27 22:28:50,366][19320] Heartbeat connected on RolloutWorker_w7 -[2023-04-27 22:28:52,258][26627] Updated weights for policy 0, policy_version 50 (0.0007) -[2023-04-27 22:28:53,815][19320] Fps is (10 sec: 18432.0, 60 sec: 11468.8, 300 sec: 11468.8). Total num frames: 229376. Throughput: 0: 2718.3. Samples: 54366. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) -[2023-04-27 22:28:53,816][19320] Avg episode reward: [(0, '4.546')] -[2023-04-27 22:28:53,817][26612] Saving new best policy, reward=4.546! -[2023-04-27 22:28:54,878][26627] Updated weights for policy 0, policy_version 60 (0.0013) -[2023-04-27 22:28:57,353][26627] Updated weights for policy 0, policy_version 70 (0.0011) -[2023-04-27 22:28:58,815][19320] Fps is (10 sec: 17612.6, 60 sec: 12615.6, 300 sec: 12615.6). Total num frames: 315392. Throughput: 0: 2677.7. Samples: 66944. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) -[2023-04-27 22:28:58,816][19320] Avg episode reward: [(0, '4.345')] -[2023-04-27 22:28:59,354][26627] Updated weights for policy 0, policy_version 80 (0.0009) -[2023-04-27 22:29:02,010][26627] Updated weights for policy 0, policy_version 90 (0.0014) -[2023-04-27 22:29:03,815][19320] Fps is (10 sec: 15974.3, 60 sec: 12970.7, 300 sec: 12970.7). Total num frames: 389120. Throughput: 0: 3074.0. Samples: 92220. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) -[2023-04-27 22:29:03,815][19320] Avg episode reward: [(0, '4.536')] -[2023-04-27 22:29:04,867][26627] Updated weights for policy 0, policy_version 100 (0.0020) -[2023-04-27 22:29:07,514][26627] Updated weights for policy 0, policy_version 110 (0.0011) -[2023-04-27 22:29:08,815][19320] Fps is (10 sec: 15974.6, 60 sec: 13575.3, 300 sec: 13575.3). Total num frames: 475136. Throughput: 0: 3287.9. Samples: 115078. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) -[2023-04-27 22:29:08,816][19320] Avg episode reward: [(0, '4.565')] -[2023-04-27 22:29:08,829][26612] Saving new best policy, reward=4.565! -[2023-04-27 22:29:09,550][26627] Updated weights for policy 0, policy_version 120 (0.0007) -[2023-04-27 22:29:11,733][26627] Updated weights for policy 0, policy_version 130 (0.0009) -[2023-04-27 22:29:13,474][26627] Updated weights for policy 0, policy_version 140 (0.0009) -[2023-04-27 22:29:13,815][19320] Fps is (10 sec: 19251.2, 60 sec: 14540.8, 300 sec: 14540.8). Total num frames: 581632. Throughput: 0: 3246.8. Samples: 129872. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) -[2023-04-27 22:29:13,816][19320] Avg episode reward: [(0, '4.521')] -[2023-04-27 22:29:15,075][26627] Updated weights for policy 0, policy_version 150 (0.0006) -[2023-04-27 22:29:16,757][26627] Updated weights for policy 0, policy_version 160 (0.0008) -[2023-04-27 22:29:18,355][26627] Updated weights for policy 0, policy_version 170 (0.0008) -[2023-04-27 22:29:18,815][19320] Fps is (10 sec: 22937.4, 60 sec: 15655.8, 300 sec: 15655.8). Total num frames: 704512. Throughput: 0: 3678.4. Samples: 165526. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) -[2023-04-27 22:29:18,816][19320] Avg episode reward: [(0, '4.458')] -[2023-04-27 22:29:20,047][26627] Updated weights for policy 0, policy_version 180 (0.0008) -[2023-04-27 22:29:21,736][26627] Updated weights for policy 0, policy_version 190 (0.0009) -[2023-04-27 22:29:23,376][26627] Updated weights for policy 0, policy_version 200 (0.0009) -[2023-04-27 22:29:23,815][19320] Fps is (10 sec: 24576.1, 60 sec: 16547.9, 300 sec: 16547.9). Total num frames: 827392. Throughput: 0: 4482.5. Samples: 202364. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) -[2023-04-27 22:29:23,815][19320] Avg episode reward: [(0, '4.534')] -[2023-04-27 22:29:25,033][26627] Updated weights for policy 0, policy_version 210 (0.0008) -[2023-04-27 22:29:26,702][26627] Updated weights for policy 0, policy_version 220 (0.0007) -[2023-04-27 22:29:28,383][26627] Updated weights for policy 0, policy_version 230 (0.0007) -[2023-04-27 22:29:28,815][19320] Fps is (10 sec: 24576.2, 60 sec: 17277.7, 300 sec: 17277.7). Total num frames: 950272. Throughput: 0: 4842.8. Samples: 221000. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) -[2023-04-27 22:29:28,816][19320] Avg episode reward: [(0, '4.485')] -[2023-04-27 22:29:30,029][26627] Updated weights for policy 0, policy_version 240 (0.0008) -[2023-04-27 22:29:31,795][26627] Updated weights for policy 0, policy_version 250 (0.0008) -[2023-04-27 22:29:33,815][19320] Fps is (10 sec: 23346.9, 60 sec: 17681.0, 300 sec: 17681.0). Total num frames: 1060864. Throughput: 0: 5146.0. Samples: 257374. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) -[2023-04-27 22:29:33,816][19320] Avg episode reward: [(0, '4.669')] -[2023-04-27 22:29:33,818][26612] Saving new best policy, reward=4.669! -[2023-04-27 22:29:34,002][26627] Updated weights for policy 0, policy_version 260 (0.0008) -[2023-04-27 22:29:36,837][26627] Updated weights for policy 0, policy_version 270 (0.0010) -[2023-04-27 22:29:38,730][26627] Updated weights for policy 0, policy_version 280 (0.0007) -[2023-04-27 22:29:38,815][19320] Fps is (10 sec: 19660.9, 60 sec: 19114.7, 300 sec: 17644.3). Total num frames: 1146880. Throughput: 0: 5076.0. Samples: 282788. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) -[2023-04-27 22:29:38,816][19320] Avg episode reward: [(0, '4.759')] -[2023-04-27 22:29:38,821][26612] Saving new best policy, reward=4.759! -[2023-04-27 22:29:41,483][26627] Updated weights for policy 0, policy_version 290 (0.0013) -[2023-04-27 22:29:43,815][19320] Fps is (10 sec: 16384.2, 60 sec: 19660.8, 300 sec: 17495.8). Total num frames: 1224704. Throughput: 0: 5061.7. Samples: 294720. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) -[2023-04-27 22:29:43,817][19320] Avg episode reward: [(0, '4.496')] -[2023-04-27 22:29:43,996][26627] Updated weights for policy 0, policy_version 300 (0.0009) -[2023-04-27 22:29:45,921][26627] Updated weights for policy 0, policy_version 310 (0.0009) -[2023-04-27 22:29:47,745][26627] Updated weights for policy 0, policy_version 320 (0.0007) -[2023-04-27 22:29:48,815][19320] Fps is (10 sec: 17203.2, 60 sec: 19660.8, 300 sec: 17585.5). Total num frames: 1318912. Throughput: 0: 5135.5. Samples: 323316. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) -[2023-04-27 22:29:48,816][19320] Avg episode reward: [(0, '4.280')] -[2023-04-27 22:29:50,625][26627] Updated weights for policy 0, policy_version 330 (0.0007) -[2023-04-27 22:29:52,896][26627] Updated weights for policy 0, policy_version 340 (0.0009) -[2023-04-27 22:29:53,815][19320] Fps is (10 sec: 17612.0, 60 sec: 19524.1, 300 sec: 17510.3). Total num frames: 1400832. Throughput: 0: 5191.6. Samples: 348700. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) -[2023-04-27 22:29:53,816][19320] Avg episode reward: [(0, '4.253')] -[2023-04-27 22:29:56,313][26627] Updated weights for policy 0, policy_version 350 (0.0015) -[2023-04-27 22:29:58,498][26627] Updated weights for policy 0, policy_version 360 (0.0007) -[2023-04-27 22:29:58,815][19320] Fps is (10 sec: 15974.4, 60 sec: 19387.8, 300 sec: 17396.0). Total num frames: 1478656. Throughput: 0: 5059.1. Samples: 357532. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) -[2023-04-27 22:29:58,816][19320] Avg episode reward: [(0, '4.385')] -[2023-04-27 22:30:01,271][26627] Updated weights for policy 0, policy_version 370 (0.0012) -[2023-04-27 22:30:03,815][19320] Fps is (10 sec: 14746.0, 60 sec: 19319.4, 300 sec: 17203.2). Total num frames: 1548288. Throughput: 0: 4768.2. Samples: 380094. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) -[2023-04-27 22:30:03,816][19320] Avg episode reward: [(0, '4.464')] -[2023-04-27 22:30:04,133][26627] Updated weights for policy 0, policy_version 380 (0.0012) -[2023-04-27 22:30:06,244][26627] Updated weights for policy 0, policy_version 390 (0.0007) -[2023-04-27 22:30:07,903][26627] Updated weights for policy 0, policy_version 400 (0.0008) -[2023-04-27 22:30:08,815][19320] Fps is (10 sec: 18022.4, 60 sec: 19729.1, 300 sec: 17461.9). Total num frames: 1658880. Throughput: 0: 4616.3. Samples: 410098. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) -[2023-04-27 22:30:08,816][19320] Avg episode reward: [(0, '4.384')] -[2023-04-27 22:30:09,574][26627] Updated weights for policy 0, policy_version 410 (0.0009) -[2023-04-27 22:30:11,225][26627] Updated weights for policy 0, policy_version 420 (0.0008) -[2023-04-27 22:30:12,922][26627] Updated weights for policy 0, policy_version 430 (0.0010) -[2023-04-27 22:30:13,815][19320] Fps is (10 sec: 23347.7, 60 sec: 20002.1, 300 sec: 17817.6). Total num frames: 1781760. Throughput: 0: 4617.6. Samples: 428792. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) -[2023-04-27 22:30:13,816][19320] Avg episode reward: [(0, '4.328')] -[2023-04-27 22:30:14,554][26627] Updated weights for policy 0, policy_version 440 (0.0007) -[2023-04-27 22:30:16,206][26627] Updated weights for policy 0, policy_version 450 (0.0007) -[2023-04-27 22:30:17,871][26627] Updated weights for policy 0, policy_version 460 (0.0009) -[2023-04-27 22:30:18,815][19320] Fps is (10 sec: 24985.5, 60 sec: 20070.4, 300 sec: 18178.4). Total num frames: 1908736. Throughput: 0: 4631.5. Samples: 465790. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) -[2023-04-27 22:30:18,816][19320] Avg episode reward: [(0, '4.316')] -[2023-04-27 22:30:19,460][26627] Updated weights for policy 0, policy_version 470 (0.0008) -[2023-04-27 22:30:21,046][26627] Updated weights for policy 0, policy_version 480 (0.0009) -[2023-04-27 22:30:22,710][26627] Updated weights for policy 0, policy_version 490 (0.0009) -[2023-04-27 22:30:23,815][19320] Fps is (10 sec: 24985.5, 60 sec: 20070.4, 300 sec: 18469.2). Total num frames: 2031616. Throughput: 0: 4906.8. Samples: 503596. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) -[2023-04-27 22:30:23,816][19320] Avg episode reward: [(0, '4.388')] -[2023-04-27 22:30:24,368][26627] Updated weights for policy 0, policy_version 500 (0.0007) -[2023-04-27 22:30:25,982][26627] Updated weights for policy 0, policy_version 510 (0.0009) -[2023-04-27 22:30:28,815][19320] Fps is (10 sec: 20889.7, 60 sec: 19456.0, 300 sec: 18414.2). Total num frames: 2117632. Throughput: 0: 5056.4. Samples: 522256. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) -[2023-04-27 22:30:28,816][19320] Avg episode reward: [(0, '4.397')] -[2023-04-27 22:30:28,820][26612] Saving /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000517_2117632.pth... -[2023-04-27 22:30:29,361][26627] Updated weights for policy 0, policy_version 520 (0.0015) -[2023-04-27 22:30:31,575][26627] Updated weights for policy 0, policy_version 530 (0.0009) -[2023-04-27 22:30:33,673][26627] Updated weights for policy 0, policy_version 540 (0.0007) -[2023-04-27 22:30:33,815][19320] Fps is (10 sec: 18022.5, 60 sec: 19183.0, 300 sec: 18432.0). Total num frames: 2211840. Throughput: 0: 4930.8. Samples: 545200. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) -[2023-04-27 22:30:33,816][19320] Avg episode reward: [(0, '4.420')] -[2023-04-27 22:30:35,975][26627] Updated weights for policy 0, policy_version 550 (0.0008) -[2023-04-27 22:30:38,045][26627] Updated weights for policy 0, policy_version 560 (0.0010) -[2023-04-27 22:30:38,815][19320] Fps is (10 sec: 18841.5, 60 sec: 19319.5, 300 sec: 18448.4). Total num frames: 2306048. Throughput: 0: 4988.7. Samples: 573190. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) -[2023-04-27 22:30:38,816][19320] Avg episode reward: [(0, '4.549')] -[2023-04-27 22:30:40,672][26627] Updated weights for policy 0, policy_version 570 (0.0011) -[2023-04-27 22:30:43,150][26627] Updated weights for policy 0, policy_version 580 (0.0012) -[2023-04-27 22:30:43,815][19320] Fps is (10 sec: 17612.8, 60 sec: 19387.7, 300 sec: 18369.0). Total num frames: 2387968. Throughput: 0: 5056.8. Samples: 585088. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) -[2023-04-27 22:30:43,816][19320] Avg episode reward: [(0, '4.520')] -[2023-04-27 22:30:45,278][26627] Updated weights for policy 0, policy_version 590 (0.0007) -[2023-04-27 22:30:47,790][26627] Updated weights for policy 0, policy_version 600 (0.0007) -[2023-04-27 22:30:48,815][19320] Fps is (10 sec: 17203.2, 60 sec: 19319.5, 300 sec: 18356.1). Total num frames: 2478080. Throughput: 0: 5159.8. Samples: 612286. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) -[2023-04-27 22:30:48,816][19320] Avg episode reward: [(0, '4.326')] -[2023-04-27 22:30:50,346][26627] Updated weights for policy 0, policy_version 610 (0.0008) -[2023-04-27 22:30:53,815][19320] Fps is (10 sec: 14745.6, 60 sec: 18910.0, 300 sec: 18110.2). Total num frames: 2535424. Throughput: 0: 4977.5. Samples: 634084. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) -[2023-04-27 22:30:53,816][19320] Avg episode reward: [(0, '4.424')] -[2023-04-27 22:30:53,893][26627] Updated weights for policy 0, policy_version 620 (0.0019) -[2023-04-27 22:30:56,047][26627] Updated weights for policy 0, policy_version 630 (0.0007) -[2023-04-27 22:30:58,480][26627] Updated weights for policy 0, policy_version 640 (0.0011) -[2023-04-27 22:30:58,815][19320] Fps is (10 sec: 14745.6, 60 sec: 19114.7, 300 sec: 18107.1). Total num frames: 2625536. Throughput: 0: 4805.1. Samples: 645020. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) -[2023-04-27 22:30:58,815][19320] Avg episode reward: [(0, '4.450')] -[2023-04-27 22:31:00,719][26627] Updated weights for policy 0, policy_version 650 (0.0008) -[2023-04-27 22:31:02,385][26627] Updated weights for policy 0, policy_version 660 (0.0008) -[2023-04-27 22:31:03,815][19320] Fps is (10 sec: 20070.4, 60 sec: 19797.4, 300 sec: 18240.9). Total num frames: 2736128. Throughput: 0: 4617.8. Samples: 673592. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) -[2023-04-27 22:31:03,816][19320] Avg episode reward: [(0, '4.602')] -[2023-04-27 22:31:04,041][26627] Updated weights for policy 0, policy_version 670 (0.0010) -[2023-04-27 22:31:05,642][26627] Updated weights for policy 0, policy_version 680 (0.0009) -[2023-04-27 22:31:07,304][26627] Updated weights for policy 0, policy_version 690 (0.0007) -[2023-04-27 22:31:08,815][19320] Fps is (10 sec: 23756.8, 60 sec: 20070.4, 300 sec: 18471.6). Total num frames: 2863104. Throughput: 0: 4611.2. Samples: 711102. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) -[2023-04-27 22:31:08,816][19320] Avg episode reward: [(0, '4.465')] -[2023-04-27 22:31:08,913][26627] Updated weights for policy 0, policy_version 700 (0.0008) -[2023-04-27 22:31:10,614][26627] Updated weights for policy 0, policy_version 710 (0.0006) -[2023-04-27 22:31:12,270][26627] Updated weights for policy 0, policy_version 720 (0.0007) -[2023-04-27 22:31:13,815][19320] Fps is (10 sec: 24985.6, 60 sec: 20070.4, 300 sec: 18662.4). Total num frames: 2985984. Throughput: 0: 4608.2. Samples: 729626. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) -[2023-04-27 22:31:13,816][19320] Avg episode reward: [(0, '4.396')] -[2023-04-27 22:31:13,869][26627] Updated weights for policy 0, policy_version 730 (0.0007) -[2023-04-27 22:31:15,506][26627] Updated weights for policy 0, policy_version 740 (0.0009) -[2023-04-27 22:31:17,160][26627] Updated weights for policy 0, policy_version 750 (0.0008) -[2023-04-27 22:31:18,815][19320] Fps is (10 sec: 24576.0, 60 sec: 20002.1, 300 sec: 18841.6). Total num frames: 3108864. Throughput: 0: 4932.6. Samples: 767168. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) -[2023-04-27 22:31:18,816][19320] Avg episode reward: [(0, '4.364')] -[2023-04-27 22:31:18,848][26627] Updated weights for policy 0, policy_version 760 (0.0009) -[2023-04-27 22:31:21,623][26627] Updated weights for policy 0, policy_version 770 (0.0013) -[2023-04-27 22:31:23,815][19320] Fps is (10 sec: 20070.4, 60 sec: 19251.2, 300 sec: 18745.2). Total num frames: 3186688. Throughput: 0: 4907.6. Samples: 794032. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) -[2023-04-27 22:31:23,816][19320] Avg episode reward: [(0, '4.513')] -[2023-04-27 22:31:24,158][26627] Updated weights for policy 0, policy_version 780 (0.0013) -[2023-04-27 22:31:26,913][26627] Updated weights for policy 0, policy_version 790 (0.0012) -[2023-04-27 22:31:28,815][19320] Fps is (10 sec: 14336.0, 60 sec: 18909.9, 300 sec: 18584.1). Total num frames: 3252224. Throughput: 0: 4932.8. Samples: 807066. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) -[2023-04-27 22:31:28,815][19320] Avg episode reward: [(0, '4.367')] -[2023-04-27 22:31:30,689][26627] Updated weights for policy 0, policy_version 800 (0.0012) -[2023-04-27 22:31:33,035][26627] Updated weights for policy 0, policy_version 810 (0.0011) -[2023-04-27 22:31:33,815][19320] Fps is (10 sec: 14745.6, 60 sec: 18705.1, 300 sec: 18523.0). Total num frames: 3334144. Throughput: 0: 4712.5. Samples: 824348. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) -[2023-04-27 22:31:33,816][19320] Avg episode reward: [(0, '4.320')] -[2023-04-27 22:31:35,061][26627] Updated weights for policy 0, policy_version 820 (0.0007) -[2023-04-27 22:31:37,474][26627] Updated weights for policy 0, policy_version 830 (0.0010) -[2023-04-27 22:31:38,815][19320] Fps is (10 sec: 16793.5, 60 sec: 18568.5, 300 sec: 18487.4). Total num frames: 3420160. Throughput: 0: 4863.2. Samples: 852926. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) -[2023-04-27 22:31:38,816][19320] Avg episode reward: [(0, '4.490')] -[2023-04-27 22:31:40,044][26627] Updated weights for policy 0, policy_version 840 (0.0007) -[2023-04-27 22:31:42,664][26627] Updated weights for policy 0, policy_version 850 (0.0014) -[2023-04-27 22:31:43,815][19320] Fps is (10 sec: 15974.4, 60 sec: 18432.0, 300 sec: 18388.9). Total num frames: 3493888. Throughput: 0: 4869.0. Samples: 864126. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) -[2023-04-27 22:31:43,816][19320] Avg episode reward: [(0, '4.454')] -[2023-04-27 22:31:45,339][26627] Updated weights for policy 0, policy_version 860 (0.0014) -[2023-04-27 22:31:47,414][26627] Updated weights for policy 0, policy_version 870 (0.0015) -[2023-04-27 22:31:48,815][19320] Fps is (10 sec: 16793.2, 60 sec: 18500.2, 300 sec: 18400.5). Total num frames: 3588096. Throughput: 0: 4780.7. Samples: 888726. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) -[2023-04-27 22:31:48,816][19320] Avg episode reward: [(0, '4.718')] -[2023-04-27 22:31:49,524][26627] Updated weights for policy 0, policy_version 880 (0.0008) -[2023-04-27 22:31:51,616][26627] Updated weights for policy 0, policy_version 890 (0.0012) -[2023-04-27 22:31:53,815][19320] Fps is (10 sec: 18841.4, 60 sec: 19114.6, 300 sec: 18411.5). Total num frames: 3682304. Throughput: 0: 4563.8. Samples: 916472. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) -[2023-04-27 22:31:53,816][19320] Avg episode reward: [(0, '4.287')] -[2023-04-27 22:31:53,847][26627] Updated weights for policy 0, policy_version 900 (0.0011) -[2023-04-27 22:31:55,513][26627] Updated weights for policy 0, policy_version 910 (0.0009) -[2023-04-27 22:31:57,111][26627] Updated weights for policy 0, policy_version 920 (0.0007) -[2023-04-27 22:31:58,754][26627] Updated weights for policy 0, policy_version 930 (0.0007) -[2023-04-27 22:31:58,815][19320] Fps is (10 sec: 22118.6, 60 sec: 19729.0, 300 sec: 18581.8). Total num frames: 3809280. Throughput: 0: 4564.9. Samples: 935046. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) -[2023-04-27 22:31:58,816][19320] Avg episode reward: [(0, '4.318')] -[2023-04-27 22:32:00,439][26627] Updated weights for policy 0, policy_version 940 (0.0007) -[2023-04-27 22:32:02,146][26627] Updated weights for policy 0, policy_version 950 (0.0007) -[2023-04-27 22:32:03,815][19320] Fps is (10 sec: 24576.2, 60 sec: 19865.6, 300 sec: 18705.1). Total num frames: 3928064. Throughput: 0: 4548.2. Samples: 971838. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) -[2023-04-27 22:32:03,816][19320] Avg episode reward: [(0, '4.347')] -[2023-04-27 22:32:03,837][26627] Updated weights for policy 0, policy_version 960 (0.0008) -[2023-04-27 22:32:05,497][26627] Updated weights for policy 0, policy_version 970 (0.0008) -[2023-04-27 22:32:06,763][26612] Stopping Batcher_0... -[2023-04-27 22:32:06,763][26612] Saving /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... -[2023-04-27 22:32:06,763][26612] Loop batcher_evt_loop terminating... -[2023-04-27 22:32:06,763][19320] Component Batcher_0 stopped! -[2023-04-27 22:32:06,764][19320] Component RolloutWorker_w1 process died already! Don't wait for it. -[2023-04-27 22:32:06,768][26638] Stopping RolloutWorker_w7... -[2023-04-27 22:32:06,769][26638] Loop rollout_proc7_evt_loop terminating... -[2023-04-27 22:32:06,769][26630] Stopping RolloutWorker_w3... -[2023-04-27 22:32:06,769][26631] Stopping RolloutWorker_w4... -[2023-04-27 22:32:06,769][26630] Loop rollout_proc3_evt_loop terminating... -[2023-04-27 22:32:06,769][26631] Loop rollout_proc4_evt_loop terminating... -[2023-04-27 22:32:06,769][26629] Stopping RolloutWorker_w2... -[2023-04-27 22:32:06,768][19320] Component RolloutWorker_w7 stopped! -[2023-04-27 22:32:06,769][26629] Loop rollout_proc2_evt_loop terminating... -[2023-04-27 22:32:06,770][19320] Component RolloutWorker_w3 stopped! -[2023-04-27 22:32:06,770][26632] Stopping RolloutWorker_w5... -[2023-04-27 22:32:06,771][26632] Loop rollout_proc5_evt_loop terminating... -[2023-04-27 22:32:06,771][19320] Component RolloutWorker_w4 stopped! -[2023-04-27 22:32:06,772][26657] Stopping RolloutWorker_w6... -[2023-04-27 22:32:06,773][26657] Loop rollout_proc6_evt_loop terminating... -[2023-04-27 22:32:06,773][19320] Component RolloutWorker_w2 stopped! -[2023-04-27 22:32:06,774][19320] Component RolloutWorker_w5 stopped! -[2023-04-27 22:32:06,775][19320] Component RolloutWorker_w6 stopped! -[2023-04-27 22:32:06,775][26627] Weights refcount: 2 0 -[2023-04-27 22:32:06,777][26627] Stopping InferenceWorker_p0-w0... -[2023-04-27 22:32:06,777][26627] Loop inference_proc0-0_evt_loop terminating... -[2023-04-27 22:32:06,777][19320] Component InferenceWorker_p0-w0 stopped! -[2023-04-27 22:32:06,786][26626] Stopping RolloutWorker_w0... -[2023-04-27 22:32:06,787][26626] Loop rollout_proc0_evt_loop terminating... -[2023-04-27 22:32:06,786][19320] Component RolloutWorker_w0 stopped! -[2023-04-27 22:32:06,811][26612] Saving /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... -[2023-04-27 22:32:06,858][26612] Stopping LearnerWorker_p0... -[2023-04-27 22:32:06,859][26612] Loop learner_proc0_evt_loop terminating... -[2023-04-27 22:32:06,859][19320] Component LearnerWorker_p0 stopped! -[2023-04-27 22:32:06,860][19320] Waiting for process learner_proc0 to stop... -[2023-04-27 22:32:07,560][19320] Waiting for process inference_proc0-0 to join... -[2023-04-27 22:32:07,561][19320] Waiting for process rollout_proc0 to join... -[2023-04-27 22:32:07,562][19320] Waiting for process rollout_proc1 to join... -[2023-04-27 22:32:07,563][19320] Waiting for process rollout_proc2 to join... -[2023-04-27 22:32:07,564][19320] Waiting for process rollout_proc3 to join... -[2023-04-27 22:32:07,564][19320] Waiting for process rollout_proc4 to join... -[2023-04-27 22:32:07,565][19320] Waiting for process rollout_proc5 to join... -[2023-04-27 22:32:07,566][19320] Waiting for process rollout_proc6 to join... -[2023-04-27 22:32:07,567][19320] Waiting for process rollout_proc7 to join... -[2023-04-27 22:32:07,567][19320] Batcher 0 profile tree view: -batching: 12.7812, releasing_batches: 0.0195 -[2023-04-27 22:32:07,568][19320] InferenceWorker_p0-w0 profile tree view: -wait_policy: 0.0000 - wait_policy_total: 2.8370 -update_model: 2.8123 - weight_update: 0.0007 -one_step: 0.0023 - handle_policy_step: 195.8112 - deserialize: 5.6395, stack: 0.7269, obs_to_device_normalize: 44.5135, forward: 68.6296, send_messages: 17.9614 - prepare_outputs: 52.0115 - to_cpu: 45.1667 -[2023-04-27 22:32:07,569][19320] Learner 0 profile tree view: -misc: 0.0037, prepare_batch: 9.4485 -train: 22.5017 - epoch_init: 0.0033, minibatch_init: 0.0045, losses_postprocess: 0.4800, kl_divergence: 0.4893, after_optimizer: 6.8805 - calculate_losses: 7.1964 - losses_init: 0.0020, forward_head: 0.5407, bptt_initial: 3.3368, tail: 0.4113, advantages_returns: 0.1193, losses: 1.7274 - bptt: 0.9502 - bptt_forward_core: 0.9150 - update: 7.1922 - clip: 0.8865 -[2023-04-27 22:32:07,569][19320] RolloutWorker_w0 profile tree view: -wait_for_trajectories: 0.0991, enqueue_policy_requests: 5.1038, env_step: 94.9605, overhead: 6.8366, complete_rollouts: 0.2024 -save_policy_outputs: 6.2491 - split_output_tensors: 2.9626 -[2023-04-27 22:32:07,570][19320] RolloutWorker_w7 profile tree view: -wait_for_trajectories: 0.0969, enqueue_policy_requests: 5.1019, env_step: 95.1988, overhead: 6.8454, complete_rollouts: 0.2055 -save_policy_outputs: 6.4640 - split_output_tensors: 3.0653 -[2023-04-27 22:32:07,571][19320] Loop Runner_EvtLoop terminating... -[2023-04-27 22:32:07,572][19320] Runner profile tree view: -main_loop: 217.2086 -[2023-04-27 22:32:07,573][19320] Collected {0: 4005888}, FPS: 18442.6 -[2023-04-27 22:33:30,876][19320] Loading existing experiment configuration from /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/config.json -[2023-04-27 22:33:30,877][19320] Overriding arg 'num_workers' with value 1 passed from command line -[2023-04-27 22:33:30,878][19320] Adding new argument 'no_render'=True that is not in the saved config file! -[2023-04-27 22:33:30,878][19320] Adding new argument 'save_video'=True that is not in the saved config file! -[2023-04-27 22:33:30,879][19320] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! -[2023-04-27 22:33:30,879][19320] Adding new argument 'video_name'=None that is not in the saved config file! -[2023-04-27 22:33:30,880][19320] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file! -[2023-04-27 22:33:30,881][19320] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! -[2023-04-27 22:33:30,882][19320] Adding new argument 'push_to_hub'=False that is not in the saved config file! -[2023-04-27 22:33:30,882][19320] Adding new argument 'hf_repository'=None that is not in the saved config file! -[2023-04-27 22:33:30,883][19320] Adding new argument 'policy_index'=0 that is not in the saved config file! -[2023-04-27 22:33:30,883][19320] Adding new argument 'eval_deterministic'=False that is not in the saved config file! -[2023-04-27 22:33:30,884][19320] Adding new argument 'train_script'=None that is not in the saved config file! -[2023-04-27 22:33:30,885][19320] Adding new argument 'enjoy_script'=None that is not in the saved config file! -[2023-04-27 22:33:30,885][19320] Using frameskip 1 and render_action_repeat=4 for evaluation -[2023-04-27 22:33:30,891][19320] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-27 22:33:30,892][19320] RunningMeanStd input shape: (3, 72, 128) -[2023-04-27 22:33:30,893][19320] RunningMeanStd input shape: (1,) -[2023-04-27 22:33:30,904][19320] ConvEncoder: input_channels=3 -[2023-04-27 22:33:30,994][19320] Conv encoder output size: 512 -[2023-04-27 22:33:30,995][19320] Policy head output size: 512 -[2023-04-27 22:33:32,427][19320] Loading state from checkpoint /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... -[2023-04-27 22:33:33,708][19320] Num frames 100... -[2023-04-27 22:33:33,796][19320] Num frames 200... -[2023-04-27 22:33:33,876][19320] Num frames 300... -[2023-04-27 22:33:33,996][19320] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840 -[2023-04-27 22:33:33,997][19320] Avg episode reward: 3.840, avg true_objective: 3.840 -[2023-04-27 22:33:34,014][19320] Num frames 400... -[2023-04-27 22:33:34,096][19320] Num frames 500... -[2023-04-27 22:33:34,173][19320] Num frames 600... -[2023-04-27 22:33:34,251][19320] Num frames 700... -[2023-04-27 22:33:34,358][19320] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840 -[2023-04-27 22:33:34,359][19320] Avg episode reward: 3.840, avg true_objective: 3.840 -[2023-04-27 22:33:34,389][19320] Num frames 800... -[2023-04-27 22:33:34,468][19320] Num frames 900... -[2023-04-27 22:33:34,542][19320] Num frames 1000... -[2023-04-27 22:33:34,617][19320] Num frames 1100... -[2023-04-27 22:33:34,736][19320] Avg episode rewards: #0: 4.280, true rewards: #0: 3.947 -[2023-04-27 22:33:34,737][19320] Avg episode reward: 4.280, avg true_objective: 3.947 -[2023-04-27 22:33:34,752][19320] Num frames 1200... -[2023-04-27 22:33:34,835][19320] Num frames 1300... -[2023-04-27 22:33:34,915][19320] Num frames 1400... -[2023-04-27 22:33:34,997][19320] Num frames 1500... -[2023-04-27 22:33:35,092][19320] Num frames 1600... -[2023-04-27 22:33:35,144][19320] Avg episode rewards: #0: 4.500, true rewards: #0: 4.000 -[2023-04-27 22:33:35,145][19320] Avg episode reward: 4.500, avg true_objective: 4.000 -[2023-04-27 22:33:35,229][19320] Num frames 1700... -[2023-04-27 22:33:35,311][19320] Num frames 1800... -[2023-04-27 22:33:35,391][19320] Num frames 1900... -[2023-04-27 22:33:35,474][19320] Num frames 2000... -[2023-04-27 22:33:35,569][19320] Avg episode rewards: #0: 4.696, true rewards: #0: 4.096 -[2023-04-27 22:33:35,570][19320] Avg episode reward: 4.696, avg true_objective: 4.096 -[2023-04-27 22:33:35,617][19320] Num frames 2100... -[2023-04-27 22:33:35,698][19320] Num frames 2200... -[2023-04-27 22:33:35,772][19320] Num frames 2300... -[2023-04-27 22:33:35,849][19320] Num frames 2400... -[2023-04-27 22:33:35,932][19320] Avg episode rewards: #0: 4.553, true rewards: #0: 4.053 -[2023-04-27 22:33:35,933][19320] Avg episode reward: 4.553, avg true_objective: 4.053 -[2023-04-27 22:33:35,989][19320] Num frames 2500... -[2023-04-27 22:33:36,070][19320] Num frames 2600... -[2023-04-27 22:33:36,157][19320] Num frames 2700... -[2023-04-27 22:33:36,236][19320] Num frames 2800... -[2023-04-27 22:33:36,319][19320] Num frames 2900... -[2023-04-27 22:33:36,396][19320] Num frames 3000... -[2023-04-27 22:33:36,479][19320] Avg episode rewards: #0: 5.200, true rewards: #0: 4.343 -[2023-04-27 22:33:36,480][19320] Avg episode reward: 5.200, avg true_objective: 4.343 -[2023-04-27 22:33:36,531][19320] Num frames 3100... -[2023-04-27 22:33:36,612][19320] Num frames 3200... -[2023-04-27 22:33:36,740][19320] Avg episode rewards: #0: 4.870, true rewards: #0: 4.120 -[2023-04-27 22:33:36,741][19320] Avg episode reward: 4.870, avg true_objective: 4.120 -[2023-04-27 22:33:36,747][19320] Num frames 3300... -[2023-04-27 22:33:36,839][19320] Num frames 3400... -[2023-04-27 22:33:36,925][19320] Num frames 3500... -[2023-04-27 22:33:36,998][19320] Num frames 3600... -[2023-04-27 22:33:37,114][19320] Avg episode rewards: #0: 4.756, true rewards: #0: 4.089 -[2023-04-27 22:33:37,115][19320] Avg episode reward: 4.756, avg true_objective: 4.089 -[2023-04-27 22:33:37,136][19320] Num frames 3700... -[2023-04-27 22:33:37,237][19320] Num frames 3800... -[2023-04-27 22:33:37,323][19320] Num frames 3900... -[2023-04-27 22:33:37,407][19320] Num frames 4000... -[2023-04-27 22:33:37,512][19320] Avg episode rewards: #0: 4.664, true rewards: #0: 4.064 -[2023-04-27 22:33:37,513][19320] Avg episode reward: 4.664, avg true_objective: 4.064 -[2023-04-27 22:33:41,947][19320] Replay video saved to /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/replay.mp4! -[2023-04-27 22:36:21,719][19320] Loading existing experiment configuration from /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/config.json -[2023-04-27 22:36:21,720][19320] Overriding arg 'num_workers' with value 1 passed from command line -[2023-04-27 22:36:21,721][19320] Adding new argument 'no_render'=True that is not in the saved config file! -[2023-04-27 22:36:21,722][19320] Adding new argument 'save_video'=True that is not in the saved config file! -[2023-04-27 22:36:21,722][19320] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! -[2023-04-27 22:36:21,723][19320] Adding new argument 'video_name'=None that is not in the saved config file! -[2023-04-27 22:36:21,724][19320] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! -[2023-04-27 22:36:21,724][19320] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! -[2023-04-27 22:36:21,725][19320] Adding new argument 'push_to_hub'=True that is not in the saved config file! -[2023-04-27 22:36:21,726][19320] Adding new argument 'hf_repository'='ItchyB/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! -[2023-04-27 22:36:21,726][19320] Adding new argument 'policy_index'=0 that is not in the saved config file! -[2023-04-27 22:36:21,726][19320] Adding new argument 'eval_deterministic'=False that is not in the saved config file! -[2023-04-27 22:36:21,727][19320] Adding new argument 'train_script'=None that is not in the saved config file! -[2023-04-27 22:36:21,728][19320] Adding new argument 'enjoy_script'=None that is not in the saved config file! -[2023-04-27 22:36:21,729][19320] Using frameskip 1 and render_action_repeat=4 for evaluation -[2023-04-27 22:36:21,732][19320] RunningMeanStd input shape: (3, 72, 128) -[2023-04-27 22:36:21,733][19320] RunningMeanStd input shape: (1,) -[2023-04-27 22:36:21,740][19320] ConvEncoder: input_channels=3 -[2023-04-27 22:36:21,763][19320] Conv encoder output size: 512 -[2023-04-27 22:36:21,764][19320] Policy head output size: 512 -[2023-04-27 22:36:21,791][19320] Loading state from checkpoint /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... -[2023-04-27 22:36:22,219][19320] Num frames 100... -[2023-04-27 22:36:22,339][19320] Num frames 200... -[2023-04-27 22:36:22,445][19320] Num frames 300... -[2023-04-27 22:36:22,572][19320] Num frames 400... -[2023-04-27 22:36:22,684][19320] Num frames 500... -[2023-04-27 22:36:22,754][19320] Avg episode rewards: #0: 7.120, true rewards: #0: 5.120 -[2023-04-27 22:36:22,755][19320] Avg episode reward: 7.120, avg true_objective: 5.120 -[2023-04-27 22:36:22,855][19320] Num frames 600... -[2023-04-27 22:36:22,974][19320] Num frames 700... -[2023-04-27 22:36:23,101][19320] Num frames 800... -[2023-04-27 22:36:23,257][19320] Avg episode rewards: #0: 5.480, true rewards: #0: 4.480 -[2023-04-27 22:36:23,258][19320] Avg episode reward: 5.480, avg true_objective: 4.480 -[2023-04-27 22:36:23,263][19320] Num frames 900... -[2023-04-27 22:36:23,371][19320] Num frames 1000... -[2023-04-27 22:36:23,473][19320] Num frames 1100... -[2023-04-27 22:36:23,587][19320] Num frames 1200... -[2023-04-27 22:36:23,727][19320] Avg episode rewards: #0: 4.933, true rewards: #0: 4.267 -[2023-04-27 22:36:23,728][19320] Avg episode reward: 4.933, avg true_objective: 4.267 -[2023-04-27 22:36:23,755][19320] Num frames 1300... -[2023-04-27 22:36:23,873][19320] Num frames 1400... -[2023-04-27 22:36:23,991][19320] Num frames 1500... -[2023-04-27 22:36:24,080][19320] Avg episode rewards: #0: 4.340, true rewards: #0: 3.840 -[2023-04-27 22:36:24,081][19320] Avg episode reward: 4.340, avg true_objective: 3.840 -[2023-04-27 22:36:24,143][19320] Num frames 1600... -[2023-04-27 22:36:24,240][19320] Num frames 1700... -[2023-04-27 22:36:24,341][19320] Num frames 1800... -[2023-04-27 22:36:24,432][19320] Num frames 1900... -[2023-04-27 22:36:24,562][19320] Avg episode rewards: #0: 4.768, true rewards: #0: 3.968 -[2023-04-27 22:36:24,563][19320] Avg episode reward: 4.768, avg true_objective: 3.968 -[2023-04-27 22:36:24,577][19320] Num frames 2000... -[2023-04-27 22:36:24,676][19320] Num frames 2100... -[2023-04-27 22:36:24,778][19320] Num frames 2200... -[2023-04-27 22:36:24,881][19320] Num frames 2300... -[2023-04-27 22:36:25,004][19320] Avg episode rewards: #0: 4.613, true rewards: #0: 3.947 -[2023-04-27 22:36:25,005][19320] Avg episode reward: 4.613, avg true_objective: 3.947 -[2023-04-27 22:36:25,036][19320] Num frames 2400... -[2023-04-27 22:36:25,138][19320] Num frames 2500... -[2023-04-27 22:36:25,237][19320] Num frames 2600... -[2023-04-27 22:36:25,340][19320] Num frames 2700... -[2023-04-27 22:36:25,446][19320] Avg episode rewards: #0: 4.503, true rewards: #0: 3.931 -[2023-04-27 22:36:25,447][19320] Avg episode reward: 4.503, avg true_objective: 3.931 -[2023-04-27 22:36:25,498][19320] Num frames 2800... -[2023-04-27 22:36:25,598][19320] Num frames 2900... -[2023-04-27 22:36:25,701][19320] Num frames 3000... -[2023-04-27 22:36:25,800][19320] Num frames 3100... -[2023-04-27 22:36:25,888][19320] Avg episode rewards: #0: 4.420, true rewards: #0: 3.920 -[2023-04-27 22:36:25,889][19320] Avg episode reward: 4.420, avg true_objective: 3.920 -[2023-04-27 22:36:25,969][19320] Num frames 3200... -[2023-04-27 22:36:26,076][19320] Num frames 3300... -[2023-04-27 22:36:26,181][19320] Num frames 3400... -[2023-04-27 22:36:26,281][19320] Num frames 3500... -[2023-04-27 22:36:26,354][19320] Avg episode rewards: #0: 4.356, true rewards: #0: 3.911 -[2023-04-27 22:36:26,355][19320] Avg episode reward: 4.356, avg true_objective: 3.911 -[2023-04-27 22:36:26,446][19320] Num frames 3600... -[2023-04-27 22:36:26,560][19320] Num frames 3700... -[2023-04-27 22:36:26,670][19320] Num frames 3800... -[2023-04-27 22:36:26,784][19320] Num frames 3900... -[2023-04-27 22:36:26,842][19320] Avg episode rewards: #0: 4.304, true rewards: #0: 3.904 -[2023-04-27 22:36:26,842][19320] Avg episode reward: 4.304, avg true_objective: 3.904 -[2023-04-27 22:36:30,880][19320] Replay video saved to /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/replay.mp4! -[2023-04-27 22:36:34,016][19320] The model has been pushed to https://huggingface.co/ItchyB/rl_course_vizdoom_health_gathering_supreme -[2023-04-29 19:05:17,493][108205] Saving configuration to /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/config.json... -[2023-04-29 19:05:17,495][108205] Rollout worker 0 uses device cpu -[2023-04-29 19:05:17,495][108205] Rollout worker 1 uses device cpu -[2023-04-29 19:05:17,496][108205] Rollout worker 2 uses device cpu -[2023-04-29 19:05:17,496][108205] Rollout worker 3 uses device cpu -[2023-04-29 19:05:17,497][108205] Rollout worker 4 uses device cpu -[2023-04-29 19:05:17,498][108205] Rollout worker 5 uses device cpu -[2023-04-29 19:05:17,498][108205] Rollout worker 6 uses device cpu -[2023-04-29 19:05:17,499][108205] Rollout worker 7 uses device cpu -[2023-04-29 19:05:17,524][108205] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-29 19:05:17,525][108205] InferenceWorker_p0-w0: min num requests: 2 -[2023-04-29 19:05:17,540][108205] Starting all processes... -[2023-04-29 19:05:17,540][108205] Starting process learner_proc0 -[2023-04-29 19:05:17,634][108205] Starting all processes... -[2023-04-29 19:05:17,637][108205] Starting process inference_proc0-0 -[2023-04-29 19:05:17,638][108205] Starting process rollout_proc0 -[2023-04-29 19:05:17,638][108205] Starting process rollout_proc1 -[2023-04-29 19:05:17,638][108205] Starting process rollout_proc2 -[2023-04-29 19:05:17,639][108205] Starting process rollout_proc3 -[2023-04-29 19:05:17,639][108205] Starting process rollout_proc4 -[2023-04-29 19:05:17,639][108205] Starting process rollout_proc5 -[2023-04-29 19:05:17,640][108205] Starting process rollout_proc6 -[2023-04-29 19:05:17,640][108205] Starting process rollout_proc7 -[2023-04-29 19:05:18,549][133597] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-29 19:05:18,549][133597] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0 -[2023-04-29 19:05:18,577][133617] Worker 6 uses CPU cores [18, 19, 20] -[2023-04-29 19:05:18,590][133597] Num visible devices: 1 -[2023-04-29 19:05:18,596][133612] Worker 1 uses CPU cores [3, 4, 5] -[2023-04-29 19:05:18,600][133610] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-29 19:05:18,600][133610] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0 -[2023-04-29 19:05:18,614][133615] Worker 4 uses CPU cores [12, 13, 14] -[2023-04-29 19:05:18,615][133610] Num visible devices: 1 -[2023-04-29 19:05:18,623][133611] Worker 0 uses CPU cores [0, 1, 2] -[2023-04-29 19:05:18,626][133616] Worker 5 uses CPU cores [15, 16, 17] -[2023-04-29 19:05:18,626][133613] Worker 3 uses CPU cores [9, 10, 11] -[2023-04-29 19:05:18,637][133597] Starting seed is not provided -[2023-04-29 19:05:18,637][133597] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-29 19:05:18,637][133597] Initializing actor-critic model on device cuda:0 -[2023-04-29 19:05:18,638][133597] RunningMeanStd input shape: (3, 72, 128) -[2023-04-29 19:05:18,638][133597] RunningMeanStd input shape: (1,) -[2023-04-29 19:05:18,640][133618] Worker 7 uses CPU cores [21, 22, 23] -[2023-04-29 19:05:18,642][133614] Worker 2 uses CPU cores [6, 7, 8] -[2023-04-29 19:05:18,648][133597] ConvEncoder: input_channels=3 -[2023-04-29 19:05:18,758][133597] Conv encoder output size: 512 -[2023-04-29 19:05:18,758][133597] Policy head output size: 512 -[2023-04-29 19:05:18,790][133597] Created Actor Critic model with architecture: -[2023-04-29 19:05:18,790][133597] ActorCriticSharedWeights( +[2023-04-30 12:43:59,297][678641] Using optimizer +[2023-04-30 12:43:59,298][678641] No checkpoints found +[2023-04-30 12:43:59,299][678641] Did not load from checkpoint, starting from scratch! +[2023-04-30 12:43:59,299][678641] Initialized policy 0 weights for model version 0 +[2023-04-30 12:43:59,300][678641] LearnerWorker_p0 finished initialization! +[2023-04-30 12:43:59,906][678704] Worker 0 uses CPU cores [0, 1, 2] +[2023-04-30 12:43:59,919][678703] RunningMeanStd input shape: (3, 72, 128) +[2023-04-30 12:43:59,920][678703] RunningMeanStd input shape: (1,) +[2023-04-30 12:43:59,928][678703] ConvEncoder: input_channels=3 +[2023-04-30 12:43:59,932][678706] Worker 2 uses CPU cores [6, 7, 8] +[2023-04-30 12:43:59,934][678711] Worker 7 uses CPU cores [21, 22, 23] +[2023-04-30 12:43:59,935][678705] Worker 1 uses CPU cores [3, 4, 5] +[2023-04-30 12:43:59,943][678708] Worker 4 uses CPU cores [12, 13, 14] +[2023-04-30 12:43:59,948][678710] Worker 6 uses CPU cores [18, 19, 20] +[2023-04-30 12:43:59,949][678707] Worker 3 uses CPU cores [9, 10, 11] +[2023-04-30 12:43:59,953][678709] Worker 5 uses CPU cores [15, 16, 17] +[2023-04-30 12:44:00,076][678703] Conv encoder output size: 512 +[2023-04-30 12:44:00,077][678703] Policy head output size: 512 +[2023-04-30 12:44:00,093][678550] Inference worker 0-0 is ready! +[2023-04-30 12:44:00,093][678550] All inference workers are ready! Signal rollout workers to start! +[2023-04-30 12:44:00,109][678704] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:44:00,111][678707] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:44:00,112][678705] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:44:00,113][678710] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:44:00,113][678706] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:44:00,113][678711] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:44:00,113][678708] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:44:00,121][678709] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:44:00,516][678704] Decorrelating experience for 0 frames... +[2023-04-30 12:44:00,516][678707] Decorrelating experience for 0 frames... +[2023-04-30 12:44:00,516][678706] Decorrelating experience for 0 frames... +[2023-04-30 12:44:00,516][678708] Decorrelating experience for 0 frames... +[2023-04-30 12:44:00,516][678710] Decorrelating experience for 0 frames... +[2023-04-30 12:44:00,516][678705] Decorrelating experience for 0 frames... +[2023-04-30 12:44:00,701][678707] Decorrelating experience for 32 frames... +[2023-04-30 12:44:00,702][678704] Decorrelating experience for 32 frames... +[2023-04-30 12:44:00,702][678705] Decorrelating experience for 32 frames... +[2023-04-30 12:44:00,703][678710] Decorrelating experience for 32 frames... +[2023-04-30 12:44:00,709][678711] Decorrelating experience for 0 frames... +[2023-04-30 12:44:00,711][678706] Decorrelating experience for 32 frames... +[2023-04-30 12:44:00,740][678709] Decorrelating experience for 0 frames... +[2023-04-30 12:44:00,891][678708] Decorrelating experience for 32 frames... +[2023-04-30 12:44:00,892][678707] Decorrelating experience for 64 frames... +[2023-04-30 12:44:00,918][678704] Decorrelating experience for 64 frames... +[2023-04-30 12:44:00,936][678709] Decorrelating experience for 32 frames... +[2023-04-30 12:44:01,119][678711] Decorrelating experience for 32 frames... +[2023-04-30 12:44:01,129][678705] Decorrelating experience for 64 frames... +[2023-04-30 12:44:01,133][678706] Decorrelating experience for 64 frames... +[2023-04-30 12:44:01,173][678708] Decorrelating experience for 64 frames... +[2023-04-30 12:44:01,319][678710] Decorrelating experience for 64 frames... +[2023-04-30 12:44:01,335][678711] Decorrelating experience for 64 frames... +[2023-04-30 12:44:01,349][678704] Decorrelating experience for 96 frames... +[2023-04-30 12:44:01,368][678706] Decorrelating experience for 96 frames... +[2023-04-30 12:44:01,538][678710] Decorrelating experience for 96 frames... +[2023-04-30 12:44:01,538][678709] Decorrelating experience for 64 frames... +[2023-04-30 12:44:01,565][678708] Decorrelating experience for 96 frames... +[2023-04-30 12:44:01,590][678711] Decorrelating experience for 96 frames... +[2023-04-30 12:44:01,635][678704] Decorrelating experience for 128 frames... +[2023-04-30 12:44:01,655][678706] Decorrelating experience for 128 frames... +[2023-04-30 12:44:01,771][678705] Decorrelating experience for 96 frames... +[2023-04-30 12:44:01,840][678550] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 0. Throughput: 0: nan. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) +[2023-04-30 12:44:01,857][678708] Decorrelating experience for 128 frames... +[2023-04-30 12:44:01,882][678711] Decorrelating experience for 128 frames... +[2023-04-30 12:44:01,931][678709] Decorrelating experience for 96 frames... +[2023-04-30 12:44:01,939][678706] Decorrelating experience for 160 frames... +[2023-04-30 12:44:02,000][678707] Decorrelating experience for 96 frames... +[2023-04-30 12:44:02,043][678705] Decorrelating experience for 128 frames... +[2023-04-30 12:44:02,044][678704] Decorrelating experience for 160 frames... +[2023-04-30 12:44:02,142][678708] Decorrelating experience for 160 frames... +[2023-04-30 12:44:02,239][678709] Decorrelating experience for 128 frames... +[2023-04-30 12:44:02,244][678710] Decorrelating experience for 128 frames... +[2023-04-30 12:44:02,265][678706] Decorrelating experience for 192 frames... +[2023-04-30 12:44:02,275][678711] Decorrelating experience for 160 frames... +[2023-04-30 12:44:02,335][678705] Decorrelating experience for 160 frames... +[2023-04-30 12:44:02,454][678708] Decorrelating experience for 192 frames... +[2023-04-30 12:44:02,491][678707] Decorrelating experience for 128 frames... +[2023-04-30 12:44:02,502][678704] Decorrelating experience for 192 frames... +[2023-04-30 12:44:02,528][678709] Decorrelating experience for 160 frames... +[2023-04-30 12:44:02,675][678710] Decorrelating experience for 160 frames... +[2023-04-30 12:44:02,721][678711] Decorrelating experience for 192 frames... +[2023-04-30 12:44:02,749][678708] Decorrelating experience for 224 frames... +[2023-04-30 12:44:02,749][678707] Decorrelating experience for 160 frames... +[2023-04-30 12:44:02,777][678706] Decorrelating experience for 224 frames... +[2023-04-30 12:44:02,865][678709] Decorrelating experience for 192 frames... +[2023-04-30 12:44:02,917][678704] Decorrelating experience for 224 frames... +[2023-04-30 12:44:02,984][678710] Decorrelating experience for 192 frames... +[2023-04-30 12:44:03,052][678705] Decorrelating experience for 192 frames... +[2023-04-30 12:44:03,056][678711] Decorrelating experience for 224 frames... +[2023-04-30 12:44:03,067][678707] Decorrelating experience for 192 frames... +[2023-04-30 12:44:03,308][678710] Decorrelating experience for 224 frames... +[2023-04-30 12:44:03,319][678709] Decorrelating experience for 224 frames... +[2023-04-30 12:44:03,354][678707] Decorrelating experience for 224 frames... +[2023-04-30 12:44:03,354][678705] Decorrelating experience for 224 frames... +[2023-04-30 12:44:04,423][678641] Signal inference workers to stop experience collection... +[2023-04-30 12:44:04,443][678703] InferenceWorker_p0-w0: stopping experience collection +[2023-04-30 12:44:05,428][678641] Signal inference workers to resume experience collection... +[2023-04-30 12:44:05,429][678703] InferenceWorker_p0-w0: resuming experience collection +[2023-04-30 12:44:06,840][678550] Fps is (10 sec: 819.2, 60 sec: 819.2, 300 sec: 819.2). Total num frames: 4096. Throughput: 0: 772.0. Samples: 3860. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0) +[2023-04-30 12:44:06,840][678550] Avg episode reward: [(0, '1.905')] +[2023-04-30 12:44:09,728][678641] Stopping Batcher_0... +[2023-04-30 12:44:09,728][678641] Loop batcher_evt_loop terminating... +[2023-04-30 12:44:09,734][678550] Component Batcher_0 stopped! +[2023-04-30 12:44:09,741][678708] Stopping RolloutWorker_w4... +[2023-04-30 12:44:09,741][678708] Loop rollout_proc4_evt_loop terminating... +[2023-04-30 12:44:09,741][678550] Component RolloutWorker_w4 stopped! +[2023-04-30 12:44:09,741][678706] Stopping RolloutWorker_w2... +[2023-04-30 12:44:09,741][678705] Stopping RolloutWorker_w1... +[2023-04-30 12:44:09,741][678550] Component RolloutWorker_w2 stopped! +[2023-04-30 12:44:09,741][678550] Component RolloutWorker_w1 stopped! +[2023-04-30 12:44:09,741][678706] Loop rollout_proc2_evt_loop terminating... +[2023-04-30 12:44:09,742][678550] Component RolloutWorker_w3 stopped! +[2023-04-30 12:44:09,742][678705] Loop rollout_proc1_evt_loop terminating... +[2023-04-30 12:44:09,742][678707] Stopping RolloutWorker_w3... +[2023-04-30 12:44:09,742][678550] Component RolloutWorker_w5 stopped! +[2023-04-30 12:44:09,742][678709] Stopping RolloutWorker_w5... +[2023-04-30 12:44:09,742][678704] Stopping RolloutWorker_w0... +[2023-04-30 12:44:09,742][678550] Component RolloutWorker_w0 stopped! +[2023-04-30 12:44:09,742][678711] Stopping RolloutWorker_w7... +[2023-04-30 12:44:09,742][678709] Loop rollout_proc5_evt_loop terminating... +[2023-04-30 12:44:09,742][678707] Loop rollout_proc3_evt_loop terminating... +[2023-04-30 12:44:09,743][678704] Loop rollout_proc0_evt_loop terminating... +[2023-04-30 12:44:09,743][678550] Component RolloutWorker_w7 stopped! +[2023-04-30 12:44:09,743][678711] Loop rollout_proc7_evt_loop terminating... +[2023-04-30 12:44:09,743][678550] Component RolloutWorker_w6 stopped! +[2023-04-30 12:44:09,743][678710] Stopping RolloutWorker_w6... +[2023-04-30 12:44:09,744][678710] Loop rollout_proc6_evt_loop terminating... +[2023-04-30 12:44:10,050][678703] Weights refcount: 2 0 +[2023-04-30 12:44:10,050][678703] Stopping InferenceWorker_p0-w0... +[2023-04-30 12:44:10,051][678703] Loop inference_proc0-0_evt_loop terminating... +[2023-04-30 12:44:10,051][678550] Component InferenceWorker_p0-w0 stopped! +[2023-04-30 12:44:11,141][678641] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000005_20480.pth... +[2023-04-30 12:44:11,164][678641] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000005_20480.pth... +[2023-04-30 12:44:11,190][678641] Stopping LearnerWorker_p0... +[2023-04-30 12:44:11,190][678550] Component LearnerWorker_p0 stopped! +[2023-04-30 12:44:11,190][678641] Loop learner_proc0_evt_loop terminating... +[2023-04-30 12:44:11,191][678550] Waiting for process learner_proc0 to stop... +[2023-04-30 12:44:11,368][678550] Waiting for process inference_proc0-0 to join... +[2023-04-30 12:44:11,369][678550] Waiting for process rollout_proc0 to join... +[2023-04-30 12:44:11,369][678550] Waiting for process rollout_proc1 to join... +[2023-04-30 12:44:11,369][678550] Waiting for process rollout_proc2 to join... +[2023-04-30 12:44:11,369][678550] Waiting for process rollout_proc3 to join... +[2023-04-30 12:44:11,369][678550] Waiting for process rollout_proc4 to join... +[2023-04-30 12:44:11,369][678550] Waiting for process rollout_proc5 to join... +[2023-04-30 12:44:11,370][678550] Waiting for process rollout_proc6 to join... +[2023-04-30 12:44:11,370][678550] Waiting for process rollout_proc7 to join... +[2023-04-30 12:44:11,370][678550] Batcher 0 profile tree view: +batching: 0.0243, releasing_batches: 0.0008 +[2023-04-30 12:44:11,370][678550] InferenceWorker_p0-w0 profile tree view: +wait_policy: 0.0051 + wait_policy_total: 2.8056 +update_model: 0.7312 + weight_update: 0.1811 +one_step: 0.0370 + handle_policy_step: 3.0217 + deserialize: 0.0424, stack: 0.0035, obs_to_device_normalize: 0.2143, forward: 2.6394, send_messages: 0.0531 + prepare_outputs: 0.0322 + to_cpu: 0.0028 +[2023-04-30 12:44:11,370][678550] Learner 0 profile tree view: +misc: 0.0000, prepare_batch: 1.5143 +train: 5.6552 + epoch_init: 0.0000, minibatch_init: 0.0000, losses_postprocess: 0.0002, kl_divergence: 0.0007, after_optimizer: 0.0041 + calculate_losses: 1.9942 + losses_init: 0.0000, forward_head: 1.4543, bptt_initial: 0.0072, tail: 0.0036, advantages_returns: 0.0006, losses: 0.0035 + bptt: 0.5243 + bptt_forward_core: 0.5233 + update: 3.6540 + clip: 0.0070 +[2023-04-30 12:44:11,370][678550] RolloutWorker_w0 profile tree view: +wait_for_trajectories: 0.0006, enqueue_policy_requests: 0.0288, env_step: 0.6001, overhead: 0.0480, complete_rollouts: 0.0007 +save_policy_outputs: 0.0333 + split_output_tensors: 0.0154 +[2023-04-30 12:44:11,371][678550] RolloutWorker_w7 profile tree view: +wait_for_trajectories: 0.0006, enqueue_policy_requests: 0.0265, env_step: 0.5396, overhead: 0.0422, complete_rollouts: 0.0007 +save_policy_outputs: 0.0293 + split_output_tensors: 0.0137 +[2023-04-30 12:44:11,371][678550] Loop Runner_EvtLoop terminating... +[2023-04-30 12:44:11,371][678550] Runner profile tree view: +main_loop: 13.3725 +[2023-04-30 12:44:11,371][678550] Collected {0: 20480}, FPS: 1531.5 +[2023-04-30 12:45:34,038][682983] Saving configuration to ./train_dir/doom_health_gathering_supreme/config.json... +[2023-04-30 12:45:34,039][682983] Rollout worker 0 uses device cpu +[2023-04-30 12:45:34,039][682983] Rollout worker 1 uses device cpu +[2023-04-30 12:45:34,039][682983] Rollout worker 2 uses device cpu +[2023-04-30 12:45:34,039][682983] Rollout worker 3 uses device cpu +[2023-04-30 12:45:34,039][682983] Rollout worker 4 uses device cpu +[2023-04-30 12:45:34,039][682983] Rollout worker 5 uses device cpu +[2023-04-30 12:45:34,039][682983] Rollout worker 6 uses device cpu +[2023-04-30 12:45:34,040][682983] Rollout worker 7 uses device cpu +[2023-04-30 12:45:34,080][682983] InferenceWorker_p0-w0: min num requests: 2 +[2023-04-30 12:45:34,139][682983] Starting all processes... +[2023-04-30 12:45:34,139][682983] Starting process learner_proc0 +[2023-04-30 12:45:34,957][682983] Starting all processes... +[2023-04-30 12:45:34,961][683074] Starting seed is not provided +[2023-04-30 12:45:34,961][683074] Initializing actor-critic model on device cpu +[2023-04-30 12:45:34,961][682983] Starting process inference_proc0-0 +[2023-04-30 12:45:34,961][683074] RunningMeanStd input shape: (3, 72, 128) +[2023-04-30 12:45:34,961][682983] Starting process rollout_proc0 +[2023-04-30 12:45:34,962][683074] RunningMeanStd input shape: (1,) +[2023-04-30 12:45:34,962][682983] Starting process rollout_proc1 +[2023-04-30 12:45:34,962][682983] Starting process rollout_proc2 +[2023-04-30 12:45:34,969][683074] ConvEncoder: input_channels=3 +[2023-04-30 12:45:34,963][682983] Starting process rollout_proc3 +[2023-04-30 12:45:34,967][682983] Starting process rollout_proc4 +[2023-04-30 12:45:34,967][682983] Starting process rollout_proc5 +[2023-04-30 12:45:34,967][682983] Starting process rollout_proc6 +[2023-04-30 12:45:34,968][682983] Starting process rollout_proc7 +[2023-04-30 12:45:35,069][683074] Conv encoder output size: 512 +[2023-04-30 12:45:35,070][683074] Policy head output size: 512 +[2023-04-30 12:45:35,079][683074] Created Actor Critic model with architecture: +[2023-04-30 12:45:35,079][683074] ActorCriticSharedWeights( (obs_normalizer): ObservationNormalizer( (running_mean_std): RunningMeanStdDictInPlace( (running_mean_std): ModuleDict( @@ -710,494 +319,983 @@ main_loop: 217.2086 (distribution_linear): Linear(in_features=512, out_features=5, bias=True) ) ) -[2023-04-29 19:05:20,571][133597] Using optimizer -[2023-04-29 19:05:20,572][133597] Loading state from checkpoint /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... -[2023-04-29 19:05:20,603][133597] Loading model from checkpoint -[2023-04-29 19:05:20,607][133597] Loaded experiment state at self.train_step=978, self.env_steps=4005888 -[2023-04-29 19:05:20,607][133597] Initialized policy 0 weights for model version 978 -[2023-04-29 19:05:20,610][133597] LearnerWorker_p0 finished initialization! -[2023-04-29 19:05:20,610][133597] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-29 19:05:20,727][133610] RunningMeanStd input shape: (3, 72, 128) -[2023-04-29 19:05:20,728][133610] RunningMeanStd input shape: (1,) -[2023-04-29 19:05:20,735][133610] ConvEncoder: input_channels=3 -[2023-04-29 19:05:20,794][133610] Conv encoder output size: 512 -[2023-04-29 19:05:20,794][133610] Policy head output size: 512 -[2023-04-29 19:05:20,912][108205] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 4005888. Throughput: 0: nan. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) -[2023-04-29 19:05:21,937][108205] Inference worker 0-0 is ready! -[2023-04-29 19:05:21,938][108205] All inference workers are ready! Signal rollout workers to start! -[2023-04-29 19:05:21,984][133613] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:05:21,985][133618] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:05:21,988][133615] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:05:21,990][133611] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:05:21,991][133612] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:05:21,993][133617] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:05:21,993][133616] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:05:22,000][133614] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:05:22,396][133611] Decorrelating experience for 0 frames... -[2023-04-29 19:05:22,396][133614] Decorrelating experience for 0 frames... -[2023-04-29 19:05:22,396][133613] Decorrelating experience for 0 frames... -[2023-04-29 19:05:22,397][133615] Decorrelating experience for 0 frames... -[2023-04-29 19:05:22,397][133612] Decorrelating experience for 0 frames... -[2023-04-29 19:05:22,398][133617] Decorrelating experience for 0 frames... -[2023-04-29 19:05:22,581][133612] Decorrelating experience for 32 frames... -[2023-04-29 19:05:22,582][133615] Decorrelating experience for 32 frames... -[2023-04-29 19:05:22,583][133613] Decorrelating experience for 32 frames... -[2023-04-29 19:05:22,616][133611] Decorrelating experience for 32 frames... -[2023-04-29 19:05:22,632][133614] Decorrelating experience for 32 frames... -[2023-04-29 19:05:22,807][133617] Decorrelating experience for 32 frames... -[2023-04-29 19:05:22,809][133616] Decorrelating experience for 0 frames... -[2023-04-29 19:05:22,836][133615] Decorrelating experience for 64 frames... -[2023-04-29 19:05:22,857][133612] Decorrelating experience for 64 frames... -[2023-04-29 19:05:22,872][133611] Decorrelating experience for 64 frames... -[2023-04-29 19:05:22,893][133618] Decorrelating experience for 0 frames... -[2023-04-29 19:05:23,027][133616] Decorrelating experience for 32 frames... -[2023-04-29 19:05:23,051][133617] Decorrelating experience for 64 frames... -[2023-04-29 19:05:23,083][133613] Decorrelating experience for 64 frames... -[2023-04-29 19:05:23,101][133612] Decorrelating experience for 96 frames... -[2023-04-29 19:05:23,280][133618] Decorrelating experience for 32 frames... -[2023-04-29 19:05:23,298][133616] Decorrelating experience for 64 frames... -[2023-04-29 19:05:23,324][133617] Decorrelating experience for 96 frames... -[2023-04-29 19:05:23,348][133613] Decorrelating experience for 96 frames... -[2023-04-29 19:05:23,508][133618] Decorrelating experience for 64 frames... -[2023-04-29 19:05:23,508][133615] Decorrelating experience for 96 frames... -[2023-04-29 19:05:23,563][133616] Decorrelating experience for 96 frames... -[2023-04-29 19:05:23,581][133614] Decorrelating experience for 64 frames... -[2023-04-29 19:05:23,773][133618] Decorrelating experience for 96 frames... -[2023-04-29 19:05:23,829][133611] Decorrelating experience for 96 frames... -[2023-04-29 19:05:23,845][133614] Decorrelating experience for 96 frames... -[2023-04-29 19:05:24,309][133597] Signal inference workers to stop experience collection... -[2023-04-29 19:05:24,312][133610] InferenceWorker_p0-w0: stopping experience collection -[2023-04-29 19:05:25,912][108205] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 4005888. Throughput: 0: 507.2. Samples: 2536. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) -[2023-04-29 19:05:25,912][108205] Avg episode reward: [(0, '2.019')] -[2023-04-29 19:05:26,586][133597] Signal inference workers to resume experience collection... -[2023-04-29 19:05:26,587][133610] InferenceWorker_p0-w0: resuming experience collection -[2023-04-29 19:05:26,587][133597] Stopping Batcher_0... -[2023-04-29 19:05:26,587][133597] Loop batcher_evt_loop terminating... -[2023-04-29 19:05:26,593][133613] Stopping RolloutWorker_w3... -[2023-04-29 19:05:26,593][133616] Stopping RolloutWorker_w5... -[2023-04-29 19:05:26,594][133613] Loop rollout_proc3_evt_loop terminating... -[2023-04-29 19:05:26,594][133616] Loop rollout_proc5_evt_loop terminating... -[2023-04-29 19:05:26,594][133612] Stopping RolloutWorker_w1... -[2023-04-29 19:05:26,595][133617] Stopping RolloutWorker_w6... -[2023-04-29 19:05:26,595][133615] Stopping RolloutWorker_w4... -[2023-04-29 19:05:26,595][133618] Stopping RolloutWorker_w7... -[2023-04-29 19:05:26,595][133612] Loop rollout_proc1_evt_loop terminating... -[2023-04-29 19:05:26,595][133614] Stopping RolloutWorker_w2... -[2023-04-29 19:05:26,595][133615] Loop rollout_proc4_evt_loop terminating... -[2023-04-29 19:05:26,595][133617] Loop rollout_proc6_evt_loop terminating... -[2023-04-29 19:05:26,595][133618] Loop rollout_proc7_evt_loop terminating... -[2023-04-29 19:05:26,595][133611] Stopping RolloutWorker_w0... -[2023-04-29 19:05:26,595][133614] Loop rollout_proc2_evt_loop terminating... -[2023-04-29 19:05:26,595][133611] Loop rollout_proc0_evt_loop terminating... -[2023-04-29 19:05:26,596][133610] Weights refcount: 2 0 -[2023-04-29 19:05:26,597][133610] Stopping InferenceWorker_p0-w0... -[2023-04-29 19:05:26,598][133610] Loop inference_proc0-0_evt_loop terminating... -[2023-04-29 19:05:26,598][108205] Component Batcher_0 stopped! -[2023-04-29 19:05:26,601][108205] Component RolloutWorker_w3 stopped! -[2023-04-29 19:05:26,602][108205] Component RolloutWorker_w5 stopped! -[2023-04-29 19:05:26,603][108205] Component RolloutWorker_w1 stopped! -[2023-04-29 19:05:26,604][108205] Component RolloutWorker_w6 stopped! -[2023-04-29 19:05:26,604][108205] Component RolloutWorker_w4 stopped! -[2023-04-29 19:05:26,605][108205] Component RolloutWorker_w7 stopped! -[2023-04-29 19:05:26,606][108205] Component RolloutWorker_w2 stopped! -[2023-04-29 19:05:26,607][108205] Component RolloutWorker_w0 stopped! -[2023-04-29 19:05:26,607][108205] Component InferenceWorker_p0-w0 stopped! -[2023-04-29 19:05:26,740][108205] Keyboard interrupt detected in the event loop EvtLoop [Runner_EvtLoop, process=main process 108205], exiting... -[2023-04-29 19:05:26,741][108205] Runner profile tree view: -main_loop: 9.2016 -[2023-04-29 19:05:26,742][108205] Collected {0: 4009984}, FPS: 445.1 -[2023-04-29 19:05:27,185][133597] Saving /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000980_4014080.pth... -[2023-04-29 19:05:27,217][133597] Removing /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000517_2117632.pth -[2023-04-29 19:05:27,218][133597] Saving /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000980_4014080.pth... -[2023-04-29 19:05:27,250][133597] Stopping LearnerWorker_p0... -[2023-04-29 19:05:27,250][133597] Loop learner_proc0_evt_loop terminating... -[2023-04-29 19:17:05,974][139883] Saving configuration to /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/config.json... -[2023-04-29 19:17:05,975][139883] Rollout worker 0 uses device cpu -[2023-04-29 19:17:05,975][139883] Rollout worker 1 uses device cpu -[2023-04-29 19:17:05,976][139883] Rollout worker 2 uses device cpu -[2023-04-29 19:17:05,976][139883] Rollout worker 3 uses device cpu -[2023-04-29 19:17:05,977][139883] Rollout worker 4 uses device cpu -[2023-04-29 19:17:05,977][139883] Rollout worker 5 uses device cpu -[2023-04-29 19:17:05,978][139883] Rollout worker 6 uses device cpu -[2023-04-29 19:17:05,978][139883] Rollout worker 7 uses device cpu -[2023-04-29 19:17:06,002][139883] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-29 19:17:06,002][139883] InferenceWorker_p0-w0: min num requests: 2 -[2023-04-29 19:17:06,060][139883] Starting all processes... -[2023-04-29 19:17:06,061][139883] Starting process learner_proc0 -[2023-04-29 19:17:06,110][139883] Starting all processes... -[2023-04-29 19:17:06,114][139883] Starting process inference_proc0-0 -[2023-04-29 19:17:06,114][139883] Starting process rollout_proc0 -[2023-04-29 19:17:06,114][139883] Starting process rollout_proc1 -[2023-04-29 19:17:06,114][139883] Starting process rollout_proc2 -[2023-04-29 19:17:06,115][139883] Starting process rollout_proc3 -[2023-04-29 19:17:06,115][139883] Starting process rollout_proc4 -[2023-04-29 19:17:06,115][139883] Starting process rollout_proc5 -[2023-04-29 19:17:06,116][139883] Starting process rollout_proc6 -[2023-04-29 19:17:06,116][139883] Starting process rollout_proc7 -[2023-04-29 19:17:06,979][141009] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-29 19:17:06,979][141009] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0 -[2023-04-29 19:17:06,997][141024] Worker 1 uses CPU cores [3, 4, 5] -[2023-04-29 19:17:07,016][141009] Num visible devices: 1 -[2023-04-29 19:17:07,024][141026] Worker 3 uses CPU cores [9, 10, 11] -[2023-04-29 19:17:07,027][141030] Worker 7 uses CPU cores [21, 22, 23] -[2023-04-29 19:17:07,038][141028] Worker 4 uses CPU cores [12, 13, 14] -[2023-04-29 19:17:07,039][141022] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-29 19:17:07,039][141022] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0 -[2023-04-29 19:17:07,045][141023] Worker 0 uses CPU cores [0, 1, 2] -[2023-04-29 19:17:07,053][141022] Num visible devices: 1 -[2023-04-29 19:17:07,054][141009] Starting seed is not provided -[2023-04-29 19:17:07,054][141009] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-29 19:17:07,054][141009] Initializing actor-critic model on device cuda:0 -[2023-04-29 19:17:07,054][141009] RunningMeanStd input shape: (3, 72, 128) -[2023-04-29 19:17:07,055][141009] RunningMeanStd input shape: (1,) -[2023-04-29 19:17:07,059][141027] Worker 5 uses CPU cores [15, 16, 17] -[2023-04-29 19:17:07,065][141009] ConvEncoder: input_channels=3 -[2023-04-29 19:17:07,066][141025] Worker 2 uses CPU cores [6, 7, 8] -[2023-04-29 19:17:07,085][141029] Worker 6 uses CPU cores [18, 19, 20] -[2023-04-29 19:17:07,208][141009] Conv encoder output size: 512 -[2023-04-29 19:17:07,208][141009] Policy head output size: 512 -[2023-04-29 19:17:07,247][141009] Created Actor Critic model with architecture: -[2023-04-29 19:17:07,247][141009] ActorCriticSharedWeights( - (obs_normalizer): ObservationNormalizer( - (running_mean_std): RunningMeanStdDictInPlace( - (running_mean_std): ModuleDict( - (obs): RunningMeanStdInPlace() - ) - ) - ) - (returns_normalizer): RecursiveScriptModule(original_name=RunningMeanStdInPlace) - (encoder): VizdoomEncoder( - (basic_encoder): ConvEncoder( - (enc): RecursiveScriptModule( - original_name=ConvEncoderImpl - (conv_head): RecursiveScriptModule( - original_name=Sequential - (0): RecursiveScriptModule(original_name=Conv2d) - (1): RecursiveScriptModule(original_name=ELU) - (2): RecursiveScriptModule(original_name=Conv2d) - (3): RecursiveScriptModule(original_name=ELU) - (4): RecursiveScriptModule(original_name=Conv2d) - (5): RecursiveScriptModule(original_name=ELU) - ) - (mlp_layers): RecursiveScriptModule( - original_name=Sequential - (0): RecursiveScriptModule(original_name=Linear) - (1): RecursiveScriptModule(original_name=ELU) - ) - ) - ) - ) - (core): ModelCoreRNN( - (core): GRU(512, 512) - ) - (decoder): MlpDecoder( - (mlp): Identity() - ) - (critic_linear): Linear(in_features=512, out_features=1, bias=True) - (action_parameterization): ActionParameterizationDefault( - (distribution_linear): Linear(in_features=512, out_features=5, bias=True) - ) -) -[2023-04-29 19:17:08,982][141009] Using optimizer -[2023-04-29 19:17:08,983][141009] Loading state from checkpoint /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000980_4014080.pth... -[2023-04-29 19:17:08,999][141009] Loading model from checkpoint -[2023-04-29 19:17:09,002][141009] Loaded experiment state at self.train_step=980, self.env_steps=4014080 -[2023-04-29 19:17:09,002][141009] Initialized policy 0 weights for model version 980 -[2023-04-29 19:17:09,005][141009] LearnerWorker_p0 finished initialization! -[2023-04-29 19:17:09,006][141009] Using GPUs [0] for process 0 (actually maps to GPUs [0]) -[2023-04-29 19:17:09,125][141022] RunningMeanStd input shape: (3, 72, 128) -[2023-04-29 19:17:09,126][141022] RunningMeanStd input shape: (1,) -[2023-04-29 19:17:09,133][141022] ConvEncoder: input_channels=3 -[2023-04-29 19:17:09,191][141022] Conv encoder output size: 512 -[2023-04-29 19:17:09,191][141022] Policy head output size: 512 -[2023-04-29 19:17:09,422][139883] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 4014080. Throughput: 0: nan. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) -[2023-04-29 19:17:10,104][139883] Inference worker 0-0 is ready! -[2023-04-29 19:17:10,104][139883] All inference workers are ready! Signal rollout workers to start! -[2023-04-29 19:17:10,122][141028] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:17:10,123][141024] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:17:10,123][141029] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:17:10,123][141030] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:17:10,123][141025] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:17:10,124][141026] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:17:10,124][141027] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:17:10,124][141023] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:17:10,323][141026] Decorrelating experience for 0 frames... -[2023-04-29 19:17:10,324][141029] Decorrelating experience for 0 frames... -[2023-04-29 19:17:10,324][141025] Decorrelating experience for 0 frames... -[2023-04-29 19:17:10,325][141023] Decorrelating experience for 0 frames... -[2023-04-29 19:17:10,327][141028] Decorrelating experience for 0 frames... -[2023-04-29 19:17:10,332][141024] Decorrelating experience for 0 frames... -[2023-04-29 19:17:10,495][141026] Decorrelating experience for 32 frames... -[2023-04-29 19:17:10,495][141025] Decorrelating experience for 32 frames... -[2023-04-29 19:17:10,519][141024] Decorrelating experience for 32 frames... -[2023-04-29 19:17:10,520][141030] Decorrelating experience for 0 frames... -[2023-04-29 19:17:10,557][141029] Decorrelating experience for 32 frames... -[2023-04-29 19:17:10,699][141026] Decorrelating experience for 64 frames... -[2023-04-29 19:17:10,700][141030] Decorrelating experience for 32 frames... -[2023-04-29 19:17:10,754][141027] Decorrelating experience for 0 frames... -[2023-04-29 19:17:10,784][141029] Decorrelating experience for 64 frames... -[2023-04-29 19:17:10,909][141023] Decorrelating experience for 32 frames... -[2023-04-29 19:17:10,909][141025] Decorrelating experience for 64 frames... -[2023-04-29 19:17:10,911][141026] Decorrelating experience for 96 frames... -[2023-04-29 19:17:10,944][141030] Decorrelating experience for 64 frames... -[2023-04-29 19:17:10,978][141027] Decorrelating experience for 32 frames... -[2023-04-29 19:17:11,007][141024] Decorrelating experience for 64 frames... -[2023-04-29 19:17:11,118][141029] Decorrelating experience for 96 frames... -[2023-04-29 19:17:11,144][141025] Decorrelating experience for 96 frames... -[2023-04-29 19:17:11,193][141030] Decorrelating experience for 96 frames... -[2023-04-29 19:17:11,214][141024] Decorrelating experience for 96 frames... -[2023-04-29 19:17:11,214][141027] Decorrelating experience for 64 frames... -[2023-04-29 19:17:11,327][141028] Decorrelating experience for 32 frames... -[2023-04-29 19:17:11,522][141027] Decorrelating experience for 96 frames... -[2023-04-29 19:17:11,550][141023] Decorrelating experience for 64 frames... -[2023-04-29 19:17:11,812][141028] Decorrelating experience for 64 frames... -[2023-04-29 19:17:11,844][141023] Decorrelating experience for 96 frames... -[2023-04-29 19:17:11,880][141009] Signal inference workers to stop experience collection... -[2023-04-29 19:17:11,883][141022] InferenceWorker_p0-w0: stopping experience collection -[2023-04-29 19:17:12,064][141028] Decorrelating experience for 96 frames... -[2023-04-29 19:17:13,307][141009] Signal inference workers to resume experience collection... -[2023-04-29 19:17:13,307][141022] InferenceWorker_p0-w0: resuming experience collection -[2023-04-29 19:17:13,308][141009] Stopping Batcher_0... -[2023-04-29 19:17:13,308][141009] Loop batcher_evt_loop terminating... -[2023-04-29 19:17:13,313][141026] Stopping RolloutWorker_w3... -[2023-04-29 19:17:13,313][141027] Stopping RolloutWorker_w5... -[2023-04-29 19:17:13,313][141023] Stopping RolloutWorker_w0... -[2023-04-29 19:17:13,313][141026] Loop rollout_proc3_evt_loop terminating... -[2023-04-29 19:17:13,313][141027] Loop rollout_proc5_evt_loop terminating... -[2023-04-29 19:17:13,313][141023] Loop rollout_proc0_evt_loop terminating... -[2023-04-29 19:17:13,313][141024] Stopping RolloutWorker_w1... -[2023-04-29 19:17:13,313][141030] Stopping RolloutWorker_w7... -[2023-04-29 19:17:13,313][141024] Loop rollout_proc1_evt_loop terminating... -[2023-04-29 19:17:13,314][141030] Loop rollout_proc7_evt_loop terminating... -[2023-04-29 19:17:13,313][141025] Stopping RolloutWorker_w2... -[2023-04-29 19:17:13,314][141025] Loop rollout_proc2_evt_loop terminating... -[2023-04-29 19:17:13,314][141028] Stopping RolloutWorker_w4... -[2023-04-29 19:17:13,314][141029] Stopping RolloutWorker_w6... -[2023-04-29 19:17:13,314][141028] Loop rollout_proc4_evt_loop terminating... -[2023-04-29 19:17:13,314][141029] Loop rollout_proc6_evt_loop terminating... -[2023-04-29 19:17:13,314][141022] Weights refcount: 2 0 -[2023-04-29 19:17:13,316][141022] Stopping InferenceWorker_p0-w0... -[2023-04-29 19:17:13,316][141022] Loop inference_proc0-0_evt_loop terminating... -[2023-04-29 19:17:13,319][139883] Component Batcher_0 stopped! -[2023-04-29 19:17:13,322][139883] Component RolloutWorker_w3 stopped! -[2023-04-29 19:17:13,322][139883] Component RolloutWorker_w5 stopped! -[2023-04-29 19:17:13,323][139883] Component RolloutWorker_w0 stopped! -[2023-04-29 19:17:13,324][139883] Component RolloutWorker_w1 stopped! -[2023-04-29 19:17:13,325][139883] Component RolloutWorker_w7 stopped! -[2023-04-29 19:17:13,325][139883] Component RolloutWorker_w2 stopped! -[2023-04-29 19:17:13,326][139883] Component RolloutWorker_w4 stopped! -[2023-04-29 19:17:13,327][139883] Component RolloutWorker_w6 stopped! -[2023-04-29 19:17:13,328][139883] Component InferenceWorker_p0-w0 stopped! -[2023-04-29 19:17:13,811][141009] Saving /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000982_4022272.pth... -[2023-04-29 19:17:13,840][141009] Removing /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth -[2023-04-29 19:17:13,842][141009] Saving /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000982_4022272.pth... -[2023-04-29 19:17:13,873][141009] Stopping LearnerWorker_p0... -[2023-04-29 19:17:13,874][141009] Loop learner_proc0_evt_loop terminating... -[2023-04-29 19:17:13,874][139883] Component LearnerWorker_p0 stopped! -[2023-04-29 19:17:13,874][139883] Waiting for process learner_proc0 to stop... -[2023-04-29 19:17:14,380][139883] Waiting for process inference_proc0-0 to join... -[2023-04-29 19:17:14,381][139883] Waiting for process rollout_proc0 to join... -[2023-04-29 19:17:14,382][139883] Waiting for process rollout_proc1 to join... -[2023-04-29 19:17:14,382][139883] Waiting for process rollout_proc2 to join... -[2023-04-29 19:17:14,383][139883] Waiting for process rollout_proc3 to join... -[2023-04-29 19:17:14,383][139883] Waiting for process rollout_proc4 to join... -[2023-04-29 19:17:14,384][139883] Waiting for process rollout_proc5 to join... -[2023-04-29 19:17:14,385][139883] Waiting for process rollout_proc6 to join... -[2023-04-29 19:17:14,385][139883] Waiting for process rollout_proc7 to join... -[2023-04-29 19:17:14,386][139883] Batcher 0 profile tree view: -batching: 0.0301, releasing_batches: 0.0006 -[2023-04-29 19:17:14,386][139883] InferenceWorker_p0-w0 profile tree view: -update_model: 0.0040 -wait_policy: 0.0000 - wait_policy_total: 0.8549 -one_step: 0.0023 - handle_policy_step: 0.8927 - deserialize: 0.0172, stack: 0.0021, obs_to_device_normalize: 0.1624, forward: 0.5764, send_messages: 0.0347 - prepare_outputs: 0.0846 - to_cpu: 0.0663 -[2023-04-29 19:17:14,387][139883] Learner 0 profile tree view: -misc: 0.0000, prepare_batch: 1.5413 -train: 0.6188 - epoch_init: 0.0000, minibatch_init: 0.0000, losses_postprocess: 0.0004, kl_divergence: 0.0005, after_optimizer: 0.0074 - calculate_losses: 0.0568 - losses_init: 0.0000, forward_head: 0.0480, bptt_initial: 0.0040, tail: 0.0008, advantages_returns: 0.0005, losses: 0.0018 - bptt: 0.0015 - bptt_forward_core: 0.0014 - update: 0.5533 - clip: 0.0038 -[2023-04-29 19:17:14,387][139883] RolloutWorker_w0 profile tree view: -wait_for_trajectories: 0.0001, enqueue_policy_requests: 0.0003 -[2023-04-29 19:17:14,388][139883] RolloutWorker_w7 profile tree view: -wait_for_trajectories: 0.0004, enqueue_policy_requests: 0.0197, env_step: 0.3010, overhead: 0.0222, complete_rollouts: 0.0005 -save_policy_outputs: 0.0155 - split_output_tensors: 0.0077 -[2023-04-29 19:17:14,389][139883] Loop Runner_EvtLoop terminating... -[2023-04-29 19:17:14,389][139883] Runner profile tree view: -main_loop: 8.3291 -[2023-04-29 19:17:14,390][139883] Collected {0: 4022272}, FPS: 983.5 -[2023-04-29 19:17:14,480][139883] Loading existing experiment configuration from /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/config.json -[2023-04-29 19:17:14,481][139883] Overriding arg 'num_workers' with value 1 passed from command line -[2023-04-29 19:17:14,482][139883] Adding new argument 'no_render'=True that is not in the saved config file! -[2023-04-29 19:17:14,482][139883] Adding new argument 'save_video'=True that is not in the saved config file! -[2023-04-29 19:17:14,483][139883] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! -[2023-04-29 19:17:14,483][139883] Adding new argument 'video_name'=None that is not in the saved config file! -[2023-04-29 19:17:14,484][139883] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file! -[2023-04-29 19:17:14,485][139883] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! -[2023-04-29 19:17:14,485][139883] Adding new argument 'push_to_hub'=False that is not in the saved config file! -[2023-04-29 19:17:14,486][139883] Adding new argument 'hf_repository'=None that is not in the saved config file! -[2023-04-29 19:17:14,486][139883] Adding new argument 'policy_index'=0 that is not in the saved config file! -[2023-04-29 19:17:14,487][139883] Adding new argument 'eval_deterministic'=False that is not in the saved config file! -[2023-04-29 19:17:14,487][139883] Adding new argument 'train_script'=None that is not in the saved config file! -[2023-04-29 19:17:14,488][139883] Adding new argument 'enjoy_script'=None that is not in the saved config file! -[2023-04-29 19:17:14,488][139883] Using frameskip 1 and render_action_repeat=4 for evaluation -[2023-04-29 19:17:14,494][139883] Doom resolution: 160x120, resize resolution: (128, 72) -[2023-04-29 19:17:14,495][139883] RunningMeanStd input shape: (3, 72, 128) -[2023-04-29 19:17:14,496][139883] RunningMeanStd input shape: (1,) -[2023-04-29 19:17:14,503][139883] ConvEncoder: input_channels=3 -[2023-04-29 19:17:14,588][139883] Conv encoder output size: 512 -[2023-04-29 19:17:14,589][139883] Policy head output size: 512 -[2023-04-29 19:17:16,265][139883] Loading state from checkpoint /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000982_4022272.pth... -[2023-04-29 19:17:17,030][139883] Num frames 100... -[2023-04-29 19:17:17,120][139883] Num frames 200... -[2023-04-29 19:17:17,212][139883] Num frames 300... -[2023-04-29 19:17:17,344][139883] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840 -[2023-04-29 19:17:17,345][139883] Avg episode reward: 3.840, avg true_objective: 3.840 -[2023-04-29 19:17:17,361][139883] Num frames 400... -[2023-04-29 19:17:17,454][139883] Num frames 500... -[2023-04-29 19:17:17,541][139883] Num frames 600... -[2023-04-29 19:17:17,628][139883] Num frames 700... -[2023-04-29 19:17:17,715][139883] Num frames 800... -[2023-04-29 19:17:17,815][139883] Num frames 900... -[2023-04-29 19:17:17,902][139883] Avg episode rewards: #0: 5.640, true rewards: #0: 4.640 -[2023-04-29 19:17:17,903][139883] Avg episode reward: 5.640, avg true_objective: 4.640 -[2023-04-29 19:17:17,978][139883] Num frames 1000... -[2023-04-29 19:17:18,072][139883] Num frames 1100... -[2023-04-29 19:17:18,162][139883] Num frames 1200... -[2023-04-29 19:17:18,256][139883] Num frames 1300... -[2023-04-29 19:17:18,380][139883] Avg episode rewards: #0: 5.587, true rewards: #0: 4.587 -[2023-04-29 19:17:18,381][139883] Avg episode reward: 5.587, avg true_objective: 4.587 -[2023-04-29 19:17:18,406][139883] Num frames 1400... -[2023-04-29 19:17:18,494][139883] Num frames 1500... -[2023-04-29 19:17:18,586][139883] Num frames 1600... -[2023-04-29 19:17:18,682][139883] Num frames 1700... -[2023-04-29 19:17:18,797][139883] Avg episode rewards: #0: 5.150, true rewards: #0: 4.400 -[2023-04-29 19:17:18,797][139883] Avg episode reward: 5.150, avg true_objective: 4.400 -[2023-04-29 19:17:18,838][139883] Num frames 1800... -[2023-04-29 19:17:18,947][139883] Num frames 1900... -[2023-04-29 19:17:19,053][139883] Num frames 2000... -[2023-04-29 19:17:19,150][139883] Num frames 2100... -[2023-04-29 19:17:19,247][139883] Avg episode rewards: #0: 4.888, true rewards: #0: 4.288 -[2023-04-29 19:17:19,248][139883] Avg episode reward: 4.888, avg true_objective: 4.288 -[2023-04-29 19:17:19,305][139883] Num frames 2200... -[2023-04-29 19:17:19,397][139883] Num frames 2300... -[2023-04-29 19:17:19,486][139883] Num frames 2400... -[2023-04-29 19:17:19,573][139883] Num frames 2500... -[2023-04-29 19:17:19,653][139883] Avg episode rewards: #0: 4.713, true rewards: #0: 4.213 -[2023-04-29 19:17:19,654][139883] Avg episode reward: 4.713, avg true_objective: 4.213 -[2023-04-29 19:17:19,722][139883] Num frames 2600... -[2023-04-29 19:17:19,815][139883] Num frames 2700... -[2023-04-29 19:17:19,907][139883] Num frames 2800... -[2023-04-29 19:17:20,017][139883] Num frames 2900... -[2023-04-29 19:17:20,095][139883] Avg episode rewards: #0: 4.589, true rewards: #0: 4.160 -[2023-04-29 19:17:20,096][139883] Avg episode reward: 4.589, avg true_objective: 4.160 -[2023-04-29 19:17:20,184][139883] Num frames 3000... -[2023-04-29 19:17:20,285][139883] Num frames 3100... -[2023-04-29 19:17:20,374][139883] Num frames 3200... -[2023-04-29 19:17:20,466][139883] Num frames 3300... -[2023-04-29 19:17:20,601][139883] Avg episode rewards: #0: 4.865, true rewards: #0: 4.240 -[2023-04-29 19:17:20,602][139883] Avg episode reward: 4.865, avg true_objective: 4.240 -[2023-04-29 19:17:20,610][139883] Num frames 3400... -[2023-04-29 19:17:20,698][139883] Num frames 3500... -[2023-04-29 19:17:20,787][139883] Num frames 3600... -[2023-04-29 19:17:20,883][139883] Num frames 3700... -[2023-04-29 19:17:20,973][139883] Num frames 3800... -[2023-04-29 19:17:21,035][139883] Avg episode rewards: #0: 4.898, true rewards: #0: 4.231 -[2023-04-29 19:17:21,036][139883] Avg episode reward: 4.898, avg true_objective: 4.231 -[2023-04-29 19:17:21,129][139883] Num frames 3900... -[2023-04-29 19:17:21,226][139883] Num frames 4000... -[2023-04-29 19:17:21,322][139883] Num frames 4100... -[2023-04-29 19:17:21,464][139883] Avg episode rewards: #0: 4.992, true rewards: #0: 4.192 -[2023-04-29 19:17:21,465][139883] Avg episode reward: 4.992, avg true_objective: 4.192 -[2023-04-29 19:17:26,026][139883] Replay video saved to /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/replay.mp4! -[2023-04-29 19:19:04,743][139883] Loading existing experiment configuration from /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/config.json -[2023-04-29 19:19:04,743][139883] Overriding arg 'num_workers' with value 1 passed from command line -[2023-04-29 19:19:04,744][139883] Adding new argument 'no_render'=True that is not in the saved config file! -[2023-04-29 19:19:04,744][139883] Adding new argument 'save_video'=True that is not in the saved config file! -[2023-04-29 19:19:04,745][139883] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! -[2023-04-29 19:19:04,746][139883] Adding new argument 'video_name'=None that is not in the saved config file! -[2023-04-29 19:19:04,746][139883] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! -[2023-04-29 19:19:04,747][139883] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! -[2023-04-29 19:19:04,747][139883] Adding new argument 'push_to_hub'=True that is not in the saved config file! -[2023-04-29 19:19:04,748][139883] Adding new argument 'hf_repository'='ItchyB/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! -[2023-04-29 19:19:04,748][139883] Adding new argument 'policy_index'=0 that is not in the saved config file! -[2023-04-29 19:19:04,749][139883] Adding new argument 'eval_deterministic'=False that is not in the saved config file! -[2023-04-29 19:19:04,750][139883] Adding new argument 'train_script'=None that is not in the saved config file! -[2023-04-29 19:19:04,750][139883] Adding new argument 'enjoy_script'=None that is not in the saved config file! -[2023-04-29 19:19:04,751][139883] Using frameskip 1 and render_action_repeat=4 for evaluation -[2023-04-29 19:19:04,754][139883] RunningMeanStd input shape: (3, 72, 128) -[2023-04-29 19:19:04,755][139883] RunningMeanStd input shape: (1,) -[2023-04-29 19:19:04,761][139883] ConvEncoder: input_channels=3 -[2023-04-29 19:19:04,782][139883] Conv encoder output size: 512 -[2023-04-29 19:19:04,783][139883] Policy head output size: 512 -[2023-04-29 19:19:04,801][139883] Loading state from checkpoint /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/checkpoint_p0/checkpoint_000000982_4022272.pth... -[2023-04-29 19:19:05,173][139883] Num frames 100... -[2023-04-29 19:19:05,315][139883] Num frames 200... -[2023-04-29 19:19:05,452][139883] Num frames 300... -[2023-04-29 19:19:05,634][139883] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840 -[2023-04-29 19:19:05,635][139883] Avg episode reward: 3.840, avg true_objective: 3.840 -[2023-04-29 19:19:05,658][139883] Num frames 400... -[2023-04-29 19:19:05,797][139883] Num frames 500... -[2023-04-29 19:19:05,955][139883] Num frames 600... -[2023-04-29 19:19:06,130][139883] Num frames 700... -[2023-04-29 19:19:06,262][139883] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840 -[2023-04-29 19:19:06,263][139883] Avg episode reward: 3.840, avg true_objective: 3.840 -[2023-04-29 19:19:06,308][139883] Num frames 800... -[2023-04-29 19:19:06,440][139883] Num frames 900... -[2023-04-29 19:19:06,579][139883] Num frames 1000... -[2023-04-29 19:19:06,697][139883] Num frames 1100... -[2023-04-29 19:19:06,855][139883] Avg episode rewards: #0: 3.947, true rewards: #0: 3.947 -[2023-04-29 19:19:06,856][139883] Avg episode reward: 3.947, avg true_objective: 3.947 -[2023-04-29 19:19:06,874][139883] Num frames 1200... -[2023-04-29 19:19:06,994][139883] Num frames 1300... -[2023-04-29 19:19:07,103][139883] Num frames 1400... -[2023-04-29 19:19:07,216][139883] Num frames 1500... -[2023-04-29 19:19:07,344][139883] Avg episode rewards: #0: 3.920, true rewards: #0: 3.920 -[2023-04-29 19:19:07,345][139883] Avg episode reward: 3.920, avg true_objective: 3.920 -[2023-04-29 19:19:07,381][139883] Num frames 1600... -[2023-04-29 19:19:07,495][139883] Num frames 1700... -[2023-04-29 19:19:07,612][139883] Num frames 1800... -[2023-04-29 19:19:07,728][139883] Num frames 1900... -[2023-04-29 19:19:07,843][139883] Avg episode rewards: #0: 3.904, true rewards: #0: 3.904 -[2023-04-29 19:19:07,844][139883] Avg episode reward: 3.904, avg true_objective: 3.904 -[2023-04-29 19:19:07,913][139883] Num frames 2000... -[2023-04-29 19:19:08,031][139883] Num frames 2100... -[2023-04-29 19:19:08,140][139883] Num frames 2200... -[2023-04-29 19:19:08,252][139883] Num frames 2300... -[2023-04-29 19:19:08,373][139883] Num frames 2400... -[2023-04-29 19:19:08,531][139883] Avg episode rewards: #0: 4.493, true rewards: #0: 4.160 -[2023-04-29 19:19:08,532][139883] Avg episode reward: 4.493, avg true_objective: 4.160 -[2023-04-29 19:19:08,539][139883] Num frames 2500... -[2023-04-29 19:19:08,658][139883] Num frames 2600... -[2023-04-29 19:19:08,778][139883] Num frames 2700... -[2023-04-29 19:19:08,915][139883] Num frames 2800... -[2023-04-29 19:19:09,064][139883] Avg episode rewards: #0: 4.400, true rewards: #0: 4.114 -[2023-04-29 19:19:09,064][139883] Avg episode reward: 4.400, avg true_objective: 4.114 -[2023-04-29 19:19:09,089][139883] Num frames 2900... -[2023-04-29 19:19:09,216][139883] Num frames 3000... -[2023-04-29 19:19:09,334][139883] Num frames 3100... -[2023-04-29 19:19:09,449][139883] Num frames 3200... -[2023-04-29 19:19:09,580][139883] Num frames 3300... -[2023-04-29 19:19:09,671][139883] Avg episode rewards: #0: 4.535, true rewards: #0: 4.160 -[2023-04-29 19:19:09,671][139883] Avg episode reward: 4.535, avg true_objective: 4.160 -[2023-04-29 19:19:09,763][139883] Num frames 3400... -[2023-04-29 19:19:09,885][139883] Num frames 3500... -[2023-04-29 19:19:10,010][139883] Num frames 3600... -[2023-04-29 19:19:10,129][139883] Num frames 3700... -[2023-04-29 19:19:10,196][139883] Avg episode rewards: #0: 4.458, true rewards: #0: 4.124 -[2023-04-29 19:19:10,197][139883] Avg episode reward: 4.458, avg true_objective: 4.124 -[2023-04-29 19:19:10,307][139883] Num frames 3800... -[2023-04-29 19:19:10,426][139883] Num frames 3900... -[2023-04-29 19:19:10,533][139883] Num frames 4000... -[2023-04-29 19:19:10,685][139883] Avg episode rewards: #0: 4.396, true rewards: #0: 4.096 -[2023-04-29 19:19:10,686][139883] Avg episode reward: 4.396, avg true_objective: 4.096 -[2023-04-29 19:19:15,388][139883] Replay video saved to /home/byron/projects/rl-learning-course/unit-08/train_dir/default_experiment/replay.mp4! +[2023-04-30 12:45:35,350][683074] Using optimizer +[2023-04-30 12:45:35,351][683074] Loading state from checkpoint ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000005_20480.pth... +[2023-04-30 12:45:35,371][683074] Loading model from checkpoint +[2023-04-30 12:45:35,395][683074] Loaded experiment state at self.train_step=5, self.env_steps=20480 +[2023-04-30 12:45:35,426][683074] Initialized policy 0 weights for model version 5 +[2023-04-30 12:45:35,442][683074] LearnerWorker_p0 finished initialization! +[2023-04-30 12:45:36,076][683138] Worker 0 uses CPU cores [0, 1, 2] +[2023-04-30 12:45:36,081][683137] RunningMeanStd input shape: (3, 72, 128) +[2023-04-30 12:45:36,081][683137] RunningMeanStd input shape: (1,) +[2023-04-30 12:45:36,084][683139] Worker 1 uses CPU cores [3, 4, 5] +[2023-04-30 12:45:36,088][683141] Worker 3 uses CPU cores [9, 10, 11] +[2023-04-30 12:45:36,089][683137] ConvEncoder: input_channels=3 +[2023-04-30 12:45:36,103][683149] Worker 7 uses CPU cores [21, 22, 23] +[2023-04-30 12:45:36,109][683144] Worker 5 uses CPU cores [15, 16, 17] +[2023-04-30 12:45:36,115][683145] Worker 6 uses CPU cores [18, 19, 20] +[2023-04-30 12:45:36,118][683142] Worker 4 uses CPU cores [12, 13, 14] +[2023-04-30 12:45:36,119][683140] Worker 2 uses CPU cores [6, 7, 8] +[2023-04-30 12:45:36,217][683137] Conv encoder output size: 512 +[2023-04-30 12:45:36,218][683137] Policy head output size: 512 +[2023-04-30 12:45:36,229][682983] Inference worker 0-0 is ready! +[2023-04-30 12:45:36,229][682983] All inference workers are ready! Signal rollout workers to start! +[2023-04-30 12:45:36,241][683138] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:45:36,242][683145] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:45:36,242][683149] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:45:36,243][683144] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:45:36,243][683141] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:45:36,260][683142] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:45:36,270][683140] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:45:36,276][683139] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-04-30 12:45:36,481][683149] Decorrelating experience for 0 frames... +[2023-04-30 12:45:36,487][683138] Decorrelating experience for 0 frames... +[2023-04-30 12:45:36,497][683144] Decorrelating experience for 0 frames... +[2023-04-30 12:45:36,501][683145] Decorrelating experience for 0 frames... +[2023-04-30 12:45:36,698][683149] Decorrelating experience for 32 frames... +[2023-04-30 12:45:36,704][683139] Decorrelating experience for 0 frames... +[2023-04-30 12:45:36,745][683144] Decorrelating experience for 32 frames... +[2023-04-30 12:45:36,755][683141] Decorrelating experience for 0 frames... +[2023-04-30 12:45:36,758][683140] Decorrelating experience for 0 frames... +[2023-04-30 12:45:36,907][683139] Decorrelating experience for 32 frames... +[2023-04-30 12:45:36,940][683144] Decorrelating experience for 64 frames... +[2023-04-30 12:45:36,940][683141] Decorrelating experience for 32 frames... +[2023-04-30 12:45:36,940][683145] Decorrelating experience for 32 frames... +[2023-04-30 12:45:36,973][683149] Decorrelating experience for 64 frames... +[2023-04-30 12:45:37,117][683140] Decorrelating experience for 32 frames... +[2023-04-30 12:45:37,142][683139] Decorrelating experience for 64 frames... +[2023-04-30 12:45:37,185][683145] Decorrelating experience for 64 frames... +[2023-04-30 12:45:37,195][683144] Decorrelating experience for 96 frames... +[2023-04-30 12:45:37,197][683149] Decorrelating experience for 96 frames... +[2023-04-30 12:45:37,198][683138] Decorrelating experience for 32 frames... +[2023-04-30 12:45:37,335][683140] Decorrelating experience for 64 frames... +[2023-04-30 12:45:37,379][683139] Decorrelating experience for 96 frames... +[2023-04-30 12:45:37,404][683142] Decorrelating experience for 0 frames... +[2023-04-30 12:45:37,431][683145] Decorrelating experience for 96 frames... +[2023-04-30 12:45:37,469][683149] Decorrelating experience for 128 frames... +[2023-04-30 12:45:37,555][683140] Decorrelating experience for 96 frames... +[2023-04-30 12:45:37,555][683138] Decorrelating experience for 64 frames... +[2023-04-30 12:45:37,592][683144] Decorrelating experience for 128 frames... +[2023-04-30 12:45:37,592][683142] Decorrelating experience for 32 frames... +[2023-04-30 12:45:37,781][683149] Decorrelating experience for 160 frames... +[2023-04-30 12:45:37,781][683138] Decorrelating experience for 96 frames... +[2023-04-30 12:45:37,785][683139] Decorrelating experience for 128 frames... +[2023-04-30 12:45:37,786][683145] Decorrelating experience for 128 frames... +[2023-04-30 12:45:37,831][683141] Decorrelating experience for 64 frames... +[2023-04-30 12:45:38,044][682983] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 20480. Throughput: 0: nan. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) +[2023-04-30 12:45:38,049][683144] Decorrelating experience for 160 frames... +[2023-04-30 12:45:38,061][683140] Decorrelating experience for 128 frames... +[2023-04-30 12:45:38,067][683138] Decorrelating experience for 128 frames... +[2023-04-30 12:45:38,088][683145] Decorrelating experience for 160 frames... +[2023-04-30 12:45:38,101][683139] Decorrelating experience for 160 frames... +[2023-04-30 12:45:38,110][683141] Decorrelating experience for 96 frames... +[2023-04-30 12:45:38,271][683142] Decorrelating experience for 64 frames... +[2023-04-30 12:45:38,324][683140] Decorrelating experience for 160 frames... +[2023-04-30 12:45:38,325][683138] Decorrelating experience for 160 frames... +[2023-04-30 12:45:38,371][683144] Decorrelating experience for 192 frames... +[2023-04-30 12:45:38,385][683145] Decorrelating experience for 192 frames... +[2023-04-30 12:45:38,403][683141] Decorrelating experience for 128 frames... +[2023-04-30 12:45:38,504][683142] Decorrelating experience for 96 frames... +[2023-04-30 12:45:38,522][683149] Decorrelating experience for 192 frames... +[2023-04-30 12:45:38,596][683139] Decorrelating experience for 192 frames... +[2023-04-30 12:45:38,615][683138] Decorrelating experience for 192 frames... +[2023-04-30 12:45:38,674][683141] Decorrelating experience for 160 frames... +[2023-04-30 12:45:38,796][683142] Decorrelating experience for 128 frames... +[2023-04-30 12:45:38,815][683149] Decorrelating experience for 224 frames... +[2023-04-30 12:45:38,815][683140] Decorrelating experience for 192 frames... +[2023-04-30 12:45:38,905][683139] Decorrelating experience for 224 frames... +[2023-04-30 12:45:38,979][683141] Decorrelating experience for 192 frames... +[2023-04-30 12:45:39,050][683138] Decorrelating experience for 224 frames... +[2023-04-30 12:45:39,067][683145] Decorrelating experience for 224 frames... +[2023-04-30 12:45:39,095][683144] Decorrelating experience for 224 frames... +[2023-04-30 12:45:39,137][683142] Decorrelating experience for 160 frames... +[2023-04-30 12:45:39,138][683140] Decorrelating experience for 224 frames... +[2023-04-30 12:45:39,308][683141] Decorrelating experience for 224 frames... +[2023-04-30 12:45:39,424][683142] Decorrelating experience for 192 frames... +[2023-04-30 12:45:39,711][683142] Decorrelating experience for 224 frames... +[2023-04-30 12:45:40,485][683074] Signal inference workers to stop experience collection... +[2023-04-30 12:45:40,505][683137] InferenceWorker_p0-w0: stopping experience collection +[2023-04-30 12:45:41,493][683074] Signal inference workers to resume experience collection... +[2023-04-30 12:45:41,494][683137] InferenceWorker_p0-w0: resuming experience collection +[2023-04-30 12:45:43,044][682983] Fps is (10 sec: 819.2, 60 sec: 819.2, 300 sec: 819.2). Total num frames: 24576. Throughput: 0: 784.0. Samples: 3920. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0) +[2023-04-30 12:45:43,045][682983] Avg episode reward: [(0, '1.796')] +[2023-04-30 12:45:48,044][682983] Fps is (10 sec: 2048.0, 60 sec: 2048.0, 300 sec: 2048.0). Total num frames: 40960. Throughput: 0: 616.0. Samples: 6160. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:45:48,044][682983] Avg episode reward: [(0, '3.036')] +[2023-04-30 12:45:53,044][682983] Fps is (10 sec: 3276.8, 60 sec: 2457.6, 300 sec: 2457.6). Total num frames: 57344. Throughput: 0: 685.9. Samples: 10288. Policy #0 lag: (min: 1.0, avg: 1.8, max: 3.0) +[2023-04-30 12:45:53,044][682983] Avg episode reward: [(0, '3.679')] +[2023-04-30 12:45:54,030][683137] Updated weights for policy 0, policy_version 15 (0.1495) +[2023-04-30 12:45:54,076][682983] Heartbeat connected on Batcher_0 +[2023-04-30 12:45:54,082][682983] Heartbeat connected on RolloutWorker_w0 +[2023-04-30 12:45:54,084][682983] Heartbeat connected on RolloutWorker_w1 +[2023-04-30 12:45:54,086][682983] Heartbeat connected on RolloutWorker_w2 +[2023-04-30 12:45:54,087][682983] Heartbeat connected on RolloutWorker_w3 +[2023-04-30 12:45:54,089][682983] Heartbeat connected on RolloutWorker_w4 +[2023-04-30 12:45:54,091][682983] Heartbeat connected on RolloutWorker_w5 +[2023-04-30 12:45:54,092][682983] Heartbeat connected on RolloutWorker_w6 +[2023-04-30 12:45:54,101][682983] Heartbeat connected on InferenceWorker_p0-w0 +[2023-04-30 12:45:54,138][682983] Heartbeat connected on RolloutWorker_w7 +[2023-04-30 12:45:56,743][682983] Heartbeat connected on LearnerWorker_p0 +[2023-04-30 12:45:58,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2457.6, 300 sec: 2457.6). Total num frames: 69632. Throughput: 0: 759.4. Samples: 15188. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:45:58,044][682983] Avg episode reward: [(0, '4.419')] +[2023-04-30 12:46:03,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2621.4, 300 sec: 2621.4). Total num frames: 86016. Throughput: 0: 689.9. Samples: 17248. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:46:03,045][682983] Avg episode reward: [(0, '4.396')] +[2023-04-30 12:46:08,044][682983] Fps is (10 sec: 3276.8, 60 sec: 2730.7, 300 sec: 2730.7). Total num frames: 102400. Throughput: 0: 710.4. Samples: 21312. Policy #0 lag: (min: 1.0, avg: 1.8, max: 3.0) +[2023-04-30 12:46:08,044][682983] Avg episode reward: [(0, '4.337')] +[2023-04-30 12:46:08,115][683137] Updated weights for policy 0, policy_version 25 (0.0839) +[2023-04-30 12:46:09,338][683074] Saving new best policy, reward=4.337! +[2023-04-30 12:46:13,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2691.7, 300 sec: 2691.7). Total num frames: 114688. Throughput: 0: 737.0. Samples: 25796. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:46:13,044][682983] Avg episode reward: [(0, '4.517')] +[2023-04-30 12:46:14,840][683074] Saving new best policy, reward=4.517! +[2023-04-30 12:46:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2764.8, 300 sec: 2764.8). Total num frames: 131072. Throughput: 0: 707.9. Samples: 28316. Policy #0 lag: (min: 1.0, avg: 1.8, max: 3.0) +[2023-04-30 12:46:18,044][682983] Avg episode reward: [(0, '4.532')] +[2023-04-30 12:46:18,990][683074] Saving new best policy, reward=4.532! +[2023-04-30 12:46:21,984][683137] Updated weights for policy 0, policy_version 35 (0.0410) +[2023-04-30 12:46:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2730.7, 300 sec: 2730.7). Total num frames: 143360. Throughput: 0: 729.9. Samples: 32844. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:46:23,045][682983] Avg episode reward: [(0, '4.647')] +[2023-04-30 12:46:24,537][683074] Saving new best policy, reward=4.647! +[2023-04-30 12:46:28,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2785.3, 300 sec: 2785.3). Total num frames: 159744. Throughput: 0: 734.3. Samples: 36964. Policy #0 lag: (min: 1.0, avg: 2.1, max: 4.0) +[2023-04-30 12:46:28,044][682983] Avg episode reward: [(0, '4.552')] +[2023-04-30 12:46:33,044][682983] Fps is (10 sec: 3276.8, 60 sec: 2830.0, 300 sec: 2830.0). Total num frames: 176128. Throughput: 0: 731.1. Samples: 39060. Policy #0 lag: (min: 1.0, avg: 1.8, max: 3.0) +[2023-04-30 12:46:33,044][682983] Avg episode reward: [(0, '4.594')] +[2023-04-30 12:46:35,701][683137] Updated weights for policy 0, policy_version 45 (0.1014) +[2023-04-30 12:46:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2798.9, 300 sec: 2798.9). Total num frames: 188416. Throughput: 0: 746.7. Samples: 43888. Policy #0 lag: (min: 1.0, avg: 2.1, max: 4.0) +[2023-04-30 12:46:38,044][682983] Avg episode reward: [(0, '4.432')] +[2023-04-30 12:46:43,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2835.7). Total num frames: 204800. Throughput: 0: 729.8. Samples: 48028. Policy #0 lag: (min: 1.0, avg: 1.8, max: 3.0) +[2023-04-30 12:46:43,044][682983] Avg episode reward: [(0, '4.388')] +[2023-04-30 12:46:48,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2867.2). Total num frames: 221184. Throughput: 0: 737.7. Samples: 50444. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:46:48,045][682983] Avg episode reward: [(0, '4.381')] +[2023-04-30 12:46:49,451][683137] Updated weights for policy 0, policy_version 55 (0.0613) +[2023-04-30 12:46:49,856][683074] Signal inference workers to stop experience collection... (50 times) +[2023-04-30 12:46:49,880][683137] InferenceWorker_p0-w0: stopping experience collection (50 times) +[2023-04-30 12:46:50,622][683074] Signal inference workers to resume experience collection... (50 times) +[2023-04-30 12:46:50,622][683137] InferenceWorker_p0-w0: resuming experience collection (50 times) +[2023-04-30 12:46:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2839.9). Total num frames: 233472. Throughput: 0: 752.2. Samples: 55160. Policy #0 lag: (min: 1.0, avg: 2.1, max: 4.0) +[2023-04-30 12:46:53,044][682983] Avg episode reward: [(0, '4.521')] +[2023-04-30 12:46:58,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2867.2). Total num frames: 249856. Throughput: 0: 748.9. Samples: 59496. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:46:58,044][682983] Avg episode reward: [(0, '4.531')] +[2023-04-30 12:47:03,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2891.3). Total num frames: 266240. Throughput: 0: 739.8. Samples: 61608. Policy #0 lag: (min: 1.0, avg: 2.1, max: 4.0) +[2023-04-30 12:47:03,044][682983] Avg episode reward: [(0, '4.595')] +[2023-04-30 12:47:03,122][683137] Updated weights for policy 0, policy_version 65 (0.1207) +[2023-04-30 12:47:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2867.2). Total num frames: 278528. Throughput: 0: 748.1. Samples: 66508. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:47:08,044][682983] Avg episode reward: [(0, '4.550')] +[2023-04-30 12:47:13,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2888.8). Total num frames: 294912. Throughput: 0: 748.4. Samples: 70644. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:47:13,044][682983] Avg episode reward: [(0, '4.613')] +[2023-04-30 12:47:16,479][683137] Updated weights for policy 0, policy_version 75 (0.0409) +[2023-04-30 12:47:18,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2908.2). Total num frames: 311296. Throughput: 0: 753.7. Samples: 72976. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:47:18,044][682983] Avg episode reward: [(0, '4.541')] +[2023-04-30 12:47:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2886.7). Total num frames: 323584. Throughput: 0: 757.2. Samples: 77960. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:47:23,044][682983] Avg episode reward: [(0, '4.580')] +[2023-04-30 12:47:28,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2904.4). Total num frames: 339968. Throughput: 0: 756.4. Samples: 82068. Policy #0 lag: (min: 1.0, avg: 2.2, max: 4.0) +[2023-04-30 12:47:28,044][682983] Avg episode reward: [(0, '4.510')] +[2023-04-30 12:47:30,039][683137] Updated weights for policy 0, policy_version 85 (0.0411) +[2023-04-30 12:47:33,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2920.6). Total num frames: 356352. Throughput: 0: 758.0. Samples: 84552. Policy #0 lag: (min: 1.0, avg: 2.2, max: 4.0) +[2023-04-30 12:47:33,044][682983] Avg episode reward: [(0, '4.608')] +[2023-04-30 12:47:34,084][683074] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000088_360448.pth... +[2023-04-30 12:47:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2901.3). Total num frames: 368640. Throughput: 0: 751.5. Samples: 88976. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:47:38,045][682983] Avg episode reward: [(0, '4.517')] +[2023-04-30 12:47:43,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2916.4). Total num frames: 385024. Throughput: 0: 745.7. Samples: 93052. Policy #0 lag: (min: 1.0, avg: 2.1, max: 4.0) +[2023-04-30 12:47:43,044][682983] Avg episode reward: [(0, '4.462')] +[2023-04-30 12:47:43,749][683137] Updated weights for policy 0, policy_version 95 (0.0606) +[2023-04-30 12:47:48,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2930.2). Total num frames: 401408. Throughput: 0: 751.3. Samples: 95416. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:47:48,044][682983] Avg episode reward: [(0, '4.606')] +[2023-04-30 12:47:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2912.7). Total num frames: 413696. Throughput: 0: 754.9. Samples: 100480. Policy #0 lag: (min: 1.0, avg: 2.1, max: 4.0) +[2023-04-30 12:47:53,044][682983] Avg episode reward: [(0, '4.601')] +[2023-04-30 12:47:57,422][683137] Updated weights for policy 0, policy_version 105 (0.0408) +[2023-04-30 12:47:57,749][683074] Signal inference workers to stop experience collection... (100 times) +[2023-04-30 12:47:57,772][683137] InferenceWorker_p0-w0: stopping experience collection (100 times) +[2023-04-30 12:47:58,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2925.7). Total num frames: 430080. Throughput: 0: 754.7. Samples: 104604. Policy #0 lag: (min: 1.0, avg: 2.1, max: 4.0) +[2023-04-30 12:47:58,044][682983] Avg episode reward: [(0, '4.597')] +[2023-04-30 12:47:58,760][683074] Signal inference workers to resume experience collection... (100 times) +[2023-04-30 12:47:58,761][683137] InferenceWorker_p0-w0: resuming experience collection (100 times) +[2023-04-30 12:48:03,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2937.8). Total num frames: 446464. Throughput: 0: 750.0. Samples: 106724. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:48:03,044][682983] Avg episode reward: [(0, '4.543')] +[2023-04-30 12:48:08,044][682983] Fps is (10 sec: 2867.1, 60 sec: 3003.7, 300 sec: 2921.8). Total num frames: 458752. Throughput: 0: 747.2. Samples: 111584. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:48:08,045][682983] Avg episode reward: [(0, '4.516')] +[2023-04-30 12:48:11,302][683137] Updated weights for policy 0, policy_version 115 (0.0814) +[2023-04-30 12:48:13,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2933.3). Total num frames: 475136. Throughput: 0: 746.0. Samples: 115636. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:48:13,044][682983] Avg episode reward: [(0, '4.398')] +[2023-04-30 12:48:18,044][682983] Fps is (10 sec: 3276.9, 60 sec: 3003.7, 300 sec: 2944.0). Total num frames: 491520. Throughput: 0: 736.6. Samples: 117700. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:48:18,044][682983] Avg episode reward: [(0, '4.413')] +[2023-04-30 12:48:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2929.3). Total num frames: 503808. Throughput: 0: 754.4. Samples: 122924. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:48:23,044][682983] Avg episode reward: [(0, '4.453')] +[2023-04-30 12:48:24,751][683137] Updated weights for policy 0, policy_version 125 (0.0214) +[2023-04-30 12:48:28,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2939.5). Total num frames: 520192. Throughput: 0: 755.1. Samples: 127032. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:48:28,044][682983] Avg episode reward: [(0, '4.499')] +[2023-04-30 12:48:33,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2949.1). Total num frames: 536576. Throughput: 0: 752.6. Samples: 129284. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:48:33,044][682983] Avg episode reward: [(0, '4.464')] +[2023-04-30 12:48:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2935.5). Total num frames: 548864. Throughput: 0: 744.0. Samples: 133960. Policy #0 lag: (min: 1.0, avg: 1.8, max: 3.0) +[2023-04-30 12:48:38,044][682983] Avg episode reward: [(0, '4.435')] +[2023-04-30 12:48:38,317][683137] Updated weights for policy 0, policy_version 135 (0.1005) +[2023-04-30 12:48:43,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2944.7). Total num frames: 565248. Throughput: 0: 744.3. Samples: 138096. Policy #0 lag: (min: 1.0, avg: 1.8, max: 3.0) +[2023-04-30 12:48:43,044][682983] Avg episode reward: [(0, '4.505')] +[2023-04-30 12:48:48,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2953.4). Total num frames: 581632. Throughput: 0: 748.4. Samples: 140404. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:48:48,044][682983] Avg episode reward: [(0, '4.554')] +[2023-04-30 12:48:52,091][683137] Updated weights for policy 0, policy_version 145 (0.0613) +[2023-04-30 12:48:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2940.7). Total num frames: 593920. Throughput: 0: 753.0. Samples: 145468. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:48:53,044][682983] Avg episode reward: [(0, '4.515')] +[2023-04-30 12:48:58,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2949.1). Total num frames: 610304. Throughput: 0: 754.0. Samples: 149564. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:48:58,045][682983] Avg episode reward: [(0, '4.678')] +[2023-04-30 12:49:00,070][683074] Saving new best policy, reward=4.678! +[2023-04-30 12:49:03,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2957.1). Total num frames: 626688. Throughput: 0: 763.2. Samples: 152044. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:49:03,045][682983] Avg episode reward: [(0, '4.617')] +[2023-04-30 12:49:05,532][683137] Updated weights for policy 0, policy_version 155 (0.0806) +[2023-04-30 12:49:05,871][683074] Signal inference workers to stop experience collection... (150 times) +[2023-04-30 12:49:05,892][683137] InferenceWorker_p0-w0: stopping experience collection (150 times) +[2023-04-30 12:49:06,865][683074] Signal inference workers to resume experience collection... (150 times) +[2023-04-30 12:49:06,865][683137] InferenceWorker_p0-w0: resuming experience collection (150 times) +[2023-04-30 12:49:08,044][682983] Fps is (10 sec: 2867.3, 60 sec: 3003.7, 300 sec: 2945.2). Total num frames: 638976. Throughput: 0: 747.5. Samples: 156560. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:49:08,044][682983] Avg episode reward: [(0, '4.339')] +[2023-04-30 12:49:13,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2952.9). Total num frames: 655360. Throughput: 0: 747.6. Samples: 160676. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:49:13,044][682983] Avg episode reward: [(0, '4.342')] +[2023-04-30 12:49:18,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2960.3). Total num frames: 671744. Throughput: 0: 748.9. Samples: 162984. Policy #0 lag: (min: 1.0, avg: 1.9, max: 3.0) +[2023-04-30 12:49:18,045][682983] Avg episode reward: [(0, '4.469')] +[2023-04-30 12:49:19,236][683137] Updated weights for policy 0, policy_version 165 (0.0610) +[2023-04-30 12:49:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2949.1). Total num frames: 684032. Throughput: 0: 754.4. Samples: 167908. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:49:23,044][682983] Avg episode reward: [(0, '4.478')] +[2023-04-30 12:49:28,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2956.2). Total num frames: 700416. Throughput: 0: 754.9. Samples: 172068. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:49:28,044][682983] Avg episode reward: [(0, '4.447')] +[2023-04-30 12:49:33,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2963.1). Total num frames: 716800. Throughput: 0: 755.5. Samples: 174400. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:49:33,045][682983] Avg episode reward: [(0, '4.556')] +[2023-04-30 12:49:33,045][683137] Updated weights for policy 0, policy_version 175 (0.1202) +[2023-04-30 12:49:34,206][683074] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000176_720896.pth... +[2023-04-30 12:49:34,229][683074] Removing ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000005_20480.pth +[2023-04-30 12:49:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2952.5). Total num frames: 729088. Throughput: 0: 750.7. Samples: 179248. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:49:38,044][682983] Avg episode reward: [(0, '4.596')] +[2023-04-30 12:49:43,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2959.2). Total num frames: 745472. Throughput: 0: 751.1. Samples: 183364. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:49:43,044][682983] Avg episode reward: [(0, '4.718')] +[2023-04-30 12:49:45,166][683074] Saving new best policy, reward=4.718! +[2023-04-30 12:49:46,754][683137] Updated weights for policy 0, policy_version 185 (0.0422) +[2023-04-30 12:49:48,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2965.5). Total num frames: 761856. Throughput: 0: 744.6. Samples: 185552. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:49:48,044][682983] Avg episode reward: [(0, '4.742')] +[2023-04-30 12:49:49,242][683074] Saving new best policy, reward=4.742! +[2023-04-30 12:49:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2955.5). Total num frames: 774144. Throughput: 0: 753.5. Samples: 190468. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:49:53,044][682983] Avg episode reward: [(0, '4.708')] +[2023-04-30 12:49:58,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2961.7). Total num frames: 790528. Throughput: 0: 754.8. Samples: 194640. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:49:58,044][682983] Avg episode reward: [(0, '4.632')] +[2023-04-30 12:50:00,153][683137] Updated weights for policy 0, policy_version 195 (0.0414) +[2023-04-30 12:50:03,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2967.7). Total num frames: 806912. Throughput: 0: 753.2. Samples: 196876. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:50:03,044][682983] Avg episode reward: [(0, '4.682')] +[2023-04-30 12:50:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2958.2). Total num frames: 819200. Throughput: 0: 753.2. Samples: 201800. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:50:08,044][682983] Avg episode reward: [(0, '4.626')] +[2023-04-30 12:50:13,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2964.0). Total num frames: 835584. Throughput: 0: 751.4. Samples: 205880. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:50:13,044][682983] Avg episode reward: [(0, '4.827')] +[2023-04-30 12:50:13,785][683137] Updated weights for policy 0, policy_version 205 (0.0413) +[2023-04-30 12:50:14,272][683074] Signal inference workers to stop experience collection... (200 times) +[2023-04-30 12:50:14,295][683137] InferenceWorker_p0-w0: stopping experience collection (200 times) +[2023-04-30 12:50:15,184][683074] Signal inference workers to resume experience collection... (200 times) +[2023-04-30 12:50:15,184][683074] Saving new best policy, reward=4.827! +[2023-04-30 12:50:15,185][683137] InferenceWorker_p0-w0: resuming experience collection (200 times) +[2023-04-30 12:50:18,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2969.6). Total num frames: 851968. Throughput: 0: 748.1. Samples: 208064. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:50:18,044][682983] Avg episode reward: [(0, '4.745')] +[2023-04-30 12:50:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2960.6). Total num frames: 864256. Throughput: 0: 744.4. Samples: 212744. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:50:23,044][682983] Avg episode reward: [(0, '4.879')] +[2023-04-30 12:50:24,786][683074] Saving new best policy, reward=4.879! +[2023-04-30 12:50:27,549][683137] Updated weights for policy 0, policy_version 215 (0.0813) +[2023-04-30 12:50:28,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2966.1). Total num frames: 880640. Throughput: 0: 753.0. Samples: 217248. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:50:28,044][682983] Avg episode reward: [(0, '4.850')] +[2023-04-30 12:50:33,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2971.3). Total num frames: 897024. Throughput: 0: 749.5. Samples: 219280. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:50:33,044][682983] Avg episode reward: [(0, '4.836')] +[2023-04-30 12:50:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 909312. Throughput: 0: 750.9. Samples: 224260. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:50:38,044][682983] Avg episode reward: [(0, '5.016')] +[2023-04-30 12:50:39,740][683074] Saving new best policy, reward=5.016! +[2023-04-30 12:50:41,149][683137] Updated weights for policy 0, policy_version 225 (0.0609) +[2023-04-30 12:50:43,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 925696. Throughput: 0: 749.5. Samples: 228368. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:50:43,044][682983] Avg episode reward: [(0, '4.806')] +[2023-04-30 12:50:48,044][682983] Fps is (10 sec: 3276.7, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 942080. Throughput: 0: 745.7. Samples: 230432. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:50:48,045][682983] Avg episode reward: [(0, '4.703')] +[2023-04-30 12:50:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 954368. Throughput: 0: 741.8. Samples: 235180. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:50:53,044][682983] Avg episode reward: [(0, '4.671')] +[2023-04-30 12:50:54,910][683137] Updated weights for policy 0, policy_version 235 (0.0817) +[2023-04-30 12:50:58,044][682983] Fps is (10 sec: 2867.3, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 970752. Throughput: 0: 751.2. Samples: 239684. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:50:58,044][682983] Avg episode reward: [(0, '4.748')] +[2023-04-30 12:51:03,044][682983] Fps is (10 sec: 3276.7, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 987136. Throughput: 0: 747.6. Samples: 241704. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:51:03,045][682983] Avg episode reward: [(0, '4.837')] +[2023-04-30 12:51:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 999424. Throughput: 0: 753.6. Samples: 246656. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:51:08,044][682983] Avg episode reward: [(0, '4.646')] +[2023-04-30 12:51:08,584][683137] Updated weights for policy 0, policy_version 245 (0.1015) +[2023-04-30 12:51:13,044][682983] Fps is (10 sec: 2867.3, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1015808. Throughput: 0: 747.6. Samples: 250888. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:51:13,044][682983] Avg episode reward: [(0, '4.586')] +[2023-04-30 12:51:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2999.1). Total num frames: 1028096. Throughput: 0: 747.6. Samples: 252924. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:51:18,044][682983] Avg episode reward: [(0, '4.674')] +[2023-04-30 12:51:22,505][683137] Updated weights for policy 0, policy_version 255 (0.1620) +[2023-04-30 12:51:22,914][683074] Signal inference workers to stop experience collection... (250 times) +[2023-04-30 12:51:22,934][683137] InferenceWorker_p0-w0: stopping experience collection (250 times) +[2023-04-30 12:51:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1044480. Throughput: 0: 735.7. Samples: 257368. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:51:23,044][682983] Avg episode reward: [(0, '4.714')] +[2023-04-30 12:51:23,675][683074] Signal inference workers to resume experience collection... (250 times) +[2023-04-30 12:51:23,675][683137] InferenceWorker_p0-w0: resuming experience collection (250 times) +[2023-04-30 12:51:28,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1060864. Throughput: 0: 751.4. Samples: 262180. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:51:28,044][682983] Avg episode reward: [(0, '4.826')] +[2023-04-30 12:51:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2999.1). Total num frames: 1073152. Throughput: 0: 750.2. Samples: 264192. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:51:33,044][682983] Avg episode reward: [(0, '4.848')] +[2023-04-30 12:51:34,628][683074] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000264_1081344.pth... +[2023-04-30 12:51:34,652][683074] Removing ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000088_360448.pth +[2023-04-30 12:51:36,018][683137] Updated weights for policy 0, policy_version 265 (0.1428) +[2023-04-30 12:51:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1089536. Throughput: 0: 748.1. Samples: 268844. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:51:38,044][682983] Avg episode reward: [(0, '4.826')] +[2023-04-30 12:51:43,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1105920. Throughput: 0: 750.0. Samples: 273436. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:51:43,044][682983] Avg episode reward: [(0, '4.918')] +[2023-04-30 12:51:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2999.1). Total num frames: 1118208. Throughput: 0: 751.0. Samples: 275500. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:51:48,045][682983] Avg episode reward: [(0, '5.068')] +[2023-04-30 12:51:49,648][683074] Saving new best policy, reward=5.068! +[2023-04-30 12:51:49,650][683137] Updated weights for policy 0, policy_version 275 (0.0210) +[2023-04-30 12:51:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1134592. Throughput: 0: 744.5. Samples: 280160. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:51:53,044][682983] Avg episode reward: [(0, '5.125')] +[2023-04-30 12:51:55,065][683074] Saving new best policy, reward=5.125! +[2023-04-30 12:51:58,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1150976. Throughput: 0: 751.6. Samples: 284712. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:51:58,045][682983] Avg episode reward: [(0, '5.239')] +[2023-04-30 12:51:59,156][683074] Saving new best policy, reward=5.239! +[2023-04-30 12:52:03,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2999.1). Total num frames: 1163264. Throughput: 0: 752.4. Samples: 286780. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:52:03,045][682983] Avg episode reward: [(0, '5.155')] +[2023-04-30 12:52:03,447][683137] Updated weights for policy 0, policy_version 285 (0.0602) +[2023-04-30 12:52:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1179648. Throughput: 0: 748.1. Samples: 291032. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:52:08,044][682983] Avg episode reward: [(0, '5.415')] +[2023-04-30 12:52:10,023][683074] Saving new best policy, reward=5.415! +[2023-04-30 12:52:13,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1196032. Throughput: 0: 751.9. Samples: 296016. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:52:13,044][682983] Avg episode reward: [(0, '5.217')] +[2023-04-30 12:52:16,848][683137] Updated weights for policy 0, policy_version 295 (0.0811) +[2023-04-30 12:52:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1208320. Throughput: 0: 753.1. Samples: 298080. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:52:18,044][682983] Avg episode reward: [(0, '5.206')] +[2023-04-30 12:52:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1224704. Throughput: 0: 746.6. Samples: 302440. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:52:23,044][682983] Avg episode reward: [(0, '5.253')] +[2023-04-30 12:52:28,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1241088. Throughput: 0: 750.8. Samples: 307224. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:52:28,044][682983] Avg episode reward: [(0, '5.301')] +[2023-04-30 12:52:30,677][683137] Updated weights for policy 0, policy_version 305 (0.0817) +[2023-04-30 12:52:31,162][683074] Signal inference workers to stop experience collection... (300 times) +[2023-04-30 12:52:31,182][683137] InferenceWorker_p0-w0: stopping experience collection (300 times) +[2023-04-30 12:52:31,828][683074] Signal inference workers to resume experience collection... (300 times) +[2023-04-30 12:52:31,828][683137] InferenceWorker_p0-w0: resuming experience collection (300 times) +[2023-04-30 12:52:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1253376. Throughput: 0: 750.2. Samples: 309260. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:52:33,044][682983] Avg episode reward: [(0, '5.249')] +[2023-04-30 12:52:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1269760. Throughput: 0: 754.3. Samples: 314104. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:52:38,044][682983] Avg episode reward: [(0, '5.004')] +[2023-04-30 12:52:43,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1286144. Throughput: 0: 753.2. Samples: 318604. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:52:43,044][682983] Avg episode reward: [(0, '5.045')] +[2023-04-30 12:52:44,046][683137] Updated weights for policy 0, policy_version 315 (0.0601) +[2023-04-30 12:52:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1298432. Throughput: 0: 752.8. Samples: 320656. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:52:48,044][682983] Avg episode reward: [(0, '5.310')] +[2023-04-30 12:52:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1314816. Throughput: 0: 768.7. Samples: 325624. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:52:53,044][682983] Avg episode reward: [(0, '5.363')] +[2023-04-30 12:52:57,551][683137] Updated weights for policy 0, policy_version 325 (0.0429) +[2023-04-30 12:52:58,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1331200. Throughput: 0: 748.9. Samples: 329716. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:52:58,044][682983] Avg episode reward: [(0, '5.291')] +[2023-04-30 12:53:03,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1343488. Throughput: 0: 748.3. Samples: 331752. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:53:03,044][682983] Avg episode reward: [(0, '5.053')] +[2023-04-30 12:53:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1359872. Throughput: 0: 763.5. Samples: 336796. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:53:08,044][682983] Avg episode reward: [(0, '4.972')] +[2023-04-30 12:53:11,249][683137] Updated weights for policy 0, policy_version 335 (0.0454) +[2023-04-30 12:53:13,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1376256. Throughput: 0: 755.7. Samples: 341232. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:53:13,044][682983] Avg episode reward: [(0, '4.880')] +[2023-04-30 12:53:18,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3072.0, 300 sec: 3013.0). Total num frames: 1392640. Throughput: 0: 756.0. Samples: 343280. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:53:18,044][682983] Avg episode reward: [(0, '4.801')] +[2023-04-30 12:53:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1404928. Throughput: 0: 752.9. Samples: 347984. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:53:23,044][682983] Avg episode reward: [(0, '5.042')] +[2023-04-30 12:53:24,765][683137] Updated weights for policy 0, policy_version 345 (0.0415) +[2023-04-30 12:53:28,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1421312. Throughput: 0: 743.2. Samples: 352048. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:53:28,044][682983] Avg episode reward: [(0, '5.282')] +[2023-04-30 12:53:33,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3072.0, 300 sec: 3013.0). Total num frames: 1437696. Throughput: 0: 742.1. Samples: 354052. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:53:33,044][682983] Avg episode reward: [(0, '5.448')] +[2023-04-30 12:53:34,319][683074] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000352_1441792.pth... +[2023-04-30 12:53:34,341][683074] Removing ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000176_720896.pth +[2023-04-30 12:53:34,344][683074] Saving new best policy, reward=5.448! +[2023-04-30 12:53:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1449984. Throughput: 0: 747.9. Samples: 359280. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:53:38,044][682983] Avg episode reward: [(0, '5.482')] +[2023-04-30 12:53:38,485][683137] Updated weights for policy 0, policy_version 355 (0.0409) +[2023-04-30 12:53:39,146][683074] Signal inference workers to stop experience collection... (350 times) +[2023-04-30 12:53:39,165][683137] InferenceWorker_p0-w0: stopping experience collection (350 times) +[2023-04-30 12:53:39,866][683074] Signal inference workers to resume experience collection... (350 times) +[2023-04-30 12:53:39,866][683074] Saving new best policy, reward=5.482! +[2023-04-30 12:53:39,866][683137] InferenceWorker_p0-w0: resuming experience collection (350 times) +[2023-04-30 12:53:43,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1466368. Throughput: 0: 757.7. Samples: 363812. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:53:43,045][682983] Avg episode reward: [(0, '5.618')] +[2023-04-30 12:53:45,355][683074] Saving new best policy, reward=5.618! +[2023-04-30 12:53:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1478656. Throughput: 0: 757.0. Samples: 365816. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:53:48,044][682983] Avg episode reward: [(0, '5.825')] +[2023-04-30 12:53:49,444][683074] Saving new best policy, reward=5.825! +[2023-04-30 12:53:52,366][683137] Updated weights for policy 0, policy_version 365 (0.1196) +[2023-04-30 12:53:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1495040. Throughput: 0: 745.8. Samples: 370356. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:53:53,045][682983] Avg episode reward: [(0, '5.970')] +[2023-04-30 12:53:54,847][683074] Saving new best policy, reward=5.970! +[2023-04-30 12:53:58,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1511424. Throughput: 0: 741.2. Samples: 374588. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:53:58,044][682983] Avg episode reward: [(0, '5.866')] +[2023-04-30 12:54:03,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3072.0, 300 sec: 3013.0). Total num frames: 1527808. Throughput: 0: 741.8. Samples: 376660. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:54:03,045][682983] Avg episode reward: [(0, '6.063')] +[2023-04-30 12:54:04,424][683074] Saving new best policy, reward=6.063! +[2023-04-30 12:54:05,786][683137] Updated weights for policy 0, policy_version 375 (0.0217) +[2023-04-30 12:54:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1540096. Throughput: 0: 751.2. Samples: 381788. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:54:08,044][682983] Avg episode reward: [(0, '5.935')] +[2023-04-30 12:54:13,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1556480. Throughput: 0: 763.1. Samples: 386388. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:54:13,044][682983] Avg episode reward: [(0, '6.127')] +[2023-04-30 12:54:15,286][683074] Saving new best policy, reward=6.127! +[2023-04-30 12:54:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2999.1). Total num frames: 1568768. Throughput: 0: 764.4. Samples: 388448. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:54:18,044][682983] Avg episode reward: [(0, '6.191')] +[2023-04-30 12:54:19,419][683074] Saving new best policy, reward=6.191! +[2023-04-30 12:54:19,421][683137] Updated weights for policy 0, policy_version 385 (0.0608) +[2023-04-30 12:54:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1585152. Throughput: 0: 749.6. Samples: 393012. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:54:23,044][682983] Avg episode reward: [(0, '6.146')] +[2023-04-30 12:54:28,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1601536. Throughput: 0: 741.1. Samples: 397160. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:54:28,044][682983] Avg episode reward: [(0, '6.405')] +[2023-04-30 12:54:28,989][683074] Saving new best policy, reward=6.405! +[2023-04-30 12:54:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2999.1). Total num frames: 1613824. Throughput: 0: 741.6. Samples: 399188. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:54:33,044][682983] Avg episode reward: [(0, '6.330')] +[2023-04-30 12:54:33,079][683137] Updated weights for policy 0, policy_version 395 (0.1200) +[2023-04-30 12:54:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1630208. Throughput: 0: 753.6. Samples: 404268. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:54:38,044][682983] Avg episode reward: [(0, '6.762')] +[2023-04-30 12:54:39,881][683074] Saving new best policy, reward=6.762! +[2023-04-30 12:54:43,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1646592. Throughput: 0: 763.1. Samples: 408928. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:54:43,045][682983] Avg episode reward: [(0, '6.827')] +[2023-04-30 12:54:44,066][683074] Saving new best policy, reward=6.827! +[2023-04-30 12:54:46,847][683137] Updated weights for policy 0, policy_version 405 (0.0817) +[2023-04-30 12:54:47,522][683074] Signal inference workers to stop experience collection... (400 times) +[2023-04-30 12:54:47,542][683137] InferenceWorker_p0-w0: stopping experience collection (400 times) +[2023-04-30 12:54:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1658880. Throughput: 0: 763.1. Samples: 411000. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:54:48,044][682983] Avg episode reward: [(0, '7.411')] +[2023-04-30 12:54:48,229][683074] Signal inference workers to resume experience collection... (400 times) +[2023-04-30 12:54:48,229][683137] InferenceWorker_p0-w0: resuming experience collection (400 times) +[2023-04-30 12:54:49,578][683074] Saving new best policy, reward=7.411! +[2023-04-30 12:54:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1675264. Throughput: 0: 744.5. Samples: 415292. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:54:53,044][682983] Avg episode reward: [(0, '7.622')] +[2023-04-30 12:54:53,699][683074] Saving new best policy, reward=7.622! +[2023-04-30 12:54:58,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1691648. Throughput: 0: 740.4. Samples: 419704. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 12:54:58,044][682983] Avg episode reward: [(0, '7.770')] +[2023-04-30 12:54:59,139][683074] Saving new best policy, reward=7.770! +[2023-04-30 12:55:00,740][683137] Updated weights for policy 0, policy_version 415 (0.0822) +[2023-04-30 12:55:03,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2999.1). Total num frames: 1703936. Throughput: 0: 740.4. Samples: 421768. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:55:03,044][682983] Avg episode reward: [(0, '7.646')] +[2023-04-30 12:55:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1720320. Throughput: 0: 738.8. Samples: 426256. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:55:08,044][682983] Avg episode reward: [(0, '7.578')] +[2023-04-30 12:55:13,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1736704. Throughput: 0: 750.7. Samples: 430940. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 12:55:13,044][682983] Avg episode reward: [(0, '8.023')] +[2023-04-30 12:55:14,336][683074] Saving new best policy, reward=8.023! +[2023-04-30 12:55:14,338][683137] Updated weights for policy 0, policy_version 425 (0.0821) +[2023-04-30 12:55:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1748992. Throughput: 0: 762.8. Samples: 433516. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:55:18,044][682983] Avg episode reward: [(0, '8.136')] +[2023-04-30 12:55:19,826][683074] Saving new best policy, reward=8.136! +[2023-04-30 12:55:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1765376. Throughput: 0: 743.1. Samples: 437708. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 12:55:23,044][682983] Avg episode reward: [(0, '8.116')] +[2023-04-30 12:55:27,989][683137] Updated weights for policy 0, policy_version 435 (0.0813) +[2023-04-30 12:55:28,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1781760. Throughput: 0: 739.4. Samples: 442200. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 12:55:28,044][682983] Avg episode reward: [(0, '8.541')] +[2023-04-30 12:55:29,363][683074] Saving new best policy, reward=8.541! +[2023-04-30 12:55:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1794048. Throughput: 0: 738.5. Samples: 444232. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:55:33,044][682983] Avg episode reward: [(0, '8.531')] +[2023-04-30 12:55:34,902][683074] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000440_1802240.pth... +[2023-04-30 12:55:34,926][683074] Removing ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000264_1081344.pth +[2023-04-30 12:55:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1810432. Throughput: 0: 734.3. Samples: 448336. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:55:38,044][682983] Avg episode reward: [(0, '8.741')] +[2023-04-30 12:55:38,965][683074] Saving new best policy, reward=8.741! +[2023-04-30 12:55:41,878][683137] Updated weights for policy 0, policy_version 445 (0.0410) +[2023-04-30 12:55:43,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2985.2). Total num frames: 1822720. Throughput: 0: 745.9. Samples: 453268. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:55:43,044][682983] Avg episode reward: [(0, '9.066')] +[2023-04-30 12:55:44,418][683074] Saving new best policy, reward=9.066! +[2023-04-30 12:55:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1839104. Throughput: 0: 761.9. Samples: 456052. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:55:48,044][682983] Avg episode reward: [(0, '8.074')] +[2023-04-30 12:55:53,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1855488. Throughput: 0: 753.9. Samples: 460180. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:55:53,044][682983] Avg episode reward: [(0, '8.291')] +[2023-04-30 12:55:55,516][683137] Updated weights for policy 0, policy_version 455 (0.1209) +[2023-04-30 12:55:56,165][683074] Signal inference workers to stop experience collection... (450 times) +[2023-04-30 12:55:56,185][683137] InferenceWorker_p0-w0: stopping experience collection (450 times) +[2023-04-30 12:55:56,699][683074] Signal inference workers to resume experience collection... (450 times) +[2023-04-30 12:55:56,700][683137] InferenceWorker_p0-w0: resuming experience collection (450 times) +[2023-04-30 12:55:58,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2985.2). Total num frames: 1867776. Throughput: 0: 750.2. Samples: 464700. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:55:58,044][682983] Avg episode reward: [(0, '8.285')] +[2023-04-30 12:56:03,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1884160. Throughput: 0: 738.3. Samples: 466740. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:56:03,044][682983] Avg episode reward: [(0, '8.085')] +[2023-04-30 12:56:08,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1900544. Throughput: 0: 736.5. Samples: 470852. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:56:08,044][682983] Avg episode reward: [(0, '8.204')] +[2023-04-30 12:56:09,097][683137] Updated weights for policy 0, policy_version 465 (0.0821) +[2023-04-30 12:56:13,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2999.1). Total num frames: 1912832. Throughput: 0: 747.8. Samples: 475852. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:56:13,044][682983] Avg episode reward: [(0, '8.138')] +[2023-04-30 12:56:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1929216. Throughput: 0: 762.1. Samples: 478528. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:56:18,044][682983] Avg episode reward: [(0, '8.551')] +[2023-04-30 12:56:22,620][683137] Updated weights for policy 0, policy_version 475 (0.0607) +[2023-04-30 12:56:23,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1945600. Throughput: 0: 763.7. Samples: 482704. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:56:23,044][682983] Avg episode reward: [(0, '9.012')] +[2023-04-30 12:56:28,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 3013.0). Total num frames: 1961984. Throughput: 0: 754.6. Samples: 487224. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:56:28,045][682983] Avg episode reward: [(0, '8.989')] +[2023-04-30 12:56:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1974272. Throughput: 0: 738.4. Samples: 489280. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:56:33,044][682983] Avg episode reward: [(0, '9.311')] +[2023-04-30 12:56:34,913][683074] Saving new best policy, reward=9.311! +[2023-04-30 12:56:36,299][683137] Updated weights for policy 0, policy_version 485 (0.0218) +[2023-04-30 12:56:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 1990656. Throughput: 0: 738.8. Samples: 493424. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:56:38,044][682983] Avg episode reward: [(0, '9.671')] +[2023-04-30 12:56:40,396][683074] Saving new best policy, reward=9.671! +[2023-04-30 12:56:43,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2002944. Throughput: 0: 740.4. Samples: 498020. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:56:43,044][682983] Avg episode reward: [(0, '9.790')] +[2023-04-30 12:56:44,571][683074] Saving new best policy, reward=9.790! +[2023-04-30 12:56:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2019328. Throughput: 0: 763.2. Samples: 501084. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:56:48,045][682983] Avg episode reward: [(0, '9.728')] +[2023-04-30 12:56:50,079][683137] Updated weights for policy 0, policy_version 495 (0.0616) +[2023-04-30 12:56:53,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2035712. Throughput: 0: 762.8. Samples: 505180. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 12:56:53,044][682983] Avg episode reward: [(0, '9.638')] +[2023-04-30 12:56:58,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2048000. Throughput: 0: 755.7. Samples: 509860. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:56:58,044][682983] Avg episode reward: [(0, '9.757')] +[2023-04-30 12:57:03,044][682983] Fps is (10 sec: 2867.1, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2064384. Throughput: 0: 741.5. Samples: 511896. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:57:03,045][682983] Avg episode reward: [(0, '10.167')] +[2023-04-30 12:57:03,703][683137] Updated weights for policy 0, policy_version 505 (0.0016) +[2023-04-30 12:57:04,333][683074] Signal inference workers to stop experience collection... (500 times) +[2023-04-30 12:57:04,354][683137] InferenceWorker_p0-w0: stopping experience collection (500 times) +[2023-04-30 12:57:05,066][683074] Signal inference workers to resume experience collection... (500 times) +[2023-04-30 12:57:05,066][683074] Saving new best policy, reward=10.167! +[2023-04-30 12:57:05,066][683137] InferenceWorker_p0-w0: resuming experience collection (500 times) +[2023-04-30 12:57:08,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2080768. Throughput: 0: 739.4. Samples: 515976. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 12:57:08,044][682983] Avg episode reward: [(0, '10.532')] +[2023-04-30 12:57:09,183][683074] Saving new best policy, reward=10.532! +[2023-04-30 12:57:13,044][682983] Fps is (10 sec: 2867.3, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2093056. Throughput: 0: 745.6. Samples: 520776. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:57:13,044][682983] Avg episode reward: [(0, '11.026')] +[2023-04-30 12:57:14,636][683074] Saving new best policy, reward=11.026! +[2023-04-30 12:57:17,352][683137] Updated weights for policy 0, policy_version 515 (0.0614) +[2023-04-30 12:57:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2109440. Throughput: 0: 759.0. Samples: 523436. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 12:57:18,045][682983] Avg episode reward: [(0, '11.371')] +[2023-04-30 12:57:20,086][683074] Saving new best policy, reward=11.371! +[2023-04-30 12:57:23,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2125824. Throughput: 0: 762.8. Samples: 527752. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 12:57:23,044][682983] Avg episode reward: [(0, '11.179')] +[2023-04-30 12:57:28,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2999.1). Total num frames: 2138112. Throughput: 0: 758.1. Samples: 532136. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:57:28,044][682983] Avg episode reward: [(0, '10.791')] +[2023-04-30 12:57:31,020][683137] Updated weights for policy 0, policy_version 525 (0.0611) +[2023-04-30 12:57:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2154496. Throughput: 0: 739.1. Samples: 534344. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:57:33,044][682983] Avg episode reward: [(0, '11.104')] +[2023-04-30 12:57:35,116][683074] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000528_2162688.pth... +[2023-04-30 12:57:35,138][683074] Removing ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000352_1441792.pth +[2023-04-30 12:57:38,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2170880. Throughput: 0: 739.2. Samples: 538444. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:57:38,045][682983] Avg episode reward: [(0, '11.005')] +[2023-04-30 12:57:43,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2183168. Throughput: 0: 740.6. Samples: 543188. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 12:57:43,044][682983] Avg episode reward: [(0, '11.569')] +[2023-04-30 12:57:44,619][683074] Saving new best policy, reward=11.569! +[2023-04-30 12:57:44,621][683137] Updated weights for policy 0, policy_version 535 (0.0408) +[2023-04-30 12:57:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2199552. Throughput: 0: 754.7. Samples: 545856. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:57:48,044][682983] Avg episode reward: [(0, '12.888')] +[2023-04-30 12:57:48,756][683074] Saving new best policy, reward=12.888! +[2023-04-30 12:57:53,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2215936. Throughput: 0: 763.0. Samples: 550312. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:57:53,045][682983] Avg episode reward: [(0, '12.593')] +[2023-04-30 12:57:58,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2228224. Throughput: 0: 757.0. Samples: 554840. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 12:57:58,044][682983] Avg episode reward: [(0, '12.292')] +[2023-04-30 12:57:58,376][683137] Updated weights for policy 0, policy_version 545 (0.0821) +[2023-04-30 12:58:03,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2244608. Throughput: 0: 743.2. Samples: 556880. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:58:03,044][682983] Avg episode reward: [(0, '11.857')] +[2023-04-30 12:58:08,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2260992. Throughput: 0: 739.6. Samples: 561032. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:58:08,044][682983] Avg episode reward: [(0, '11.286')] +[2023-04-30 12:58:11,949][683137] Updated weights for policy 0, policy_version 555 (0.0800) +[2023-04-30 12:58:12,642][683074] Signal inference workers to stop experience collection... (550 times) +[2023-04-30 12:58:12,664][683137] InferenceWorker_p0-w0: stopping experience collection (550 times) +[2023-04-30 12:58:13,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 2273280. Throughput: 0: 748.3. Samples: 565808. Policy #0 lag: (min: 1.0, avg: 2.3, max: 3.0) +[2023-04-30 12:58:13,044][682983] Avg episode reward: [(0, '11.360')] +[2023-04-30 12:58:13,322][683074] Signal inference workers to resume experience collection... (550 times) +[2023-04-30 12:58:13,322][683137] InferenceWorker_p0-w0: resuming experience collection (550 times) +[2023-04-30 12:58:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2289664. Throughput: 0: 749.2. Samples: 568056. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:58:18,045][682983] Avg episode reward: [(0, '11.548')] +[2023-04-30 12:58:23,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2306048. Throughput: 0: 764.5. Samples: 572848. Policy #0 lag: (min: 1.0, avg: 2.3, max: 4.0) +[2023-04-30 12:58:23,045][682983] Avg episode reward: [(0, '11.677')] +[2023-04-30 12:58:25,579][683137] Updated weights for policy 0, policy_version 565 (0.1000) +[2023-04-30 12:58:28,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 2318336. Throughput: 0: 753.6. Samples: 577100. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:58:28,044][682983] Avg episode reward: [(0, '11.566')] +[2023-04-30 12:58:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2334720. Throughput: 0: 747.5. Samples: 579492. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:58:33,044][682983] Avg episode reward: [(0, '13.022')] +[2023-04-30 12:58:35,207][683074] Saving new best policy, reward=13.022! +[2023-04-30 12:58:38,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2351104. Throughput: 0: 739.3. Samples: 583580. Policy #0 lag: (min: 1.0, avg: 2.3, max: 4.0) +[2023-04-30 12:58:38,044][682983] Avg episode reward: [(0, '13.044')] +[2023-04-30 12:58:39,308][683074] Saving new best policy, reward=13.044! +[2023-04-30 12:58:39,310][683137] Updated weights for policy 0, policy_version 575 (0.0609) +[2023-04-30 12:58:43,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2363392. Throughput: 0: 737.1. Samples: 588008. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:58:43,044][682983] Avg episode reward: [(0, '13.431')] +[2023-04-30 12:58:44,839][683074] Saving new best policy, reward=13.431! +[2023-04-30 12:58:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2379776. Throughput: 0: 750.4. Samples: 590648. Policy #0 lag: (min: 1.0, avg: 2.3, max: 4.0) +[2023-04-30 12:58:48,044][682983] Avg episode reward: [(0, '13.832')] +[2023-04-30 12:58:48,913][683074] Saving new best policy, reward=13.832! +[2023-04-30 12:58:52,991][683137] Updated weights for policy 0, policy_version 585 (0.0615) +[2023-04-30 12:58:53,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2396160. Throughput: 0: 762.6. Samples: 595348. Policy #0 lag: (min: 1.0, avg: 2.2, max: 4.0) +[2023-04-30 12:58:53,044][682983] Avg episode reward: [(0, '14.697')] +[2023-04-30 12:58:54,362][683074] Saving new best policy, reward=14.697! +[2023-04-30 12:58:58,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 2408448. Throughput: 0: 748.2. Samples: 599476. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:58:58,044][682983] Avg episode reward: [(0, '14.173')] +[2023-04-30 12:59:03,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2424832. Throughput: 0: 751.2. Samples: 601860. Policy #0 lag: (min: 1.0, avg: 2.2, max: 4.0) +[2023-04-30 12:59:03,044][682983] Avg episode reward: [(0, '13.818')] +[2023-04-30 12:59:06,870][683137] Updated weights for policy 0, policy_version 595 (0.0827) +[2023-04-30 12:59:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2985.2). Total num frames: 2437120. Throughput: 0: 738.8. Samples: 606096. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:59:08,045][682983] Avg episode reward: [(0, '13.007')] +[2023-04-30 12:59:13,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2453504. Throughput: 0: 737.0. Samples: 610264. Policy #0 lag: (min: 1.0, avg: 2.3, max: 4.0) +[2023-04-30 12:59:13,044][682983] Avg episode reward: [(0, '13.464')] +[2023-04-30 12:59:18,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2469888. Throughput: 0: 741.8. Samples: 612872. Policy #0 lag: (min: 1.0, avg: 2.2, max: 4.0) +[2023-04-30 12:59:18,044][682983] Avg episode reward: [(0, '13.153')] +[2023-04-30 12:59:20,409][683137] Updated weights for policy 0, policy_version 605 (0.0814) +[2023-04-30 12:59:21,031][683074] Signal inference workers to stop experience collection... (600 times) +[2023-04-30 12:59:21,059][683137] InferenceWorker_p0-w0: stopping experience collection (600 times) +[2023-04-30 12:59:21,783][683074] Signal inference workers to resume experience collection... (600 times) +[2023-04-30 12:59:21,783][683137] InferenceWorker_p0-w0: resuming experience collection (600 times) +[2023-04-30 12:59:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2985.2). Total num frames: 2482176. Throughput: 0: 760.2. Samples: 617788. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:59:23,044][682983] Avg episode reward: [(0, '13.263')] +[2023-04-30 12:59:28,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2498560. Throughput: 0: 754.0. Samples: 621940. Policy #0 lag: (min: 1.0, avg: 2.2, max: 4.0) +[2023-04-30 12:59:28,044][682983] Avg episode reward: [(0, '13.268')] +[2023-04-30 12:59:33,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2514944. Throughput: 0: 752.7. Samples: 624520. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:59:33,045][682983] Avg episode reward: [(0, '14.427')] +[2023-04-30 12:59:34,004][683074] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000615_2519040.pth... +[2023-04-30 12:59:34,006][683137] Updated weights for policy 0, policy_version 615 (0.0612) +[2023-04-30 12:59:34,028][683074] Removing ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000440_1802240.pth +[2023-04-30 12:59:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2985.2). Total num frames: 2527232. Throughput: 0: 739.6. Samples: 628632. Policy #0 lag: (min: 1.0, avg: 2.2, max: 4.0) +[2023-04-30 12:59:38,045][682983] Avg episode reward: [(0, '14.446')] +[2023-04-30 12:59:43,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2543616. Throughput: 0: 741.2. Samples: 632832. Policy #0 lag: (min: 1.0, avg: 2.2, max: 4.0) +[2023-04-30 12:59:43,044][682983] Avg episode reward: [(0, '13.970')] +[2023-04-30 12:59:47,882][683137] Updated weights for policy 0, policy_version 625 (0.0820) +[2023-04-30 12:59:48,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2560000. Throughput: 0: 739.3. Samples: 635128. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 12:59:48,044][682983] Avg episode reward: [(0, '14.162')] +[2023-04-30 12:59:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2985.2). Total num frames: 2572288. Throughput: 0: 758.7. Samples: 640236. Policy #0 lag: (min: 1.0, avg: 2.2, max: 4.0) +[2023-04-30 12:59:53,044][682983] Avg episode reward: [(0, '14.640')] +[2023-04-30 12:59:58,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2588672. Throughput: 0: 759.6. Samples: 644448. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 12:59:58,044][682983] Avg episode reward: [(0, '14.552')] +[2023-04-30 13:00:01,358][683137] Updated weights for policy 0, policy_version 635 (0.0613) +[2023-04-30 13:00:03,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2605056. Throughput: 0: 756.4. Samples: 646912. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:00:03,044][682983] Avg episode reward: [(0, '14.921')] +[2023-04-30 13:00:04,140][683074] Saving new best policy, reward=14.921! +[2023-04-30 13:00:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 2617344. Throughput: 0: 744.6. Samples: 651296. Policy #0 lag: (min: 1.0, avg: 2.2, max: 4.0) +[2023-04-30 13:00:08,044][682983] Avg episode reward: [(0, '15.287')] +[2023-04-30 13:00:09,541][683074] Saving new best policy, reward=15.287! +[2023-04-30 13:00:13,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2633728. Throughput: 0: 743.8. Samples: 655412. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:00:13,044][682983] Avg episode reward: [(0, '14.729')] +[2023-04-30 13:00:15,055][683137] Updated weights for policy 0, policy_version 645 (0.0411) +[2023-04-30 13:00:18,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2650112. Throughput: 0: 740.9. Samples: 657860. Policy #0 lag: (min: 1.0, avg: 2.3, max: 4.0) +[2023-04-30 13:00:18,044][682983] Avg episode reward: [(0, '14.097')] +[2023-04-30 13:00:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 2662400. Throughput: 0: 754.8. Samples: 662596. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:00:23,044][682983] Avg episode reward: [(0, '14.728')] +[2023-04-30 13:00:28,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2678784. Throughput: 0: 758.5. Samples: 666964. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:00:28,044][682983] Avg episode reward: [(0, '13.782')] +[2023-04-30 13:00:28,819][683137] Updated weights for policy 0, policy_version 655 (0.1023) +[2023-04-30 13:00:29,397][683074] Signal inference workers to stop experience collection... (650 times) +[2023-04-30 13:00:29,418][683137] InferenceWorker_p0-w0: stopping experience collection (650 times) +[2023-04-30 13:00:30,208][683074] Signal inference workers to resume experience collection... (650 times) +[2023-04-30 13:00:30,208][683137] InferenceWorker_p0-w0: resuming experience collection (650 times) +[2023-04-30 13:00:33,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2695168. Throughput: 0: 753.7. Samples: 669044. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:00:33,044][682983] Avg episode reward: [(0, '14.370')] +[2023-04-30 13:00:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2707456. Throughput: 0: 743.7. Samples: 673704. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:00:38,044][682983] Avg episode reward: [(0, '14.548')] +[2023-04-30 13:00:42,774][683137] Updated weights for policy 0, policy_version 665 (0.1226) +[2023-04-30 13:00:43,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2723840. Throughput: 0: 742.2. Samples: 677848. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:00:43,044][682983] Avg episode reward: [(0, '14.702')] +[2023-04-30 13:00:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2985.2). Total num frames: 2736128. Throughput: 0: 732.7. Samples: 679884. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:00:48,044][682983] Avg episode reward: [(0, '15.510')] +[2023-04-30 13:00:49,457][683074] Saving new best policy, reward=15.510! +[2023-04-30 13:00:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2752512. Throughput: 0: 737.2. Samples: 684472. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:00:53,044][682983] Avg episode reward: [(0, '15.263')] +[2023-04-30 13:00:56,263][683137] Updated weights for policy 0, policy_version 675 (0.0828) +[2023-04-30 13:00:58,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2768896. Throughput: 0: 758.5. Samples: 689544. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:00:58,044][682983] Avg episode reward: [(0, '14.507')] +[2023-04-30 13:01:03,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2985.2). Total num frames: 2781184. Throughput: 0: 749.7. Samples: 691596. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:01:03,044][682983] Avg episode reward: [(0, '14.790')] +[2023-04-30 13:01:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2797568. Throughput: 0: 748.9. Samples: 696296. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:01:08,044][682983] Avg episode reward: [(0, '14.394')] +[2023-04-30 13:01:09,921][683137] Updated weights for policy 0, policy_version 685 (0.0806) +[2023-04-30 13:01:13,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2813952. Throughput: 0: 742.1. Samples: 700360. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:01:13,044][682983] Avg episode reward: [(0, '14.092')] +[2023-04-30 13:01:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2985.2). Total num frames: 2826240. Throughput: 0: 742.1. Samples: 702440. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:01:18,044][682983] Avg episode reward: [(0, '14.554')] +[2023-04-30 13:01:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 2842624. Throughput: 0: 735.1. Samples: 706784. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:01:23,044][682983] Avg episode reward: [(0, '14.575')] +[2023-04-30 13:01:23,656][683137] Updated weights for policy 0, policy_version 695 (0.0820) +[2023-04-30 13:01:28,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2859008. Throughput: 0: 760.0. Samples: 712048. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:01:28,044][682983] Avg episode reward: [(0, '15.110')] +[2023-04-30 13:01:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2985.2). Total num frames: 2871296. Throughput: 0: 760.4. Samples: 714100. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:01:33,045][682983] Avg episode reward: [(0, '14.599')] +[2023-04-30 13:01:34,482][683074] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000703_2879488.pth... +[2023-04-30 13:01:34,507][683074] Removing ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000528_2162688.pth +[2023-04-30 13:01:37,211][683137] Updated weights for policy 0, policy_version 705 (0.1002) +[2023-04-30 13:01:37,869][683074] Signal inference workers to stop experience collection... (700 times) +[2023-04-30 13:01:37,890][683137] InferenceWorker_p0-w0: stopping experience collection (700 times) +[2023-04-30 13:01:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2887680. Throughput: 0: 762.8. Samples: 718800. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:01:38,044][682983] Avg episode reward: [(0, '15.883')] +[2023-04-30 13:01:38,612][683074] Signal inference workers to resume experience collection... (700 times) +[2023-04-30 13:01:38,613][683137] InferenceWorker_p0-w0: resuming experience collection (700 times) +[2023-04-30 13:01:39,946][683074] Saving new best policy, reward=15.883! +[2023-04-30 13:01:43,044][682983] Fps is (10 sec: 3276.9, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2904064. Throughput: 0: 740.4. Samples: 722860. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:01:43,044][682983] Avg episode reward: [(0, '16.496')] +[2023-04-30 13:01:44,027][683074] Saving new best policy, reward=16.496! +[2023-04-30 13:01:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 2916352. Throughput: 0: 739.5. Samples: 724872. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:01:48,044][682983] Avg episode reward: [(0, '16.137')] +[2023-04-30 13:01:50,854][683137] Updated weights for policy 0, policy_version 715 (0.0612) +[2023-04-30 13:01:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2932736. Throughput: 0: 745.3. Samples: 729836. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:01:53,044][682983] Avg episode reward: [(0, '15.545')] +[2023-04-30 13:01:58,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2949120. Throughput: 0: 759.7. Samples: 734548. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:01:58,044][682983] Avg episode reward: [(0, '15.810')] +[2023-04-30 13:02:03,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 2961408. Throughput: 0: 759.4. Samples: 736612. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:02:03,044][682983] Avg episode reward: [(0, '14.536')] +[2023-04-30 13:02:04,465][683137] Updated weights for policy 0, policy_version 725 (0.0413) +[2023-04-30 13:02:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2977792. Throughput: 0: 764.7. Samples: 741196. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:02:08,044][682983] Avg episode reward: [(0, '13.761')] +[2023-04-30 13:02:13,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 2994176. Throughput: 0: 740.9. Samples: 745388. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:02:13,044][682983] Avg episode reward: [(0, '14.195')] +[2023-04-30 13:02:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 3006464. Throughput: 0: 740.5. Samples: 747424. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:02:18,044][682983] Avg episode reward: [(0, '14.984')] +[2023-04-30 13:02:18,312][683137] Updated weights for policy 0, policy_version 735 (0.0815) +[2023-04-30 13:02:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3022848. Throughput: 0: 737.2. Samples: 751976. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:02:23,044][682983] Avg episode reward: [(0, '15.762')] +[2023-04-30 13:02:28,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3039232. Throughput: 0: 761.7. Samples: 757136. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:02:28,044][682983] Avg episode reward: [(0, '16.268')] +[2023-04-30 13:02:31,945][683137] Updated weights for policy 0, policy_version 745 (0.1199) +[2023-04-30 13:02:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 3051520. Throughput: 0: 762.2. Samples: 759172. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:02:33,044][682983] Avg episode reward: [(0, '16.758')] +[2023-04-30 13:02:34,443][683074] Saving new best policy, reward=16.758! +[2023-04-30 13:02:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3067904. Throughput: 0: 756.0. Samples: 763856. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:02:38,044][682983] Avg episode reward: [(0, '18.018')] +[2023-04-30 13:02:39,978][683074] Saving new best policy, reward=18.018! +[2023-04-30 13:02:43,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3084288. Throughput: 0: 742.8. Samples: 767976. Policy #0 lag: (min: 1.0, avg: 2.3, max: 3.0) +[2023-04-30 13:02:43,044][682983] Avg episode reward: [(0, '17.980')] +[2023-04-30 13:02:45,496][683137] Updated weights for policy 0, policy_version 755 (0.0422) +[2023-04-30 13:02:46,083][683074] Signal inference workers to stop experience collection... (750 times) +[2023-04-30 13:02:46,104][683137] InferenceWorker_p0-w0: stopping experience collection (750 times) +[2023-04-30 13:02:46,844][683074] Signal inference workers to resume experience collection... (750 times) +[2023-04-30 13:02:46,845][683137] InferenceWorker_p0-w0: resuming experience collection (750 times) +[2023-04-30 13:02:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 3096576. Throughput: 0: 743.1. Samples: 770052. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:02:48,044][682983] Avg episode reward: [(0, '16.947')] +[2023-04-30 13:02:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3112960. Throughput: 0: 736.5. Samples: 774340. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:02:53,045][682983] Avg episode reward: [(0, '16.887')] +[2023-04-30 13:02:58,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3129344. Throughput: 0: 760.3. Samples: 779600. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:02:58,044][682983] Avg episode reward: [(0, '16.813')] +[2023-04-30 13:02:59,125][683137] Updated weights for policy 0, policy_version 765 (0.1199) +[2023-04-30 13:03:03,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 3141632. Throughput: 0: 760.3. Samples: 781636. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:03:03,044][682983] Avg episode reward: [(0, '15.977')] +[2023-04-30 13:03:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3158016. Throughput: 0: 765.2. Samples: 786412. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:03:08,044][682983] Avg episode reward: [(0, '16.698')] +[2023-04-30 13:03:12,932][683137] Updated weights for policy 0, policy_version 775 (0.0816) +[2023-04-30 13:03:13,044][682983] Fps is (10 sec: 3276.7, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3174400. Throughput: 0: 741.2. Samples: 790492. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:03:13,045][682983] Avg episode reward: [(0, '17.640')] +[2023-04-30 13:03:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 3186688. Throughput: 0: 741.9. Samples: 792556. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:03:18,044][682983] Avg episode reward: [(0, '16.939')] +[2023-04-30 13:03:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3203072. Throughput: 0: 733.8. Samples: 796876. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:03:23,045][682983] Avg episode reward: [(0, '17.946')] +[2023-04-30 13:03:26,430][683137] Updated weights for policy 0, policy_version 785 (0.1016) +[2023-04-30 13:03:28,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3219456. Throughput: 0: 759.5. Samples: 802152. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:03:28,045][682983] Avg episode reward: [(0, '18.516')] +[2023-04-30 13:03:29,161][683074] Saving new best policy, reward=18.516! +[2023-04-30 13:03:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 3231744. Throughput: 0: 759.0. Samples: 804208. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:03:33,044][682983] Avg episode reward: [(0, '18.890')] +[2023-04-30 13:03:34,662][683074] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000791_3239936.pth... +[2023-04-30 13:03:34,686][683074] Removing ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000615_2519040.pth +[2023-04-30 13:03:34,689][683074] Saving new best policy, reward=18.890! +[2023-04-30 13:03:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3248128. Throughput: 0: 764.7. Samples: 808752. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:03:38,044][682983] Avg episode reward: [(0, '18.836')] +[2023-04-30 13:03:40,104][683137] Updated weights for policy 0, policy_version 795 (0.0212) +[2023-04-30 13:03:43,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3264512. Throughput: 0: 743.5. Samples: 813056. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:03:43,045][682983] Avg episode reward: [(0, '19.253')] +[2023-04-30 13:03:44,192][683074] Saving new best policy, reward=19.253! +[2023-04-30 13:03:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2985.2). Total num frames: 3276800. Throughput: 0: 743.9. Samples: 815112. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:03:48,044][682983] Avg episode reward: [(0, '20.032')] +[2023-04-30 13:03:49,634][683074] Saving new best policy, reward=20.032! +[2023-04-30 13:03:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3293184. Throughput: 0: 731.0. Samples: 819308. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:03:53,044][682983] Avg episode reward: [(0, '20.225')] +[2023-04-30 13:03:53,716][683137] Updated weights for policy 0, policy_version 805 (0.0798) +[2023-04-30 13:03:54,275][683074] Signal inference workers to stop experience collection... (800 times) +[2023-04-30 13:03:54,296][683137] InferenceWorker_p0-w0: stopping experience collection (800 times) +[2023-04-30 13:03:55,067][683074] Signal inference workers to resume experience collection... (800 times) +[2023-04-30 13:03:55,068][683074] Saving new best policy, reward=20.225! +[2023-04-30 13:03:55,068][683137] InferenceWorker_p0-w0: resuming experience collection (800 times) +[2023-04-30 13:03:58,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3309568. Throughput: 0: 759.3. Samples: 824660. Policy #0 lag: (min: 1.0, avg: 2.3, max: 3.0) +[2023-04-30 13:03:58,044][682983] Avg episode reward: [(0, '20.326')] +[2023-04-30 13:03:59,155][683074] Saving new best policy, reward=20.326! +[2023-04-30 13:04:03,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3321856. Throughput: 0: 759.3. Samples: 826724. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:04:03,044][682983] Avg episode reward: [(0, '20.112')] +[2023-04-30 13:04:07,458][683137] Updated weights for policy 0, policy_version 815 (0.1039) +[2023-04-30 13:04:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3338240. Throughput: 0: 756.4. Samples: 830916. Policy #0 lag: (min: 1.0, avg: 2.3, max: 4.0) +[2023-04-30 13:04:08,044][682983] Avg episode reward: [(0, '19.533')] +[2023-04-30 13:04:13,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3354624. Throughput: 0: 743.8. Samples: 835624. Policy #0 lag: (min: 1.0, avg: 2.3, max: 4.0) +[2023-04-30 13:04:13,044][682983] Avg episode reward: [(0, '20.020')] +[2023-04-30 13:04:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3366912. Throughput: 0: 743.4. Samples: 837660. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:04:18,044][682983] Avg episode reward: [(0, '18.374')] +[2023-04-30 13:04:21,109][683137] Updated weights for policy 0, policy_version 825 (0.0998) +[2023-04-30 13:04:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3383296. Throughput: 0: 739.2. Samples: 842016. Policy #0 lag: (min: 1.0, avg: 2.3, max: 3.0) +[2023-04-30 13:04:23,044][682983] Avg episode reward: [(0, '18.362')] +[2023-04-30 13:04:28,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3399680. Throughput: 0: 751.0. Samples: 846852. Policy #0 lag: (min: 1.0, avg: 2.1, max: 4.0) +[2023-04-30 13:04:28,044][682983] Avg episode reward: [(0, '18.226')] +[2023-04-30 13:04:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3411968. Throughput: 0: 759.4. Samples: 849284. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:04:33,044][682983] Avg episode reward: [(0, '19.551')] +[2023-04-30 13:04:34,720][683137] Updated weights for policy 0, policy_version 835 (0.1216) +[2023-04-30 13:04:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3428352. Throughput: 0: 757.3. Samples: 853388. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:04:38,044][682983] Avg episode reward: [(0, '20.090')] +[2023-04-30 13:04:43,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3444736. Throughput: 0: 742.8. Samples: 858084. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:04:43,045][682983] Avg episode reward: [(0, '19.302')] +[2023-04-30 13:04:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3457024. Throughput: 0: 742.1. Samples: 860120. Policy #0 lag: (min: 1.0, avg: 2.2, max: 3.0) +[2023-04-30 13:04:48,044][682983] Avg episode reward: [(0, '20.120')] +[2023-04-30 13:04:48,281][683137] Updated weights for policy 0, policy_version 845 (0.0995) +[2023-04-30 13:04:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3473408. Throughput: 0: 749.9. Samples: 864660. Policy #0 lag: (min: 1.0, avg: 2.2, max: 4.0) +[2023-04-30 13:04:53,044][682983] Avg episode reward: [(0, '20.576')] +[2023-04-30 13:04:55,093][683074] Saving new best policy, reward=20.576! +[2023-04-30 13:04:58,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3489792. Throughput: 0: 758.1. Samples: 869740. Policy #0 lag: (min: 1.0, avg: 2.2, max: 4.0) +[2023-04-30 13:04:58,044][682983] Avg episode reward: [(0, '20.160')] +[2023-04-30 13:05:02,072][683137] Updated weights for policy 0, policy_version 855 (0.1027) +[2023-04-30 13:05:02,733][683074] Signal inference workers to stop experience collection... (850 times) +[2023-04-30 13:05:02,753][683137] InferenceWorker_p0-w0: stopping experience collection (850 times) +[2023-04-30 13:05:03,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3502080. Throughput: 0: 758.0. Samples: 871772. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:05:03,044][682983] Avg episode reward: [(0, '19.109')] +[2023-04-30 13:05:03,253][683074] Signal inference workers to resume experience collection... (850 times) +[2023-04-30 13:05:03,254][683137] InferenceWorker_p0-w0: resuming experience collection (850 times) +[2023-04-30 13:05:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3518464. Throughput: 0: 755.6. Samples: 876020. Policy #0 lag: (min: 1.0, avg: 2.1, max: 4.0) +[2023-04-30 13:05:08,044][682983] Avg episode reward: [(0, '19.191')] +[2023-04-30 13:05:13,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3534848. Throughput: 0: 745.9. Samples: 880416. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:05:13,045][682983] Avg episode reward: [(0, '19.888')] +[2023-04-30 13:05:15,467][683137] Updated weights for policy 0, policy_version 865 (0.0811) +[2023-04-30 13:05:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3547136. Throughput: 0: 737.7. Samples: 882480. Policy #0 lag: (min: 1.0, avg: 2.1, max: 4.0) +[2023-04-30 13:05:18,044][682983] Avg episode reward: [(0, '20.366')] +[2023-04-30 13:05:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3563520. Throughput: 0: 751.9. Samples: 887224. Policy #0 lag: (min: 1.0, avg: 2.1, max: 4.0) +[2023-04-30 13:05:23,045][682983] Avg episode reward: [(0, '18.907')] +[2023-04-30 13:05:28,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3579904. Throughput: 0: 757.3. Samples: 892164. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:05:28,044][682983] Avg episode reward: [(0, '18.653')] +[2023-04-30 13:05:29,298][683137] Updated weights for policy 0, policy_version 875 (0.0815) +[2023-04-30 13:05:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3592192. Throughput: 0: 757.8. Samples: 894220. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:05:33,044][682983] Avg episode reward: [(0, '19.605')] +[2023-04-30 13:05:34,608][683074] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000879_3600384.pth... +[2023-04-30 13:05:34,633][683074] Removing ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000703_2879488.pth +[2023-04-30 13:05:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3608576. Throughput: 0: 758.6. Samples: 898796. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:05:38,044][682983] Avg episode reward: [(0, '18.191')] +[2023-04-30 13:05:42,970][683137] Updated weights for policy 0, policy_version 885 (0.0601) +[2023-04-30 13:05:43,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 3013.0). Total num frames: 3624960. Throughput: 0: 739.1. Samples: 903000. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:05:43,044][682983] Avg episode reward: [(0, '17.661')] +[2023-04-30 13:05:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3637248. Throughput: 0: 738.8. Samples: 905016. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:05:48,044][682983] Avg episode reward: [(0, '17.378')] +[2023-04-30 13:05:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3653632. Throughput: 0: 750.1. Samples: 909776. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:05:53,044][682983] Avg episode reward: [(0, '17.987')] +[2023-04-30 13:05:56,402][683137] Updated weights for policy 0, policy_version 895 (0.0829) +[2023-04-30 13:05:58,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 3013.0). Total num frames: 3670016. Throughput: 0: 762.1. Samples: 914712. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:05:58,044][682983] Avg episode reward: [(0, '18.096')] +[2023-04-30 13:06:03,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3682304. Throughput: 0: 762.1. Samples: 916776. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:06:03,045][682983] Avg episode reward: [(0, '17.302')] +[2023-04-30 13:06:08,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3698688. Throughput: 0: 750.6. Samples: 921000. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:06:08,044][682983] Avg episode reward: [(0, '18.105')] +[2023-04-30 13:06:10,027][683137] Updated weights for policy 0, policy_version 905 (0.0615) +[2023-04-30 13:06:10,627][683074] Signal inference workers to stop experience collection... (900 times) +[2023-04-30 13:06:10,649][683137] InferenceWorker_p0-w0: stopping experience collection (900 times) +[2023-04-30 13:06:11,391][683074] Signal inference workers to resume experience collection... (900 times) +[2023-04-30 13:06:11,392][683137] InferenceWorker_p0-w0: resuming experience collection (900 times) +[2023-04-30 13:06:13,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 3013.0). Total num frames: 3715072. Throughput: 0: 742.3. Samples: 925568. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:06:13,045][682983] Avg episode reward: [(0, '18.338')] +[2023-04-30 13:06:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3727360. Throughput: 0: 742.7. Samples: 927640. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:06:18,044][682983] Avg episode reward: [(0, '18.425')] +[2023-04-30 13:06:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3743744. Throughput: 0: 741.1. Samples: 932144. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:06:23,044][682983] Avg episode reward: [(0, '18.785')] +[2023-04-30 13:06:23,750][683137] Updated weights for policy 0, policy_version 915 (0.1016) +[2023-04-30 13:06:28,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 3013.0). Total num frames: 3760128. Throughput: 0: 755.0. Samples: 936976. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:06:28,044][682983] Avg episode reward: [(0, '18.571')] +[2023-04-30 13:06:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3772416. Throughput: 0: 762.0. Samples: 939304. Policy #0 lag: (min: 1.0, avg: 2.0, max: 3.0) +[2023-04-30 13:06:33,045][682983] Avg episode reward: [(0, '19.278')] +[2023-04-30 13:06:37,462][683137] Updated weights for policy 0, policy_version 925 (0.0820) +[2023-04-30 13:06:38,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3788800. Throughput: 0: 748.8. Samples: 943472. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:06:38,044][682983] Avg episode reward: [(0, '19.292')] +[2023-04-30 13:06:43,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 3013.0). Total num frames: 3805184. Throughput: 0: 742.5. Samples: 948124. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:06:43,044][682983] Avg episode reward: [(0, '19.691')] +[2023-04-30 13:06:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3817472. Throughput: 0: 742.7. Samples: 950196. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:06:48,044][682983] Avg episode reward: [(0, '19.590')] +[2023-04-30 13:06:51,378][683137] Updated weights for policy 0, policy_version 935 (0.0809) +[2023-04-30 13:06:53,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3833856. Throughput: 0: 741.9. Samples: 954384. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:06:53,045][682983] Avg episode reward: [(0, '20.647')] +[2023-04-30 13:06:55,307][683074] Saving new best policy, reward=20.647! +[2023-04-30 13:06:58,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2999.1). Total num frames: 3846144. Throughput: 0: 756.2. Samples: 959596. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:06:58,044][682983] Avg episode reward: [(0, '20.474')] +[2023-04-30 13:07:03,044][682983] Fps is (10 sec: 2867.3, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3862528. Throughput: 0: 758.2. Samples: 961760. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:07:03,044][682983] Avg episode reward: [(0, '20.640')] +[2023-04-30 13:07:04,875][683137] Updated weights for policy 0, policy_version 945 (0.1224) +[2023-04-30 13:07:08,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3878912. Throughput: 0: 749.9. Samples: 965888. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:07:08,044][682983] Avg episode reward: [(0, '20.677')] +[2023-04-30 13:07:08,923][683074] Saving new best policy, reward=20.677! +[2023-04-30 13:07:13,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 3013.0). Total num frames: 3895296. Throughput: 0: 749.4. Samples: 970700. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:07:13,044][682983] Avg episode reward: [(0, '20.801')] +[2023-04-30 13:07:14,378][683074] Saving new best policy, reward=20.801! +[2023-04-30 13:07:18,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3907584. Throughput: 0: 743.6. Samples: 972764. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:07:18,044][682983] Avg episode reward: [(0, '20.694')] +[2023-04-30 13:07:18,475][683137] Updated weights for policy 0, policy_version 955 (0.0998) +[2023-04-30 13:07:19,050][683074] Signal inference workers to stop experience collection... (950 times) +[2023-04-30 13:07:19,071][683137] InferenceWorker_p0-w0: stopping experience collection (950 times) +[2023-04-30 13:07:19,863][683074] Signal inference workers to resume experience collection... (950 times) +[2023-04-30 13:07:19,863][683137] InferenceWorker_p0-w0: resuming experience collection (950 times) +[2023-04-30 13:07:23,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3923968. Throughput: 0: 744.0. Samples: 976952. Policy #0 lag: (min: 1.0, avg: 2.1, max: 4.0) +[2023-04-30 13:07:23,044][682983] Avg episode reward: [(0, '20.036')] +[2023-04-30 13:07:28,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2999.1). Total num frames: 3936256. Throughput: 0: 748.4. Samples: 981800. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:07:28,044][682983] Avg episode reward: [(0, '19.883')] +[2023-04-30 13:07:32,132][683137] Updated weights for policy 0, policy_version 965 (0.0822) +[2023-04-30 13:07:33,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3952640. Throughput: 0: 758.1. Samples: 984312. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:07:33,044][682983] Avg episode reward: [(0, '18.899')] +[2023-04-30 13:07:34,898][683074] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000967_3960832.pth... +[2023-04-30 13:07:34,925][683074] Removing ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000791_3239936.pth +[2023-04-30 13:07:38,044][682983] Fps is (10 sec: 3276.8, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3969024. Throughput: 0: 755.7. Samples: 988392. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:07:38,044][682983] Avg episode reward: [(0, '18.993')] +[2023-04-30 13:07:43,044][682983] Fps is (10 sec: 2867.2, 60 sec: 2935.5, 300 sec: 2999.1). Total num frames: 3981312. Throughput: 0: 736.7. Samples: 992748. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:07:43,044][682983] Avg episode reward: [(0, '18.584')] +[2023-04-30 13:07:46,267][683137] Updated weights for policy 0, policy_version 975 (0.0823) +[2023-04-30 13:07:48,044][682983] Fps is (10 sec: 2867.2, 60 sec: 3003.7, 300 sec: 2999.1). Total num frames: 3997696. Throughput: 0: 745.8. Samples: 995320. Policy #0 lag: (min: 1.0, avg: 2.1, max: 3.0) +[2023-04-30 13:07:48,044][682983] Avg episode reward: [(0, '19.073')] +[2023-04-30 13:07:50,206][683074] Stopping Batcher_0... +[2023-04-30 13:07:50,207][683074] Loop batcher_evt_loop terminating... +[2023-04-30 13:07:50,214][682983] Component Batcher_0 stopped! +[2023-04-30 13:07:50,223][683139] Stopping RolloutWorker_w1... +[2023-04-30 13:07:50,223][682983] Component RolloutWorker_w1 stopped! +[2023-04-30 13:07:50,223][683139] Loop rollout_proc1_evt_loop terminating... +[2023-04-30 13:07:50,223][683149] Stopping RolloutWorker_w7... +[2023-04-30 13:07:50,223][683141] Stopping RolloutWorker_w3... +[2023-04-30 13:07:50,223][683145] Stopping RolloutWorker_w6... +[2023-04-30 13:07:50,223][683140] Stopping RolloutWorker_w2... +[2023-04-30 13:07:50,223][683149] Loop rollout_proc7_evt_loop terminating... +[2023-04-30 13:07:50,223][682983] Component RolloutWorker_w7 stopped! +[2023-04-30 13:07:50,224][682983] Component RolloutWorker_w3 stopped! +[2023-04-30 13:07:50,224][683145] Loop rollout_proc6_evt_loop terminating... +[2023-04-30 13:07:50,224][683141] Loop rollout_proc3_evt_loop terminating... +[2023-04-30 13:07:50,224][683140] Loop rollout_proc2_evt_loop terminating... +[2023-04-30 13:07:50,224][682983] Component RolloutWorker_w6 stopped! +[2023-04-30 13:07:50,224][683144] Stopping RolloutWorker_w5... +[2023-04-30 13:07:50,224][683142] Stopping RolloutWorker_w4... +[2023-04-30 13:07:50,224][682983] Component RolloutWorker_w2 stopped! +[2023-04-30 13:07:50,224][682983] Component RolloutWorker_w5 stopped! +[2023-04-30 13:07:50,224][683142] Loop rollout_proc4_evt_loop terminating... +[2023-04-30 13:07:50,224][683144] Loop rollout_proc5_evt_loop terminating... +[2023-04-30 13:07:50,225][682983] Component RolloutWorker_w4 stopped! +[2023-04-30 13:07:50,240][683138] Stopping RolloutWorker_w0... +[2023-04-30 13:07:50,240][682983] Component RolloutWorker_w0 stopped! +[2023-04-30 13:07:50,241][683138] Loop rollout_proc0_evt_loop terminating... +[2023-04-30 13:07:50,280][683137] Weights refcount: 2 0 +[2023-04-30 13:07:50,281][683137] Stopping InferenceWorker_p0-w0... +[2023-04-30 13:07:50,282][683137] Loop inference_proc0-0_evt_loop terminating... +[2023-04-30 13:07:50,282][682983] Component InferenceWorker_p0-w0 stopped! +[2023-04-30 13:07:51,610][683074] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000979_4009984.pth... +[2023-04-30 13:07:51,633][683074] Removing ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000879_3600384.pth +[2023-04-30 13:07:51,636][683074] Saving ./train_dir/doom_health_gathering_supreme/checkpoint_p0/checkpoint_000000979_4009984.pth... +[2023-04-30 13:07:51,667][683074] Stopping LearnerWorker_p0... +[2023-04-30 13:07:51,667][682983] Component LearnerWorker_p0 stopped! +[2023-04-30 13:07:51,667][683074] Loop learner_proc0_evt_loop terminating... +[2023-04-30 13:07:51,667][682983] Waiting for process learner_proc0 to stop... +[2023-04-30 13:07:51,842][682983] Waiting for process inference_proc0-0 to join... +[2023-04-30 13:07:51,843][682983] Waiting for process rollout_proc0 to join... +[2023-04-30 13:07:51,843][682983] Waiting for process rollout_proc1 to join... +[2023-04-30 13:07:51,843][682983] Waiting for process rollout_proc2 to join... +[2023-04-30 13:07:51,843][682983] Waiting for process rollout_proc3 to join... +[2023-04-30 13:07:51,843][682983] Waiting for process rollout_proc4 to join... +[2023-04-30 13:07:51,843][682983] Waiting for process rollout_proc5 to join... +[2023-04-30 13:07:51,844][682983] Waiting for process rollout_proc6 to join... +[2023-04-30 13:07:51,844][682983] Waiting for process rollout_proc7 to join... +[2023-04-30 13:07:51,844][682983] Batcher 0 profile tree view: +batching: 4.6847, releasing_batches: 0.1589 +[2023-04-30 13:07:51,844][682983] InferenceWorker_p0-w0 profile tree view: +wait_policy: 0.0051 + wait_policy_total: 6.7719 +update_model: 74.6759 + weight_update: 0.1017 +one_step: 0.0334 + handle_policy_step: 455.4509 + deserialize: 5.5774, stack: 0.4649, obs_to_device_normalize: 29.5796, forward: 403.7178, send_messages: 8.4459 + prepare_outputs: 4.5571 + to_cpu: 0.3871 +[2023-04-30 13:07:51,844][682983] Learner 0 profile tree view: +misc: 0.0030, prepare_batch: 230.8644 +train: 1097.8844 + epoch_init: 0.0034, minibatch_init: 0.0043, losses_postprocess: 0.0338, kl_divergence: 0.1467, after_optimizer: 0.7092 + calculate_losses: 382.0104 + losses_init: 0.0019, forward_head: 277.3595, bptt_initial: 0.9937, tail: 0.7954, advantages_returns: 0.0669, losses: 0.3979 + bptt: 102.2738 + bptt_forward_core: 102.0908 + update: 714.7093 + clip: 1.3180 +[2023-04-30 13:07:51,844][682983] RolloutWorker_w0 profile tree view: +wait_for_trajectories: 0.0646, enqueue_policy_requests: 3.8134, env_step: 90.4966, overhead: 7.0037, complete_rollouts: 0.1099 +save_policy_outputs: 4.7685 + split_output_tensors: 2.2309 +[2023-04-30 13:07:51,844][682983] RolloutWorker_w7 profile tree view: +wait_for_trajectories: 0.0636, enqueue_policy_requests: 3.7639, env_step: 90.4639, overhead: 7.0318, complete_rollouts: 0.1070 +save_policy_outputs: 4.7276 + split_output_tensors: 2.2120 +[2023-04-30 13:07:51,845][682983] Loop Runner_EvtLoop terminating... +[2023-04-30 13:07:51,845][682983] Runner profile tree view: +main_loop: 1337.7066 +[2023-04-30 13:07:51,845][682983] Collected {0: 4009984}, FPS: 2982.3