[2023-01-11 19:11:48,464][457818] Saving configuration to ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/04_v083_brax_basic_benchmark_see_2322090_env_halfcheetah_u.rnn_False_n.epo_5/config.json...
[2023-01-11 19:11:48,641][457818] Using GPUs [0] for process 0 (actually maps to GPUs [0])
[2023-01-11 19:11:48,642][457818] Rollout worker 0 uses device cuda:0
[2023-01-11 19:11:48,643][457818] In synchronous mode, we only accumulate one batch. Setting num_batches_to_accumulate to 1
[2023-01-11 19:11:48,675][457818] Using GPUs [0] for process 0 (actually maps to GPUs [0])
[2023-01-11 19:11:48,676][457818] InferenceWorker_p0-w0: min num requests: 1
[2023-01-11 19:11:48,677][457818] Using GPUs [0] for process 0 (actually maps to GPUs [0])
[2023-01-11 19:11:48,679][457818] WARNING! It is generally recommended to enable Fixed KL loss (https://arxiv.org/pdf/1707.06347.pdf) for continuous action tasks to avoid potential numerical issues. I.e. set --kl_loss_coeff=0.1
[2023-01-11 19:11:48,679][457818] Setting fixed seed 2322090
[2023-01-11 19:11:48,680][457818] Using GPUs [0] for process 0 (actually maps to GPUs [0])
[2023-01-11 19:11:48,680][457818] Initializing actor-critic model on device cuda:0
[2023-01-11 19:11:48,681][457818] RunningMeanStd input shape: (18,)
[2023-01-11 19:11:48,682][457818] RunningMeanStd input shape: (1,)
[2023-01-11 19:11:48,763][457818] Created Actor Critic model with architecture:
[2023-01-11 19:11:48,764][457818] ActorCriticSharedWeights(
  (obs_normalizer): ObservationNormalizer(
    (running_mean_std): RunningMeanStdDictInPlace(
      (running_mean_std): ModuleDict(
        (obs): RunningMeanStdInPlace()
      )
    )
  )
  (returns_normalizer): RecursiveScriptModule(original_name=RunningMeanStdInPlace)
  (encoder): MultiInputEncoder(
    (encoders): ModuleDict(
      (obs): MlpEncoder(
        (mlp_head): RecursiveScriptModule(
          original_name=Sequential
          (0): RecursiveScriptModule(original_name=Linear)
          (1): RecursiveScriptModule(original_name=ELU)
          (2): RecursiveScriptModule(original_name=Linear)
          (3): RecursiveScriptModule(original_name=ELU)
          (4): RecursiveScriptModule(original_name=Linear)
          (5): RecursiveScriptModule(original_name=ELU)
        )
      )
    )
  )
  (core): ModelCoreIdentity()
  (decoder): MlpDecoder(
    (mlp): Identity()
  )
  (critic_linear): Linear(in_features=64, out_features=1, bias=True)
  (action_parameterization): ActionParameterizationContinuousNonAdaptiveStddev(
    (distribution_linear): Linear(in_features=64, out_features=6, bias=True)
  )
)
[2023-01-11 19:11:48,767][457818] Using optimizer <class 'torch.optim.adam.Adam'>
[2023-01-11 19:11:48,770][457818] No checkpoints found
[2023-01-11 19:11:48,771][457818] Did not load from checkpoint, starting from scratch!
[2023-01-11 19:11:48,772][457818] Initialized policy 0 weights for model version 0
[2023-01-11 19:11:48,772][457818] LearnerWorker_p0 finished initialization!
[2023-01-11 19:11:48,773][457818] Using GPUs [0] for process 0 (actually maps to GPUs [0])
[2023-01-11 19:11:48,778][457818] Inference worker 0-0 is ready!
[2023-01-11 19:11:48,779][457818] All inference workers are ready! Signal rollout workers to start!
[2023-01-11 19:11:48,780][457818] EnvRunner 0-0 uses policy 0
[2023-01-11 19:11:50,292][457818] Resetting env <VectorGymWrapper instance> with 2048 parallel agents...
[2023-01-11 19:11:56,260][457818] reset() done, obs.shape=torch.Size([2048, 18])!
[2023-01-11 19:11:56,271][457818] 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)
[2023-01-11 19:12:09,903][457818] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 150.2. Samples: 2048. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
[2023-01-11 19:12:09,914][457818] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 300.2. Samples: 4096. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
[2023-01-11 19:12:09,917][457818] Heartbeat connected on Batcher_0
[2023-01-11 19:12:09,917][457818] Heartbeat connected on LearnerWorker_p0
[2023-01-11 19:12:09,918][457818] Heartbeat connected on InferenceWorker_p0-w0
[2023-01-11 19:12:09,918][457818] Heartbeat connected on RolloutWorker_w0
[2023-01-11 19:12:09,918][457818] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 300.1. Samples: 4096. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
[2023-01-11 19:12:13,067][457818] Fps is (10 sec: 145474.4, 60 sec: 27312.7, 300 sec: 27312.7). Total num frames: 458752. Throughput: 0: 23410.9. Samples: 393216. Policy #0 lag: (min: 2.0, avg: 2.0, max: 2.0)
[2023-01-11 19:12:18,043][457818] Fps is (10 sec: 209709.3, 60 sec: 78261.0, 300 sec: 78261.0). Total num frames: 1703936. Throughput: 0: 52769.7. Samples: 1148928. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:12:23,041][457818] Fps is (10 sec: 249685.6, 60 sec: 110163.8, 300 sec: 110163.8). Total num frames: 2949120. Throughput: 0: 100142.0. Samples: 2680832. Policy #0 lag: (min: 1.0, avg: 1.0, max: 1.0)
[2023-01-11 19:12:23,042][457818] Avg episode reward: [(0, '125.668')]
[2023-01-11 19:12:28,039][457818] Fps is (10 sec: 249141.6, 60 sec: 132028.1, 300 sec: 132028.1). Total num frames: 4194304. Throughput: 0: 131383.4. Samples: 4173824. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:12:28,040][457818] Avg episode reward: [(0, '2973.944')]
[2023-01-11 19:12:33,087][457818] Fps is (10 sec: 254425.7, 60 sec: 149527.7, 300 sec: 149527.7). Total num frames: 5505024. Throughput: 0: 133673.7. Samples: 4921344. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:12:33,088][457818] Avg episode reward: [(0, '2973.944')]
[2023-01-11 19:12:33,097][457818] Saving new best policy, reward=2973.944!
[2023-01-11 19:12:38,040][457818] Fps is (10 sec: 249028.2, 60 sec: 160040.5, 300 sec: 160040.5). Total num frames: 6684672. Throughput: 0: 154058.6. Samples: 6434816. Policy #0 lag: (min: 1.0, avg: 1.0, max: 1.0)
[2023-01-11 19:12:38,040][457818] Avg episode reward: [(0, '6932.740')]
[2023-01-11 19:12:38,102][457818] Saving new best policy, reward=6932.740!
[2023-01-11 19:12:43,044][457818] Fps is (10 sec: 243537.0, 60 sec: 169539.9, 300 sec: 169539.9). Total num frames: 7929856. Throughput: 0: 239154.4. Samples: 7927808. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:12:43,044][457818] Avg episode reward: [(0, '6932.740')]
[2023-01-11 19:12:48,040][457818] Fps is (10 sec: 249025.8, 60 sec: 177230.2, 300 sec: 177230.2). Total num frames: 9175040. Throughput: 0: 226951.2. Samples: 8656896. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:12:48,041][457818] Avg episode reward: [(0, '9671.154')]
[2023-01-11 19:12:48,048][457818] Saving new best policy, reward=9671.154!
[2023-01-11 19:12:53,043][457818] Fps is (10 sec: 249045.5, 60 sec: 183543.6, 300 sec: 183543.6). Total num frames: 10420224. Throughput: 0: 235453.5. Samples: 10158080. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:12:53,044][457818] Avg episode reward: [(0, '10904.311')]
[2023-01-11 19:12:53,052][457818] Saving new best policy, reward=10904.311!
[2023-01-11 19:12:58,043][457818] Fps is (10 sec: 248972.3, 60 sec: 242324.6, 300 sec: 188847.3). Total num frames: 11665408. Throughput: 0: 250311.6. Samples: 11651072. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:12:58,043][457818] Avg episode reward: [(0, '10904.311')]
[2023-01-11 19:13:03,043][457818] Fps is (10 sec: 249054.8, 60 sec: 243004.9, 300 sec: 193354.1). Total num frames: 12910592. Throughput: 0: 249723.5. Samples: 12386304. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:13:03,044][457818] Avg episode reward: [(0, '11910.594')]
[2023-01-11 19:13:03,051][457818] Saving new best policy, reward=11910.594!
[2023-01-11 19:13:08,044][457818] Fps is (10 sec: 249012.5, 60 sec: 243538.6, 300 sec: 197230.8). Total num frames: 14155776. Throughput: 0: 248932.9. Samples: 13883392. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:13:08,044][457818] Avg episode reward: [(0, '11910.594')]
[2023-01-11 19:13:13,043][457818] Fps is (10 sec: 249041.0, 60 sec: 249139.4, 300 sec: 200607.6). Total num frames: 15400960. Throughput: 0: 249428.1. Samples: 15398912. Policy #0 lag: (min: 1.0, avg: 1.0, max: 1.0)
[2023-01-11 19:13:13,043][457818] Avg episode reward: [(0, '12797.121')]
[2023-01-11 19:13:13,055][457818] Saving new best policy, reward=12797.121!
[2023-01-11 19:13:18,043][457818] Fps is (10 sec: 249047.4, 60 sec: 249037.9, 300 sec: 203567.3). Total num frames: 16646144. Throughput: 0: 249143.2. Samples: 16121856. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:13:18,044][457818] Avg episode reward: [(0, '13501.586')]
[2023-01-11 19:13:18,051][457818] Saving new best policy, reward=13501.586!
[2023-01-11 19:13:23,042][457818] Fps is (10 sec: 249049.9, 60 sec: 249033.7, 300 sec: 206190.1). Total num frames: 17891328. Throughput: 0: 248750.3. Samples: 17629184. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:13:23,043][457818] Avg episode reward: [(0, '13501.586')]
[2023-01-11 19:13:28,040][457818] Fps is (10 sec: 249120.1, 60 sec: 249034.4, 300 sec: 208529.5). Total num frames: 19136512. Throughput: 0: 248421.5. Samples: 19105792. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0)
[2023-01-11 19:13:28,040][457818] Avg episode reward: [(0, '14243.232')]
[2023-01-11 19:13:28,047][457818] Saving new best policy, reward=14243.232!
[2023-01-11 19:13:33,039][457818] Fps is (10 sec: 249112.8, 60 sec: 248143.4, 300 sec: 210624.3). Total num frames: 20381696. Throughput: 0: 248906.1. Samples: 19857408. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0)
[2023-01-11 19:13:33,040][457818] Avg episode reward: [(0, '14243.232')]
[2023-01-11 19:13:38,042][457818] Fps is (10 sec: 248974.0, 60 sec: 249025.3, 300 sec: 212504.5). Total num frames: 21626880. Throughput: 0: 248860.7. Samples: 21356544. Policy #0 lag: (min: 4.0, avg: 4.0, max: 4.0)
[2023-01-11 19:13:38,043][457818] Avg episode reward: [(0, '14880.855')]
[2023-01-11 19:13:38,045][457818] Saving new best policy, reward=14880.855!
[2023-01-11 19:13:43,040][457818] Fps is (10 sec: 249005.7, 60 sec: 249051.6, 300 sec: 214219.6). Total num frames: 22872064. Throughput: 0: 249277.7. Samples: 22867968. Policy #0 lag: (min: 4.0, avg: 4.0, max: 4.0)
[2023-01-11 19:13:43,041][457818] Avg episode reward: [(0, '15424.260')]
[2023-01-11 19:13:43,053][457818] Saving ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/04_v083_brax_basic_benchmark_see_2322090_env_halfcheetah_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000003490_22872064.pth...
[2023-01-11 19:13:43,503][457818] Saving new best policy, reward=15424.260!
[2023-01-11 19:13:48,043][457818] Fps is (10 sec: 242479.9, 60 sec: 247934.4, 300 sec: 215186.4). Total num frames: 24051712. Throughput: 0: 246899.0. Samples: 23496704. Policy #0 lag: (min: 4.0, avg: 4.0, max: 4.0)
[2023-01-11 19:13:48,044][457818] Avg episode reward: [(0, '15424.260')]
[2023-01-11 19:13:53,041][457818] Fps is (10 sec: 242469.0, 60 sec: 247955.4, 300 sec: 216639.0). Total num frames: 25296896. Throughput: 0: 247186.2. Samples: 25006080. Policy #0 lag: (min: 4.0, avg: 4.0, max: 4.0)
[2023-01-11 19:13:53,041][457818] Avg episode reward: [(0, '16100.578')]
[2023-01-11 19:13:53,054][457818] Saving new best policy, reward=16100.578!
[2023-01-11 19:13:58,044][457818] Fps is (10 sec: 249005.3, 60 sec: 247939.8, 300 sec: 217964.0). Total num frames: 26542080. Throughput: 0: 246163.0. Samples: 26476544. Policy #0 lag: (min: 7.0, avg: 7.0, max: 7.0)
[2023-01-11 19:13:58,044][457818] Avg episode reward: [(0, '16100.578')]
[2023-01-11 19:14:03,044][457818] Fps is (10 sec: 242413.1, 60 sec: 246848.2, 300 sec: 218672.7). Total num frames: 27721728. Throughput: 0: 246394.3. Samples: 27209728. Policy #0 lag: (min: 7.0, avg: 7.0, max: 7.0)
[2023-01-11 19:14:03,044][457818] Avg episode reward: [(0, '16658.027')]
[2023-01-11 19:14:03,058][457818] Saving new best policy, reward=16658.027!
[2023-01-11 19:14:08,042][457818] Fps is (10 sec: 242525.3, 60 sec: 246858.8, 300 sec: 219827.6). Total num frames: 28966912. Throughput: 0: 245851.0. Samples: 28692480. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:14:08,043][457818] Avg episode reward: [(0, '17339.150')]
[2023-01-11 19:14:08,050][457818] Saving new best policy, reward=17339.150!
[2023-01-11 19:14:13,043][457818] Fps is (10 sec: 249044.6, 60 sec: 246848.8, 300 sec: 220893.2). Total num frames: 30212096. Throughput: 0: 246605.2. Samples: 30203904. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:14:13,044][457818] Avg episode reward: [(0, '17339.150')]
[2023-01-11 19:14:18,042][457818] Fps is (10 sec: 249024.9, 60 sec: 246855.1, 300 sec: 221887.2). Total num frames: 31457280. Throughput: 0: 246104.8. Samples: 30932992. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:14:18,043][457818] Avg episode reward: [(0, '17867.812')]
[2023-01-11 19:14:18,050][457818] Saving new best policy, reward=17867.812!
[2023-01-11 19:14:23,043][457818] Fps is (10 sec: 249051.9, 60 sec: 246849.2, 300 sec: 222811.6). Total num frames: 32702464. Throughput: 0: 246303.8. Samples: 32440320. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:14:23,043][457818] Avg episode reward: [(0, '17867.812')]
[2023-01-11 19:14:28,041][457818] Fps is (10 sec: 249081.8, 60 sec: 246848.8, 300 sec: 223678.7). Total num frames: 33947648. Throughput: 0: 245985.0. Samples: 33937408. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:14:28,041][457818] Avg episode reward: [(0, '18219.049')]
[2023-01-11 19:14:28,048][457818] Saving new best policy, reward=18219.049!
[2023-01-11 19:14:33,043][457818] Fps is (10 sec: 249041.6, 60 sec: 246837.4, 300 sec: 224484.7). Total num frames: 35192832. Throughput: 0: 248262.6. Samples: 34668544. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:14:33,043][457818] Avg episode reward: [(0, '18429.346')]
[2023-01-11 19:14:33,049][457818] Saving new best policy, reward=18429.346!
[2023-01-11 19:14:38,076][457818] Fps is (10 sec: 254685.5, 60 sec: 247804.7, 300 sec: 225601.8). Total num frames: 36503552. Throughput: 0: 248204.4. Samples: 36184064. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:14:38,077][457818] Avg episode reward: [(0, '18429.346')]
[2023-01-11 19:14:43,040][457818] Fps is (10 sec: 249103.3, 60 sec: 246853.5, 300 sec: 225960.5). Total num frames: 37683200. Throughput: 0: 248876.0. Samples: 37675008. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:14:43,041][457818] Avg episode reward: [(0, '18644.486')]
[2023-01-11 19:14:43,052][457818] Saving new best policy, reward=18644.486!
[2023-01-11 19:14:48,042][457818] Fps is (10 sec: 243307.2, 60 sec: 247945.1, 300 sec: 226629.1). Total num frames: 38928384. Throughput: 0: 248907.7. Samples: 38410240. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:14:48,043][457818] Avg episode reward: [(0, '18897.219')]
[2023-01-11 19:14:48,050][457818] Saving new best policy, reward=18897.219!
[2023-01-11 19:14:51,825][457818] Early stopping after 4 epochs (8 sgd steps), loss delta 0.0000010
[2023-01-11 19:14:53,040][457818] Fps is (10 sec: 249043.9, 60 sec: 247949.4, 300 sec: 227266.4). Total num frames: 40173568. Throughput: 0: 249323.1. Samples: 39911424. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:14:53,040][457818] Avg episode reward: [(0, '18897.219')]
[2023-01-11 19:14:58,096][457818] Fps is (10 sec: 254218.6, 60 sec: 248818.9, 300 sec: 228154.6). Total num frames: 41484288. Throughput: 0: 248926.1. Samples: 41418752. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:14:58,097][457818] Avg episode reward: [(0, '19037.414')]
[2023-01-11 19:14:58,099][457818] Saving new best policy, reward=19037.414!
[2023-01-11 19:15:03,042][457818] Fps is (10 sec: 242436.1, 60 sec: 247953.3, 300 sec: 228078.7). Total num frames: 42598400. Throughput: 0: 247722.0. Samples: 42080256. Policy #0 lag: (min: 8.0, avg: 8.0, max: 8.0)
[2023-01-11 19:15:03,042][457818] Avg episode reward: [(0, '19037.414')]
[2023-01-11 19:15:08,042][457818] Fps is (10 sec: 237224.6, 60 sec: 247945.8, 300 sec: 228625.0). Total num frames: 43843584. Throughput: 0: 247540.8. Samples: 43579392. Policy #0 lag: (min: 8.0, avg: 8.0, max: 8.0)
[2023-01-11 19:15:08,042][457818] Avg episode reward: [(0, '19141.273')]
[2023-01-11 19:15:08,049][457818] Saving new best policy, reward=19141.273!
[2023-01-11 19:15:13,043][457818] Fps is (10 sec: 242457.3, 60 sec: 246855.3, 300 sec: 228809.5). Total num frames: 45023232. Throughput: 0: 245112.2. Samples: 44967936. Policy #0 lag: (min: 8.0, avg: 8.0, max: 8.0)
[2023-01-11 19:15:13,043][457818] Avg episode reward: [(0, '19141.273')]
[2023-01-11 19:15:18,042][457818] Fps is (10 sec: 242477.9, 60 sec: 246854.6, 300 sec: 229311.6). Total num frames: 46268416. Throughput: 0: 245672.5. Samples: 45723648. Policy #0 lag: (min: 7.0, avg: 7.0, max: 7.0)
[2023-01-11 19:15:18,043][457818] Avg episode reward: [(0, '19317.938')]
[2023-01-11 19:15:18,049][457818] Saving new best policy, reward=19317.938!
[2023-01-11 19:15:23,041][457818] Fps is (10 sec: 242515.3, 60 sec: 245766.0, 300 sec: 229472.3). Total num frames: 47448064. Throughput: 0: 243582.3. Samples: 47136768. Policy #0 lag: (min: 7.0, avg: 7.0, max: 7.0)
[2023-01-11 19:15:23,042][457818] Avg episode reward: [(0, '19508.840')]
[2023-01-11 19:15:23,057][457818] Saving new best policy, reward=19508.840!
[2023-01-11 19:15:28,044][457818] Fps is (10 sec: 235891.0, 60 sec: 244656.0, 300 sec: 229622.3). Total num frames: 48627712. Throughput: 0: 242008.4. Samples: 48566272. Policy #0 lag: (min: 6.0, avg: 6.0, max: 6.0)
[2023-01-11 19:15:28,044][457818] Avg episode reward: [(0, '19508.840')]
[2023-01-11 19:15:33,041][457818] Fps is (10 sec: 242488.1, 60 sec: 244673.7, 300 sec: 230072.7). Total num frames: 49872896. Throughput: 0: 242853.9. Samples: 49338368. Policy #0 lag: (min: 6.0, avg: 6.0, max: 6.0)
[2023-01-11 19:15:33,042][457818] Avg episode reward: [(0, '19688.805')]
[2023-01-11 19:15:33,053][457818] Saving new best policy, reward=19688.805!
[2023-01-11 19:15:38,040][457818] Fps is (10 sec: 242563.4, 60 sec: 242628.6, 300 sec: 230205.6). Total num frames: 51052544. Throughput: 0: 241524.0. Samples: 50780160. Policy #0 lag: (min: 6.0, avg: 6.0, max: 6.0)
[2023-01-11 19:15:38,041][457818] Avg episode reward: [(0, '19688.805')]
[2023-01-11 19:15:43,039][457818] Fps is (10 sec: 242522.6, 60 sec: 243577.2, 300 sec: 230621.6). Total num frames: 52297728. Throughput: 0: 241787.2. Samples: 52285440. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:15:43,040][457818] Avg episode reward: [(0, '19800.465')]
[2023-01-11 19:15:43,052][457818] Saving ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/04_v083_brax_basic_benchmark_see_2322090_env_halfcheetah_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000007978_52297728.pth...
[2023-01-11 19:15:43,291][457818] Saving new best policy, reward=19800.465!
[2023-01-11 19:15:48,043][457818] Fps is (10 sec: 235874.7, 60 sec: 241389.9, 300 sec: 230450.3). Total num frames: 53411840. Throughput: 0: 240156.7. Samples: 52887552. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:15:48,043][457818] Avg episode reward: [(0, '20117.086')]
[2023-01-11 19:15:48,050][457818] Saving new best policy, reward=20117.086!
[2023-01-11 19:15:53,072][457818] Fps is (10 sec: 235160.0, 60 sec: 241259.9, 300 sec: 230813.9). Total num frames: 54657024. Throughput: 0: 238862.5. Samples: 54335488. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:15:53,073][457818] Avg episode reward: [(0, '20117.086')]
[2023-01-11 19:15:58,041][457818] Fps is (10 sec: 249066.5, 60 sec: 240518.8, 300 sec: 231220.2). Total num frames: 55902208. Throughput: 0: 241670.7. Samples: 55842816. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:15:58,042][457818] Avg episode reward: [(0, '20120.244')]
[2023-01-11 19:15:58,049][457818] Saving new best policy, reward=20120.244!
[2023-01-11 19:16:02,787][457818] Early stopping after 2 epochs (4 sgd steps), loss delta 0.0000008
[2023-01-11 19:16:03,048][457818] Fps is (10 sec: 249643.5, 60 sec: 242457.6, 300 sec: 231575.1). Total num frames: 57147392. Throughput: 0: 241358.9. Samples: 56586240. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:16:03,049][457818] Avg episode reward: [(0, '20120.244')]
[2023-01-11 19:16:08,086][457818] Fps is (10 sec: 247928.1, 60 sec: 242303.9, 300 sec: 231886.7). Total num frames: 58392576. Throughput: 0: 242742.2. Samples: 58071040. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:16:08,087][457818] Avg episode reward: [(0, '20412.082')]
[2023-01-11 19:16:08,089][457818] Saving new best policy, reward=20412.082!
[2023-01-11 19:16:13,060][457818] Fps is (10 sec: 248745.2, 60 sec: 243506.5, 300 sec: 232244.5). Total num frames: 59637760. Throughput: 0: 244489.4. Samples: 59572224. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:16:13,060][457818] Avg episode reward: [(0, '20515.316')]
[2023-01-11 19:16:13,069][457818] Saving new best policy, reward=20515.316!
[2023-01-11 19:16:18,091][457818] Fps is (10 sec: 248923.0, 60 sec: 243377.7, 300 sec: 232537.7). Total num frames: 60882944. Throughput: 0: 243534.7. Samples: 60309504. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:16:18,092][457818] Avg episode reward: [(0, '20515.316')]
[2023-01-11 19:16:23,042][457818] Fps is (10 sec: 249467.3, 60 sec: 244663.4, 300 sec: 232889.0). Total num frames: 62128128. Throughput: 0: 245293.3. Samples: 61818880. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:16:23,044][457818] Avg episode reward: [(0, '20655.256')]
[2023-01-11 19:16:23,058][457818] Saving new best policy, reward=20655.256!
[2023-01-11 19:16:28,043][457818] Fps is (10 sec: 250227.3, 60 sec: 245761.9, 300 sec: 233185.5). Total num frames: 63373312. Throughput: 0: 245285.2. Samples: 63324160. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:16:28,044][457818] Avg episode reward: [(0, '20655.256')]
[2023-01-11 19:16:33,041][457818] Fps is (10 sec: 249067.2, 60 sec: 245759.8, 300 sec: 233473.5). Total num frames: 64618496. Throughput: 0: 248680.6. Samples: 64077824. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:16:33,042][457818] Avg episode reward: [(0, '20727.250')]
[2023-01-11 19:16:33,053][457818] Saving new best policy, reward=20727.250!
[2023-01-11 19:16:38,043][457818] Fps is (10 sec: 249038.3, 60 sec: 246840.8, 300 sec: 233748.1). Total num frames: 65863680. Throughput: 0: 250200.2. Samples: 65587200. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:16:38,044][457818] Avg episode reward: [(0, '20896.125')]
[2023-01-11 19:16:38,050][457818] Saving new best policy, reward=20896.125!
[2023-01-11 19:16:43,039][457818] Fps is (10 sec: 249082.6, 60 sec: 246853.0, 300 sec: 234017.7). Total num frames: 67108864. Throughput: 0: 249776.5. Samples: 67082240. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:16:43,040][457818] Avg episode reward: [(0, '20896.125')]
[2023-01-11 19:16:48,042][457818] Fps is (10 sec: 249052.9, 60 sec: 249037.6, 300 sec: 234272.6). Total num frames: 68354048. Throughput: 0: 249841.1. Samples: 67827712. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:16:48,043][457818] Avg episode reward: [(0, '21059.682')]
[2023-01-11 19:16:48,050][457818] Saving new best policy, reward=21059.682!
[2023-01-11 19:16:53,087][457818] Fps is (10 sec: 247850.0, 60 sec: 248974.6, 300 sec: 245773.7). Total num frames: 69599232. Throughput: 0: 249030.8. Samples: 69277696. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:16:53,088][457818] Avg episode reward: [(0, '21059.682')]
[2023-01-11 19:16:58,042][457818] Fps is (10 sec: 242488.8, 60 sec: 247941.3, 300 sec: 245650.5). Total num frames: 70778880. Throughput: 0: 248769.3. Samples: 70762496. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:16:58,043][457818] Avg episode reward: [(0, '21069.305')]
[2023-01-11 19:16:58,049][457818] Saving new best policy, reward=21069.305!
[2023-01-11 19:17:03,042][457818] Fps is (10 sec: 243572.6, 60 sec: 247967.0, 300 sec: 245711.7). Total num frames: 72024064. Throughput: 0: 248939.4. Samples: 71499776. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:17:03,043][457818] Avg episode reward: [(0, '21291.115')]
[2023-01-11 19:17:03,050][457818] Saving new best policy, reward=21291.115!
[2023-01-11 19:17:08,045][457818] Fps is (10 sec: 248975.5, 60 sec: 248116.1, 300 sec: 246834.2). Total num frames: 73269248. Throughput: 0: 247841.1. Samples: 72972288. Policy #0 lag: (min: 5.0, avg: 5.0, max: 5.0)
[2023-01-11 19:17:08,045][457818] Avg episode reward: [(0, '21291.115')]
[2023-01-11 19:17:13,042][457818] Fps is (10 sec: 249056.6, 60 sec: 248018.8, 300 sec: 246816.7). Total num frames: 74514432. Throughput: 0: 247770.4. Samples: 74473472. Policy #0 lag: (min: 5.0, avg: 5.0, max: 5.0)
[2023-01-11 19:17:13,042][457818] Avg episode reward: [(0, '21271.732')]
[2023-01-11 19:17:18,041][457818] Fps is (10 sec: 249125.4, 60 sec: 248149.7, 300 sec: 246815.4). Total num frames: 75759616. Throughput: 0: 247717.4. Samples: 75225088. Policy #0 lag: (min: 5.0, avg: 5.0, max: 5.0)
[2023-01-11 19:17:18,042][457818] Avg episode reward: [(0, '21271.732')]
[2023-01-11 19:17:23,040][457818] Fps is (10 sec: 249083.2, 60 sec: 247955.1, 300 sec: 246814.8). Total num frames: 77004800. Throughput: 0: 247188.6. Samples: 76709888. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:17:23,040][457818] Avg episode reward: [(0, '21196.262')]
[2023-01-11 19:17:28,043][457818] Fps is (10 sec: 248985.7, 60 sec: 247944.4, 300 sec: 246629.8). Total num frames: 78249984. Throughput: 0: 246831.4. Samples: 78190592. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:17:28,044][457818] Avg episode reward: [(0, '21287.096')]
[2023-01-11 19:17:33,079][457818] Fps is (10 sec: 248060.4, 60 sec: 247787.5, 300 sec: 246782.1). Total num frames: 79495168. Throughput: 0: 246605.2. Samples: 78934016. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:17:33,080][457818] Avg episode reward: [(0, '21287.096')]
[2023-01-11 19:17:38,091][457818] Fps is (10 sec: 247857.5, 60 sec: 247747.7, 300 sec: 246776.0). Total num frames: 80740352. Throughput: 0: 247561.1. Samples: 80418816. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:17:38,091][457818] Avg episode reward: [(0, '21565.305')]
[2023-01-11 19:17:38,094][457818] Saving new best policy, reward=21565.305!
[2023-01-11 19:17:43,044][457818] Fps is (10 sec: 249925.5, 60 sec: 247926.8, 300 sec: 246812.2). Total num frames: 81985536. Throughput: 0: 247936.6. Samples: 81920000. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:17:43,044][457818] Avg episode reward: [(0, '21661.523')]
[2023-01-11 19:17:43,053][457818] Saving ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/04_v083_brax_basic_benchmark_see_2322090_env_halfcheetah_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000012502_81985536.pth...
[2023-01-11 19:17:43,067][457818] Removing ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/04_v083_brax_basic_benchmark_see_2322090_env_halfcheetah_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000003490_22872064.pth
[2023-01-11 19:17:43,069][457818] Saving new best policy, reward=21661.523!
[2023-01-11 19:17:48,043][457818] Fps is (10 sec: 250231.8, 60 sec: 247942.2, 300 sec: 246815.6). Total num frames: 83230720. Throughput: 0: 248306.0. Samples: 82673664. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:17:48,044][457818] Avg episode reward: [(0, '21661.523')]
[2023-01-11 19:17:53,076][457818] Fps is (10 sec: 248241.7, 60 sec: 247992.3, 300 sec: 246787.6). Total num frames: 84475904. Throughput: 0: 248546.9. Samples: 84164608. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:17:53,076][457818] Avg episode reward: [(0, '21755.340')]
[2023-01-11 19:17:53,085][457818] Saving new best policy, reward=21755.340!
[2023-01-11 19:17:58,039][457818] Fps is (10 sec: 242578.1, 60 sec: 247957.5, 300 sec: 246596.2). Total num frames: 85655552. Throughput: 0: 248368.7. Samples: 85649408. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:17:58,040][457818] Avg episode reward: [(0, '21755.340')]
[2023-01-11 19:18:03,090][457818] Fps is (10 sec: 248681.0, 60 sec: 248839.9, 300 sec: 246776.5). Total num frames: 86966272. Throughput: 0: 247721.0. Samples: 86384640. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:18:03,091][457818] Avg episode reward: [(0, '21714.219')]
[2023-01-11 19:18:08,040][457818] Fps is (10 sec: 249000.9, 60 sec: 247961.7, 300 sec: 246594.8). Total num frames: 88145920. Throughput: 0: 248304.9. Samples: 87883776. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:18:08,041][457818] Avg episode reward: [(0, '21863.023')]
[2023-01-11 19:18:08,048][457818] Saving new best policy, reward=21863.023!
[2023-01-11 19:18:13,042][457818] Fps is (10 sec: 243639.5, 60 sec: 247941.2, 300 sec: 246593.7). Total num frames: 89391104. Throughput: 0: 248130.1. Samples: 89356288. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:18:13,043][457818] Avg episode reward: [(0, '21863.023')]
[2023-01-11 19:18:18,043][457818] Fps is (10 sec: 248970.5, 60 sec: 247936.0, 300 sec: 246592.1). Total num frames: 90636288. Throughput: 0: 248689.8. Samples: 90116096. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:18:18,044][457818] Avg episode reward: [(0, '21893.676')]
[2023-01-11 19:18:18,046][457818] Saving new best policy, reward=21893.676!
[2023-01-11 19:18:23,078][457818] Fps is (10 sec: 254683.4, 60 sec: 248878.0, 300 sec: 246783.2). Total num frames: 91947008. Throughput: 0: 249061.0. Samples: 91623424. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
[2023-01-11 19:18:23,079][457818] Avg episode reward: [(0, '21893.676')]
[2023-01-11 19:18:28,043][457818] Fps is (10 sec: 249050.8, 60 sec: 247946.8, 300 sec: 246590.1). Total num frames: 93126656. Throughput: 0: 249042.5. Samples: 93126656. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:18:28,043][457818] Avg episode reward: [(0, '21788.742')]
[2023-01-11 19:18:33,041][457818] Fps is (10 sec: 243376.3, 60 sec: 248100.7, 300 sec: 246593.9). Total num frames: 94371840. Throughput: 0: 248317.1. Samples: 93847552. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:18:33,042][457818] Avg episode reward: [(0, '22082.066')]
[2023-01-11 19:18:33,054][457818] Saving new best policy, reward=22082.066!
[2023-01-11 19:18:38,042][457818] Fps is (10 sec: 249041.1, 60 sec: 248144.3, 300 sec: 246591.2). Total num frames: 95617024. Throughput: 0: 248856.5. Samples: 95354880. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:18:38,043][457818] Avg episode reward: [(0, '22082.066')]
[2023-01-11 19:18:43,040][457818] Fps is (10 sec: 249063.0, 60 sec: 247958.0, 300 sec: 246817.0). Total num frames: 96862208. Throughput: 0: 248893.0. Samples: 96849920. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:18:43,041][457818] Avg episode reward: [(0, '22204.691')]
[2023-01-11 19:18:43,055][457818] Saving new best policy, reward=22204.691!
[2023-01-11 19:18:48,044][457818] Fps is (10 sec: 255562.0, 60 sec: 249034.4, 300 sec: 247035.1). Total num frames: 98172928. Throughput: 0: 249977.4. Samples: 97622016. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:18:48,044][457818] Avg episode reward: [(0, '22204.691')]
[2023-01-11 19:18:53,041][457818] Fps is (10 sec: 255574.3, 60 sec: 249180.7, 300 sec: 247039.7). Total num frames: 99418112. Throughput: 0: 249625.6. Samples: 99117056. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
[2023-01-11 19:18:53,042][457818] Avg episode reward: [(0, '22253.053')]
[2023-01-11 19:18:53,054][457818] Saving new best policy, reward=22253.053!
[2023-01-11 19:18:55,695][457818] Saving ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/04_v083_brax_basic_benchmark_see_2322090_env_halfcheetah_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000015262_100073472.pth...
[2023-01-11 19:18:55,729][457818] Removing ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/04_v083_brax_basic_benchmark_see_2322090_env_halfcheetah_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000007978_52297728.pth
[2023-01-11 19:18:55,731][457818] Stopping Batcher_0...
[2023-01-11 19:18:55,732][457818] Stopping InferenceWorker_p0-w0...
[2023-01-11 19:18:55,732][457818] Stopping RolloutWorker_w0...
[2023-01-11 19:18:55,732][457818] Component Batcher_0 stopped!
[2023-01-11 19:18:55,733][457818] Saving ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/04_v083_brax_basic_benchmark_see_2322090_env_halfcheetah_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000015262_100073472.pth...
[2023-01-11 19:18:55,748][457818] Stopping LearnerWorker_p0...
[2023-01-11 19:18:55,749][457818] Component InferenceWorker_p0-w0 stopped!
[2023-01-11 19:18:55,749][457818] Component RolloutWorker_w0 stopped!
[2023-01-11 19:18:55,749][457818] Component LearnerWorker_p0 stopped!
[2023-01-11 19:18:55,749][457818] Batcher 0 profile tree view:
batching: 0.3653, releasing_batches: 0.0719
[2023-01-11 19:18:55,749][457818] InferenceWorker_p0-w0 profile tree view:
update_model: 0.4851
one_step: 0.0012
  handle_policy_step: 60.9075
    deserialize: 0.5112, stack: 0.0665, obs_to_device_normalize: 10.8553, forward: 38.7097, prepare_outputs: 6.6728, send_messages: 0.8327
[2023-01-11 19:18:55,750][457818] Learner 0 profile tree view:
misc: 0.0056, prepare_batch: 5.7227
train: 89.4918
  epoch_init: 0.0671, minibatch_init: 1.0447, losses_postprocess: 2.4470, kl_divergence: 5.8758, after_optimizer: 0.3590
  calculate_losses: 18.5979
    losses_init: 0.0364, forward_head: 3.0156, bptt_initial: 0.1332, bptt: 0.1403, tail: 9.1234, advantages_returns: 1.1915, losses: 3.6083
  update: 59.1597
    clip: 8.9046
[2023-01-11 19:18:55,750][457818] RolloutWorker_w0 profile tree view:
wait_for_trajectories: 0.0901, enqueue_policy_requests: 5.6020, process_policy_outputs: 3.4572, env_step: 223.0124, finalize_trajectories: 0.1451, complete_rollouts: 0.0697
post_env_step: 14.4411
  process_env_step: 2.8677
[2023-01-11 19:18:55,750][457818] Loop Runner_EvtLoop terminating...
[2023-01-11 19:18:55,750][457818] Runner profile tree view:
main_loop: 427.0721
[2023-01-11 19:18:55,751][457818] Collected {0: 100073472}, FPS: 234324.5