diff --git "a/sf_log.txt" "b/sf_log.txt" new file mode 100644--- /dev/null +++ "b/sf_log.txt" @@ -0,0 +1,2341 @@ +[2023-02-22 18:30:14,984][01245] Saving configuration to /content/train_dir/default_experiment/config.json... +[2023-02-22 18:30:14,987][01245] Rollout worker 0 uses device cpu +[2023-02-22 18:30:14,989][01245] Rollout worker 1 uses device cpu +[2023-02-22 18:30:14,992][01245] Rollout worker 2 uses device cpu +[2023-02-22 18:30:14,994][01245] Rollout worker 3 uses device cpu +[2023-02-22 18:30:14,997][01245] Rollout worker 4 uses device cpu +[2023-02-22 18:30:14,999][01245] Rollout worker 5 uses device cpu +[2023-02-22 18:30:15,000][01245] Rollout worker 6 uses device cpu +[2023-02-22 18:30:15,002][01245] Rollout worker 7 uses device cpu +[2023-02-22 18:30:15,191][01245] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-22 18:30:15,192][01245] InferenceWorker_p0-w0: min num requests: 2 +[2023-02-22 18:30:15,224][01245] Starting all processes... +[2023-02-22 18:30:15,226][01245] Starting process learner_proc0 +[2023-02-22 18:30:15,278][01245] Starting all processes... +[2023-02-22 18:30:15,290][01245] Starting process inference_proc0-0 +[2023-02-22 18:30:15,290][01245] Starting process rollout_proc0 +[2023-02-22 18:30:15,292][01245] Starting process rollout_proc1 +[2023-02-22 18:30:15,292][01245] Starting process rollout_proc2 +[2023-02-22 18:30:15,292][01245] Starting process rollout_proc3 +[2023-02-22 18:30:15,292][01245] Starting process rollout_proc4 +[2023-02-22 18:30:15,292][01245] Starting process rollout_proc5 +[2023-02-22 18:30:15,292][01245] Starting process rollout_proc6 +[2023-02-22 18:30:15,292][01245] Starting process rollout_proc7 +[2023-02-22 18:30:26,318][15062] Worker 4 uses CPU cores [0] +[2023-02-22 18:30:26,608][15039] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-22 18:30:26,612][15039] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0 +[2023-02-22 18:30:26,860][15059] Worker 1 uses CPU cores [1] +[2023-02-22 18:30:26,893][15058] Worker 0 uses CPU cores [0] +[2023-02-22 18:30:27,034][15064] Worker 6 uses CPU cores [0] +[2023-02-22 18:30:26,984][15063] Worker 3 uses CPU cores [1] +[2023-02-22 18:30:27,039][15060] Worker 2 uses CPU cores [0] +[2023-02-22 18:30:27,144][15057] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-22 18:30:27,146][15057] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0 +[2023-02-22 18:30:27,200][15065] Worker 7 uses CPU cores [1] +[2023-02-22 18:30:27,201][15061] Worker 5 uses CPU cores [1] +[2023-02-22 18:30:27,592][15039] Num visible devices: 1 +[2023-02-22 18:30:27,592][15057] Num visible devices: 1 +[2023-02-22 18:30:27,608][15039] Starting seed is not provided +[2023-02-22 18:30:27,608][15039] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-22 18:30:27,608][15039] Initializing actor-critic model on device cuda:0 +[2023-02-22 18:30:27,608][15039] RunningMeanStd input shape: (3, 72, 128) +[2023-02-22 18:30:27,610][15039] RunningMeanStd input shape: (1,) +[2023-02-22 18:30:27,623][15039] ConvEncoder: input_channels=3 +[2023-02-22 18:30:27,906][15039] Conv encoder output size: 512 +[2023-02-22 18:30:27,906][15039] Policy head output size: 512 +[2023-02-22 18:30:27,958][15039] Created Actor Critic model with architecture: +[2023-02-22 18:30:27,958][15039] 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) + ) +) +[2023-02-22 18:30:35,015][15039] Using optimizer +[2023-02-22 18:30:35,016][15039] No checkpoints found +[2023-02-22 18:30:35,017][15039] Did not load from checkpoint, starting from scratch! +[2023-02-22 18:30:35,017][15039] Initialized policy 0 weights for model version 0 +[2023-02-22 18:30:35,020][15039] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-22 18:30:35,031][15039] LearnerWorker_p0 finished initialization! +[2023-02-22 18:30:35,187][01245] Heartbeat connected on LearnerWorker_p0 +[2023-02-22 18:30:35,204][01245] Heartbeat connected on RolloutWorker_w0 +[2023-02-22 18:30:35,209][01245] Heartbeat connected on RolloutWorker_w1 +[2023-02-22 18:30:35,211][01245] Heartbeat connected on RolloutWorker_w2 +[2023-02-22 18:30:35,213][01245] Heartbeat connected on RolloutWorker_w3 +[2023-02-22 18:30:35,215][01245] Heartbeat connected on RolloutWorker_w4 +[2023-02-22 18:30:35,217][01245] Heartbeat connected on RolloutWorker_w5 +[2023-02-22 18:30:35,223][01245] Heartbeat connected on RolloutWorker_w7 +[2023-02-22 18:30:35,226][01245] Heartbeat connected on RolloutWorker_w6 +[2023-02-22 18:30:35,239][01245] Heartbeat connected on Batcher_0 +[2023-02-22 18:30:35,255][15057] RunningMeanStd input shape: (3, 72, 128) +[2023-02-22 18:30:35,256][15057] RunningMeanStd input shape: (1,) +[2023-02-22 18:30:35,268][15057] ConvEncoder: input_channels=3 +[2023-02-22 18:30:35,365][15057] Conv encoder output size: 512 +[2023-02-22 18:30:35,365][15057] Policy head output size: 512 +[2023-02-22 18:30:35,594][01245] 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) +[2023-02-22 18:30:38,287][01245] Inference worker 0-0 is ready! +[2023-02-22 18:30:38,293][01245] All inference workers are ready! Signal rollout workers to start! +[2023-02-22 18:30:38,297][01245] Heartbeat connected on InferenceWorker_p0-w0 +[2023-02-22 18:30:38,454][15065] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:30:38,468][15063] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:30:38,478][15061] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:30:38,498][15059] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:30:38,521][15060] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:30:38,562][15062] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:30:38,597][15064] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:30:38,605][15058] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:30:40,128][15064] Decorrelating experience for 0 frames... +[2023-02-22 18:30:40,132][15062] Decorrelating experience for 0 frames... +[2023-02-22 18:30:40,242][15061] Decorrelating experience for 0 frames... +[2023-02-22 18:30:40,248][15065] Decorrelating experience for 0 frames... +[2023-02-22 18:30:40,250][15063] Decorrelating experience for 0 frames... +[2023-02-22 18:30:40,258][15059] Decorrelating experience for 0 frames... +[2023-02-22 18:30:40,594][01245] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 0.0. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) +[2023-02-22 18:30:41,384][15062] Decorrelating experience for 32 frames... +[2023-02-22 18:30:41,632][15064] Decorrelating experience for 32 frames... +[2023-02-22 18:30:41,637][15060] Decorrelating experience for 0 frames... +[2023-02-22 18:30:41,697][15061] Decorrelating experience for 32 frames... +[2023-02-22 18:30:41,699][15065] Decorrelating experience for 32 frames... +[2023-02-22 18:30:41,703][15063] Decorrelating experience for 32 frames... +[2023-02-22 18:30:42,708][15058] Decorrelating experience for 0 frames... +[2023-02-22 18:30:42,715][15062] Decorrelating experience for 64 frames... +[2023-02-22 18:30:42,864][15064] Decorrelating experience for 64 frames... +[2023-02-22 18:30:42,881][15059] Decorrelating experience for 32 frames... +[2023-02-22 18:30:43,050][15061] Decorrelating experience for 64 frames... +[2023-02-22 18:30:43,096][15063] Decorrelating experience for 64 frames... +[2023-02-22 18:30:43,781][15058] Decorrelating experience for 32 frames... +[2023-02-22 18:30:43,887][15062] Decorrelating experience for 96 frames... +[2023-02-22 18:30:44,038][15060] Decorrelating experience for 32 frames... +[2023-02-22 18:30:44,356][15065] Decorrelating experience for 64 frames... +[2023-02-22 18:30:44,525][15061] Decorrelating experience for 96 frames... +[2023-02-22 18:30:44,611][15063] Decorrelating experience for 96 frames... +[2023-02-22 18:30:45,118][15058] Decorrelating experience for 64 frames... +[2023-02-22 18:30:45,135][15059] Decorrelating experience for 64 frames... +[2023-02-22 18:30:45,165][15064] Decorrelating experience for 96 frames... +[2023-02-22 18:30:45,483][15065] Decorrelating experience for 96 frames... +[2023-02-22 18:30:45,594][01245] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 0.0. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) +[2023-02-22 18:30:45,796][15059] Decorrelating experience for 96 frames... +[2023-02-22 18:30:46,132][15060] Decorrelating experience for 64 frames... +[2023-02-22 18:30:46,163][15058] Decorrelating experience for 96 frames... +[2023-02-22 18:30:46,546][15060] Decorrelating experience for 96 frames... +[2023-02-22 18:30:50,549][15039] Signal inference workers to stop experience collection... +[2023-02-22 18:30:50,556][15057] InferenceWorker_p0-w0: stopping experience collection +[2023-02-22 18:30:50,594][01245] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 138.5. Samples: 2078. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) +[2023-02-22 18:30:50,600][01245] Avg episode reward: [(0, '1.678')] +[2023-02-22 18:30:53,343][15039] Signal inference workers to resume experience collection... +[2023-02-22 18:30:53,344][15057] InferenceWorker_p0-w0: resuming experience collection +[2023-02-22 18:30:55,594][01245] Fps is (10 sec: 409.6, 60 sec: 204.8, 300 sec: 204.8). Total num frames: 4096. Throughput: 0: 113.9. Samples: 2278. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0) +[2023-02-22 18:30:55,596][01245] Avg episode reward: [(0, '2.358')] +[2023-02-22 18:31:00,594][01245] Fps is (10 sec: 2457.6, 60 sec: 983.0, 300 sec: 983.0). Total num frames: 24576. Throughput: 0: 226.6. Samples: 5664. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0) +[2023-02-22 18:31:00,599][01245] Avg episode reward: [(0, '3.545')] +[2023-02-22 18:31:05,088][15057] Updated weights for policy 0, policy_version 10 (0.0030) +[2023-02-22 18:31:05,598][01245] Fps is (10 sec: 3684.7, 60 sec: 1365.1, 300 sec: 1365.1). Total num frames: 40960. Throughput: 0: 356.0. Samples: 10682. Policy #0 lag: (min: 0.0, avg: 0.1, max: 1.0) +[2023-02-22 18:31:05,601][01245] Avg episode reward: [(0, '3.873')] +[2023-02-22 18:31:10,597][01245] Fps is (10 sec: 2866.2, 60 sec: 1521.2, 300 sec: 1521.2). Total num frames: 53248. Throughput: 0: 370.6. Samples: 12974. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:31:10,601][01245] Avg episode reward: [(0, '4.477')] +[2023-02-22 18:31:15,594][01245] Fps is (10 sec: 2868.6, 60 sec: 1740.8, 300 sec: 1740.8). Total num frames: 69632. Throughput: 0: 425.4. Samples: 17014. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0) +[2023-02-22 18:31:15,596][01245] Avg episode reward: [(0, '4.623')] +[2023-02-22 18:31:19,600][15057] Updated weights for policy 0, policy_version 20 (0.0059) +[2023-02-22 18:31:20,594][01245] Fps is (10 sec: 2868.2, 60 sec: 1820.4, 300 sec: 1820.4). Total num frames: 81920. Throughput: 0: 474.1. Samples: 21334. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0) +[2023-02-22 18:31:20,596][01245] Avg episode reward: [(0, '4.567')] +[2023-02-22 18:31:25,594][01245] Fps is (10 sec: 3686.4, 60 sec: 2129.9, 300 sec: 2129.9). Total num frames: 106496. Throughput: 0: 553.5. Samples: 24906. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:31:25,597][01245] Avg episode reward: [(0, '4.389')] +[2023-02-22 18:31:25,606][15039] Saving new best policy, reward=4.389! +[2023-02-22 18:31:28,621][15057] Updated weights for policy 0, policy_version 30 (0.0028) +[2023-02-22 18:31:30,594][01245] Fps is (10 sec: 4915.2, 60 sec: 2383.1, 300 sec: 2383.1). Total num frames: 131072. Throughput: 0: 712.3. Samples: 32052. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:31:30,596][01245] Avg episode reward: [(0, '4.475')] +[2023-02-22 18:31:30,601][15039] Saving new best policy, reward=4.475! +[2023-02-22 18:31:35,594][01245] Fps is (10 sec: 3686.4, 60 sec: 2389.3, 300 sec: 2389.3). Total num frames: 143360. Throughput: 0: 771.9. Samples: 36812. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:31:35,596][01245] Avg episode reward: [(0, '4.472')] +[2023-02-22 18:31:40,594][01245] Fps is (10 sec: 2867.2, 60 sec: 2662.4, 300 sec: 2457.6). Total num frames: 159744. Throughput: 0: 814.5. Samples: 38930. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:31:40,600][01245] Avg episode reward: [(0, '4.479')] +[2023-02-22 18:31:40,609][15039] Saving new best policy, reward=4.479! +[2023-02-22 18:31:41,134][15057] Updated weights for policy 0, policy_version 40 (0.0028) +[2023-02-22 18:31:45,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3072.0, 300 sec: 2633.1). Total num frames: 184320. Throughput: 0: 879.2. Samples: 45226. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:31:45,600][01245] Avg episode reward: [(0, '4.337')] +[2023-02-22 18:31:49,875][15057] Updated weights for policy 0, policy_version 50 (0.0022) +[2023-02-22 18:31:50,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3413.3, 300 sec: 2730.7). Total num frames: 204800. Throughput: 0: 920.2. Samples: 52088. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:31:50,597][01245] Avg episode reward: [(0, '4.334')] +[2023-02-22 18:31:55,594][01245] Fps is (10 sec: 3686.3, 60 sec: 3618.1, 300 sec: 2764.8). Total num frames: 221184. Throughput: 0: 919.0. Samples: 54328. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:31:55,600][01245] Avg episode reward: [(0, '4.377')] +[2023-02-22 18:32:00,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 2794.9). Total num frames: 237568. Throughput: 0: 930.0. Samples: 58864. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:32:00,599][01245] Avg episode reward: [(0, '4.376')] +[2023-02-22 18:32:02,319][15057] Updated weights for policy 0, policy_version 60 (0.0014) +[2023-02-22 18:32:05,594][01245] Fps is (10 sec: 3686.5, 60 sec: 3618.4, 300 sec: 2867.2). Total num frames: 258048. Throughput: 0: 984.8. Samples: 65648. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:32:05,600][01245] Avg episode reward: [(0, '4.569')] +[2023-02-22 18:32:05,607][15039] Saving new best policy, reward=4.569! +[2023-02-22 18:32:10,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3823.2, 300 sec: 2975.0). Total num frames: 282624. Throughput: 0: 982.3. Samples: 69110. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:32:10,596][01245] Avg episode reward: [(0, '4.601')] +[2023-02-22 18:32:10,612][15039] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000069_282624.pth... +[2023-02-22 18:32:10,730][15039] Saving new best policy, reward=4.601! +[2023-02-22 18:32:11,711][15057] Updated weights for policy 0, policy_version 70 (0.0024) +[2023-02-22 18:32:15,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3754.7, 300 sec: 2949.1). Total num frames: 294912. Throughput: 0: 935.1. Samples: 74132. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:32:15,597][01245] Avg episode reward: [(0, '4.507')] +[2023-02-22 18:32:20,594][01245] Fps is (10 sec: 2457.6, 60 sec: 3754.7, 300 sec: 2925.7). Total num frames: 307200. Throughput: 0: 905.7. Samples: 77568. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:32:20,597][01245] Avg episode reward: [(0, '4.375')] +[2023-02-22 18:32:24,548][15057] Updated weights for policy 0, policy_version 80 (0.0038) +[2023-02-22 18:32:25,595][01245] Fps is (10 sec: 3685.9, 60 sec: 3754.6, 300 sec: 3016.1). Total num frames: 331776. Throughput: 0: 936.3. Samples: 81066. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:32:25,601][01245] Avg episode reward: [(0, '4.316')] +[2023-02-22 18:32:30,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3686.4, 300 sec: 3063.1). Total num frames: 352256. Throughput: 0: 954.2. Samples: 88166. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:32:30,598][01245] Avg episode reward: [(0, '4.466')] +[2023-02-22 18:32:34,650][15057] Updated weights for policy 0, policy_version 90 (0.0040) +[2023-02-22 18:32:35,594][01245] Fps is (10 sec: 3686.9, 60 sec: 3754.7, 300 sec: 3072.0). Total num frames: 368640. Throughput: 0: 909.6. Samples: 93020. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:32:35,598][01245] Avg episode reward: [(0, '4.400')] +[2023-02-22 18:32:40,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3754.7, 300 sec: 3080.2). Total num frames: 385024. Throughput: 0: 910.1. Samples: 95280. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:32:40,597][01245] Avg episode reward: [(0, '4.593')] +[2023-02-22 18:32:45,526][15057] Updated weights for policy 0, policy_version 100 (0.0028) +[2023-02-22 18:32:45,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3754.7, 300 sec: 3150.8). Total num frames: 409600. Throughput: 0: 948.0. Samples: 101526. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:32:45,596][01245] Avg episode reward: [(0, '4.463')] +[2023-02-22 18:32:50,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3754.7, 300 sec: 3185.8). Total num frames: 430080. Throughput: 0: 948.0. Samples: 108306. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:32:50,603][01245] Avg episode reward: [(0, '4.448')] +[2023-02-22 18:32:55,594][01245] Fps is (10 sec: 2867.2, 60 sec: 3618.2, 300 sec: 3130.5). Total num frames: 438272. Throughput: 0: 897.1. Samples: 109480. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:32:55,601][01245] Avg episode reward: [(0, '4.494')] +[2023-02-22 18:32:59,439][15057] Updated weights for policy 0, policy_version 110 (0.0022) +[2023-02-22 18:33:00,594][01245] Fps is (10 sec: 2457.6, 60 sec: 3618.1, 300 sec: 3135.6). Total num frames: 454656. Throughput: 0: 865.1. Samples: 113062. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:33:00,599][01245] Avg episode reward: [(0, '4.739')] +[2023-02-22 18:33:00,610][15039] Saving new best policy, reward=4.739! +[2023-02-22 18:33:05,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3618.1, 300 sec: 3167.6). Total num frames: 475136. Throughput: 0: 941.8. Samples: 119950. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:33:05,599][01245] Avg episode reward: [(0, '4.502')] +[2023-02-22 18:33:08,215][15057] Updated weights for policy 0, policy_version 120 (0.0019) +[2023-02-22 18:33:10,597][01245] Fps is (10 sec: 4504.2, 60 sec: 3618.0, 300 sec: 3223.9). Total num frames: 499712. Throughput: 0: 944.2. Samples: 123556. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) +[2023-02-22 18:33:10,600][01245] Avg episode reward: [(0, '4.562')] +[2023-02-22 18:33:15,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3618.1, 300 sec: 3200.0). Total num frames: 512000. Throughput: 0: 895.0. Samples: 128442. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:33:15,598][01245] Avg episode reward: [(0, '4.671')] +[2023-02-22 18:33:20,594][01245] Fps is (10 sec: 2868.0, 60 sec: 3686.4, 300 sec: 3202.3). Total num frames: 528384. Throughput: 0: 889.9. Samples: 133066. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:33:20,597][01245] Avg episode reward: [(0, '4.528')] +[2023-02-22 18:33:20,906][15057] Updated weights for policy 0, policy_version 130 (0.0021) +[2023-02-22 18:33:25,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3686.5, 300 sec: 3252.7). Total num frames: 552960. Throughput: 0: 913.7. Samples: 136396. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:33:25,596][01245] Avg episode reward: [(0, '4.443')] +[2023-02-22 18:33:30,144][15057] Updated weights for policy 0, policy_version 140 (0.0025) +[2023-02-22 18:33:30,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3686.4, 300 sec: 3276.8). Total num frames: 573440. Throughput: 0: 928.0. Samples: 143284. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:33:30,598][01245] Avg episode reward: [(0, '4.480')] +[2023-02-22 18:33:35,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3618.1, 300 sec: 3254.0). Total num frames: 585728. Throughput: 0: 877.6. Samples: 147798. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:33:35,602][01245] Avg episode reward: [(0, '4.427')] +[2023-02-22 18:33:40,594][01245] Fps is (10 sec: 2867.2, 60 sec: 3618.1, 300 sec: 3254.7). Total num frames: 602112. Throughput: 0: 900.2. Samples: 149990. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:33:40,599][01245] Avg episode reward: [(0, '4.443')] +[2023-02-22 18:33:42,182][15057] Updated weights for policy 0, policy_version 150 (0.0013) +[2023-02-22 18:33:45,597][01245] Fps is (10 sec: 4094.8, 60 sec: 3618.0, 300 sec: 3298.3). Total num frames: 626688. Throughput: 0: 967.0. Samples: 156582. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:33:45,602][01245] Avg episode reward: [(0, '4.514')] +[2023-02-22 18:33:50,594][01245] Fps is (10 sec: 4915.2, 60 sec: 3686.4, 300 sec: 3339.8). Total num frames: 651264. Throughput: 0: 960.2. Samples: 163158. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:33:50,597][01245] Avg episode reward: [(0, '4.537')] +[2023-02-22 18:33:51,999][15057] Updated weights for policy 0, policy_version 160 (0.0011) +[2023-02-22 18:33:55,594][01245] Fps is (10 sec: 3687.5, 60 sec: 3754.7, 300 sec: 3317.8). Total num frames: 663552. Throughput: 0: 927.7. Samples: 165298. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:33:55,599][01245] Avg episode reward: [(0, '4.488')] +[2023-02-22 18:34:00,594][01245] Fps is (10 sec: 2867.2, 60 sec: 3754.7, 300 sec: 3316.8). Total num frames: 679936. Throughput: 0: 916.6. Samples: 169688. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:34:00,599][01245] Avg episode reward: [(0, '4.451')] +[2023-02-22 18:34:03,692][15057] Updated weights for policy 0, policy_version 170 (0.0058) +[2023-02-22 18:34:05,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3354.8). Total num frames: 704512. Throughput: 0: 966.6. Samples: 176564. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:34:05,596][01245] Avg episode reward: [(0, '4.552')] +[2023-02-22 18:34:10,600][01245] Fps is (10 sec: 4503.0, 60 sec: 3754.5, 300 sec: 3372.0). Total num frames: 724992. Throughput: 0: 971.9. Samples: 180138. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:34:10,602][01245] Avg episode reward: [(0, '4.513')] +[2023-02-22 18:34:10,620][15039] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000177_724992.pth... +[2023-02-22 18:34:14,103][15057] Updated weights for policy 0, policy_version 180 (0.0013) +[2023-02-22 18:34:15,596][01245] Fps is (10 sec: 3685.7, 60 sec: 3822.8, 300 sec: 3369.9). Total num frames: 741376. Throughput: 0: 925.2. Samples: 184918. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:34:15,599][01245] Avg episode reward: [(0, '4.557')] +[2023-02-22 18:34:20,598][01245] Fps is (10 sec: 3277.4, 60 sec: 3822.7, 300 sec: 3367.8). Total num frames: 757760. Throughput: 0: 929.0. Samples: 189608. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:34:20,608][01245] Avg episode reward: [(0, '4.737')] +[2023-02-22 18:34:25,030][15057] Updated weights for policy 0, policy_version 190 (0.0023) +[2023-02-22 18:34:25,594][01245] Fps is (10 sec: 3687.1, 60 sec: 3754.7, 300 sec: 3383.7). Total num frames: 778240. Throughput: 0: 956.3. Samples: 193024. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:34:25,597][01245] Avg episode reward: [(0, '4.669')] +[2023-02-22 18:34:30,594][01245] Fps is (10 sec: 4097.6, 60 sec: 3754.7, 300 sec: 3398.8). Total num frames: 798720. Throughput: 0: 968.4. Samples: 200158. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:34:30,600][01245] Avg episode reward: [(0, '4.449')] +[2023-02-22 18:34:35,598][01245] Fps is (10 sec: 3684.7, 60 sec: 3822.6, 300 sec: 3396.2). Total num frames: 815104. Throughput: 0: 923.5. Samples: 204720. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:34:35,601][01245] Avg episode reward: [(0, '4.517')] +[2023-02-22 18:34:36,456][15057] Updated weights for policy 0, policy_version 200 (0.0012) +[2023-02-22 18:34:40,597][01245] Fps is (10 sec: 3275.8, 60 sec: 3822.7, 300 sec: 3393.8). Total num frames: 831488. Throughput: 0: 923.3. Samples: 206850. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:34:40,604][01245] Avg episode reward: [(0, '4.527')] +[2023-02-22 18:34:45,596][01245] Fps is (10 sec: 4096.9, 60 sec: 3823.0, 300 sec: 3424.2). Total num frames: 856064. Throughput: 0: 974.6. Samples: 213546. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:34:45,600][01245] Avg episode reward: [(0, '4.420')] +[2023-02-22 18:34:46,110][15057] Updated weights for policy 0, policy_version 210 (0.0015) +[2023-02-22 18:34:50,598][01245] Fps is (10 sec: 4505.1, 60 sec: 3754.4, 300 sec: 3437.4). Total num frames: 876544. Throughput: 0: 969.5. Samples: 220194. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:34:50,605][01245] Avg episode reward: [(0, '4.510')] +[2023-02-22 18:34:55,594][01245] Fps is (10 sec: 3687.3, 60 sec: 3822.9, 300 sec: 3434.3). Total num frames: 892928. Throughput: 0: 939.1. Samples: 222392. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:34:55,601][01245] Avg episode reward: [(0, '4.594')] +[2023-02-22 18:34:58,150][15057] Updated weights for policy 0, policy_version 220 (0.0013) +[2023-02-22 18:35:00,594][01245] Fps is (10 sec: 3278.2, 60 sec: 3822.9, 300 sec: 3431.4). Total num frames: 909312. Throughput: 0: 934.6. Samples: 226972. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:35:00,596][01245] Avg episode reward: [(0, '4.515')] +[2023-02-22 18:35:05,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3458.8). Total num frames: 933888. Throughput: 0: 988.4. Samples: 234080. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:35:05,602][01245] Avg episode reward: [(0, '4.595')] +[2023-02-22 18:35:06,927][15057] Updated weights for policy 0, policy_version 230 (0.0024) +[2023-02-22 18:35:10,597][01245] Fps is (10 sec: 4503.9, 60 sec: 3823.1, 300 sec: 3470.4). Total num frames: 954368. Throughput: 0: 991.4. Samples: 237640. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:35:10,600][01245] Avg episode reward: [(0, '4.649')] +[2023-02-22 18:35:15,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3823.1, 300 sec: 3467.0). Total num frames: 970752. Throughput: 0: 939.4. Samples: 242432. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:35:15,599][01245] Avg episode reward: [(0, '4.570')] +[2023-02-22 18:35:19,280][15057] Updated weights for policy 0, policy_version 240 (0.0020) +[2023-02-22 18:35:20,594][01245] Fps is (10 sec: 3278.0, 60 sec: 3823.2, 300 sec: 3463.6). Total num frames: 987136. Throughput: 0: 947.7. Samples: 247364. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:35:20,596][01245] Avg episode reward: [(0, '4.721')] +[2023-02-22 18:35:25,594][01245] Fps is (10 sec: 4095.9, 60 sec: 3891.2, 300 sec: 3488.7). Total num frames: 1011712. Throughput: 0: 978.6. Samples: 250884. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:35:25,596][01245] Avg episode reward: [(0, '4.573')] +[2023-02-22 18:35:28,126][15057] Updated weights for policy 0, policy_version 250 (0.0013) +[2023-02-22 18:35:30,598][01245] Fps is (10 sec: 4503.5, 60 sec: 3890.9, 300 sec: 3498.9). Total num frames: 1032192. Throughput: 0: 989.5. Samples: 258076. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:35:30,606][01245] Avg episode reward: [(0, '4.444')] +[2023-02-22 18:35:35,594][01245] Fps is (10 sec: 3276.9, 60 sec: 3823.2, 300 sec: 3540.6). Total num frames: 1044480. Throughput: 0: 941.6. Samples: 262564. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:35:35,600][01245] Avg episode reward: [(0, '4.547')] +[2023-02-22 18:35:40,557][15057] Updated weights for policy 0, policy_version 260 (0.0023) +[2023-02-22 18:35:40,594][01245] Fps is (10 sec: 3278.1, 60 sec: 3891.4, 300 sec: 3610.0). Total num frames: 1064960. Throughput: 0: 942.5. Samples: 264806. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:35:40,598][01245] Avg episode reward: [(0, '4.563')] +[2023-02-22 18:35:45,599][01245] Fps is (10 sec: 3275.0, 60 sec: 3686.2, 300 sec: 3651.6). Total num frames: 1077248. Throughput: 0: 950.0. Samples: 269728. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:35:45,605][01245] Avg episode reward: [(0, '4.542')] +[2023-02-22 18:35:50,594][01245] Fps is (10 sec: 2867.4, 60 sec: 3618.4, 300 sec: 3693.3). Total num frames: 1093632. Throughput: 0: 888.8. Samples: 274078. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:35:50,599][01245] Avg episode reward: [(0, '4.434')] +[2023-02-22 18:35:54,788][15057] Updated weights for policy 0, policy_version 270 (0.0015) +[2023-02-22 18:35:55,594][01245] Fps is (10 sec: 2868.8, 60 sec: 3549.9, 300 sec: 3665.6). Total num frames: 1105920. Throughput: 0: 853.8. Samples: 276060. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:35:55,596][01245] Avg episode reward: [(0, '4.483')] +[2023-02-22 18:36:00,594][01245] Fps is (10 sec: 2867.3, 60 sec: 3549.9, 300 sec: 3665.6). Total num frames: 1122304. Throughput: 0: 844.9. Samples: 280452. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:36:00,600][01245] Avg episode reward: [(0, '4.508')] +[2023-02-22 18:36:05,264][15057] Updated weights for policy 0, policy_version 280 (0.0014) +[2023-02-22 18:36:05,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3549.9, 300 sec: 3707.3). Total num frames: 1146880. Throughput: 0: 888.8. Samples: 287362. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:36:05,596][01245] Avg episode reward: [(0, '4.542')] +[2023-02-22 18:36:10,600][01245] Fps is (10 sec: 4502.6, 60 sec: 3549.7, 300 sec: 3721.0). Total num frames: 1167360. Throughput: 0: 890.1. Samples: 290946. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:36:10,603][01245] Avg episode reward: [(0, '4.764')] +[2023-02-22 18:36:10,619][15039] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000285_1167360.pth... +[2023-02-22 18:36:10,827][15039] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000069_282624.pth +[2023-02-22 18:36:10,851][15039] Saving new best policy, reward=4.764! +[2023-02-22 18:36:15,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3735.0). Total num frames: 1183744. Throughput: 0: 838.8. Samples: 295820. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:36:15,598][01245] Avg episode reward: [(0, '4.643')] +[2023-02-22 18:36:16,587][15057] Updated weights for policy 0, policy_version 290 (0.0013) +[2023-02-22 18:36:20,594][01245] Fps is (10 sec: 3279.0, 60 sec: 3549.9, 300 sec: 3707.2). Total num frames: 1200128. Throughput: 0: 843.3. Samples: 300514. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:36:20,596][01245] Avg episode reward: [(0, '4.469')] +[2023-02-22 18:36:25,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3481.6, 300 sec: 3693.3). Total num frames: 1220608. Throughput: 0: 871.1. Samples: 304006. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:36:25,596][01245] Avg episode reward: [(0, '4.550')] +[2023-02-22 18:36:26,431][15057] Updated weights for policy 0, policy_version 300 (0.0012) +[2023-02-22 18:36:30,594][01245] Fps is (10 sec: 4505.5, 60 sec: 3550.1, 300 sec: 3735.0). Total num frames: 1245184. Throughput: 0: 922.6. Samples: 311238. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:36:30,603][01245] Avg episode reward: [(0, '4.623')] +[2023-02-22 18:36:35,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3618.1, 300 sec: 3735.0). Total num frames: 1261568. Throughput: 0: 930.2. Samples: 315938. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:36:35,601][01245] Avg episode reward: [(0, '4.560')] +[2023-02-22 18:36:38,075][15057] Updated weights for policy 0, policy_version 310 (0.0019) +[2023-02-22 18:36:40,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3707.2). Total num frames: 1277952. Throughput: 0: 935.2. Samples: 318146. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:36:40,602][01245] Avg episode reward: [(0, '4.493')] +[2023-02-22 18:36:45,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3686.7, 300 sec: 3707.2). Total num frames: 1298432. Throughput: 0: 981.7. Samples: 324628. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:36:45,601][01245] Avg episode reward: [(0, '4.590')] +[2023-02-22 18:36:47,373][15057] Updated weights for policy 0, policy_version 320 (0.0018) +[2023-02-22 18:36:50,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3822.9, 300 sec: 3735.0). Total num frames: 1323008. Throughput: 0: 979.3. Samples: 331430. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:36:50,596][01245] Avg episode reward: [(0, '4.483')] +[2023-02-22 18:36:55,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3891.2, 300 sec: 3735.0). Total num frames: 1339392. Throughput: 0: 949.2. Samples: 333652. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:36:55,597][01245] Avg episode reward: [(0, '4.561')] +[2023-02-22 18:36:59,596][15057] Updated weights for policy 0, policy_version 330 (0.0013) +[2023-02-22 18:37:00,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3891.2, 300 sec: 3721.1). Total num frames: 1355776. Throughput: 0: 942.3. Samples: 338224. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:37:00,598][01245] Avg episode reward: [(0, '4.679')] +[2023-02-22 18:37:05,594][01245] Fps is (10 sec: 3686.3, 60 sec: 3822.9, 300 sec: 3707.2). Total num frames: 1376256. Throughput: 0: 991.9. Samples: 345152. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:37:05,597][01245] Avg episode reward: [(0, '4.513')] +[2023-02-22 18:37:08,330][15057] Updated weights for policy 0, policy_version 340 (0.0025) +[2023-02-22 18:37:10,597][01245] Fps is (10 sec: 4503.9, 60 sec: 3891.4, 300 sec: 3748.8). Total num frames: 1400832. Throughput: 0: 991.6. Samples: 348630. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:37:10,600][01245] Avg episode reward: [(0, '4.526')] +[2023-02-22 18:37:15,594][01245] Fps is (10 sec: 3686.5, 60 sec: 3822.9, 300 sec: 3748.9). Total num frames: 1413120. Throughput: 0: 942.7. Samples: 353658. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:37:15,596][01245] Avg episode reward: [(0, '4.554')] +[2023-02-22 18:37:20,594][01245] Fps is (10 sec: 2868.3, 60 sec: 3822.9, 300 sec: 3721.1). Total num frames: 1429504. Throughput: 0: 942.7. Samples: 358360. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:37:20,596][01245] Avg episode reward: [(0, '4.451')] +[2023-02-22 18:37:20,833][15057] Updated weights for policy 0, policy_version 350 (0.0027) +[2023-02-22 18:37:25,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3891.2, 300 sec: 3735.0). Total num frames: 1454080. Throughput: 0: 971.0. Samples: 361840. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:37:25,598][01245] Avg episode reward: [(0, '4.522')] +[2023-02-22 18:37:29,466][15057] Updated weights for policy 0, policy_version 360 (0.0024) +[2023-02-22 18:37:30,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3822.9, 300 sec: 3748.9). Total num frames: 1474560. Throughput: 0: 985.0. Samples: 368954. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:37:30,604][01245] Avg episode reward: [(0, '4.621')] +[2023-02-22 18:37:35,594][01245] Fps is (10 sec: 3686.3, 60 sec: 3822.9, 300 sec: 3748.9). Total num frames: 1490944. Throughput: 0: 936.2. Samples: 373558. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:37:35,598][01245] Avg episode reward: [(0, '4.635')] +[2023-02-22 18:37:40,594][01245] Fps is (10 sec: 3276.7, 60 sec: 3822.9, 300 sec: 3721.1). Total num frames: 1507328. Throughput: 0: 937.5. Samples: 375842. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:37:40,597][01245] Avg episode reward: [(0, '4.643')] +[2023-02-22 18:37:41,721][15057] Updated weights for policy 0, policy_version 370 (0.0016) +[2023-02-22 18:37:45,594][01245] Fps is (10 sec: 4096.1, 60 sec: 3891.2, 300 sec: 3735.0). Total num frames: 1531904. Throughput: 0: 982.1. Samples: 382418. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:37:45,596][01245] Avg episode reward: [(0, '4.661')] +[2023-02-22 18:37:50,594][01245] Fps is (10 sec: 4505.7, 60 sec: 3822.9, 300 sec: 3776.7). Total num frames: 1552384. Throughput: 0: 978.2. Samples: 389170. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:37:50,596][01245] Avg episode reward: [(0, '4.567')] +[2023-02-22 18:37:50,880][15057] Updated weights for policy 0, policy_version 380 (0.0017) +[2023-02-22 18:37:55,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3776.7). Total num frames: 1568768. Throughput: 0: 950.1. Samples: 391382. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:37:55,598][01245] Avg episode reward: [(0, '4.430')] +[2023-02-22 18:38:00,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3822.9, 300 sec: 3762.8). Total num frames: 1585152. Throughput: 0: 937.3. Samples: 395836. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:38:00,601][01245] Avg episode reward: [(0, '4.453')] +[2023-02-22 18:38:02,784][15057] Updated weights for policy 0, policy_version 390 (0.0035) +[2023-02-22 18:38:05,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3891.2, 300 sec: 3762.8). Total num frames: 1609728. Throughput: 0: 989.0. Samples: 402864. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) +[2023-02-22 18:38:05,599][01245] Avg episode reward: [(0, '4.747')] +[2023-02-22 18:38:10,598][01245] Fps is (10 sec: 4503.5, 60 sec: 3822.9, 300 sec: 3790.5). Total num frames: 1630208. Throughput: 0: 990.3. Samples: 406406. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:38:10,601][01245] Avg episode reward: [(0, '4.403')] +[2023-02-22 18:38:10,616][15039] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000398_1630208.pth... +[2023-02-22 18:38:10,797][15039] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000177_724992.pth +[2023-02-22 18:38:12,790][15057] Updated weights for policy 0, policy_version 400 (0.0021) +[2023-02-22 18:38:15,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3891.2, 300 sec: 3790.5). Total num frames: 1646592. Throughput: 0: 940.2. Samples: 411262. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:38:15,596][01245] Avg episode reward: [(0, '4.397')] +[2023-02-22 18:38:20,594][01245] Fps is (10 sec: 3278.3, 60 sec: 3891.2, 300 sec: 3762.8). Total num frames: 1662976. Throughput: 0: 946.9. Samples: 416168. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:38:20,600][01245] Avg episode reward: [(0, '4.760')] +[2023-02-22 18:38:23,839][15057] Updated weights for policy 0, policy_version 410 (0.0017) +[2023-02-22 18:38:25,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3762.8). Total num frames: 1683456. Throughput: 0: 974.8. Samples: 419706. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:38:25,596][01245] Avg episode reward: [(0, '4.992')] +[2023-02-22 18:38:25,684][15039] Saving new best policy, reward=4.992! +[2023-02-22 18:38:30,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3891.2, 300 sec: 3804.4). Total num frames: 1708032. Throughput: 0: 982.5. Samples: 426630. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:38:30,598][01245] Avg episode reward: [(0, '4.989')] +[2023-02-22 18:38:34,685][15057] Updated weights for policy 0, policy_version 420 (0.0027) +[2023-02-22 18:38:35,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3790.5). Total num frames: 1720320. Throughput: 0: 933.6. Samples: 431180. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:38:35,601][01245] Avg episode reward: [(0, '4.941')] +[2023-02-22 18:38:40,597][01245] Fps is (10 sec: 2866.1, 60 sec: 3822.7, 300 sec: 3762.8). Total num frames: 1736704. Throughput: 0: 933.6. Samples: 433398. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:38:40,605][01245] Avg episode reward: [(0, '4.785')] +[2023-02-22 18:38:45,038][15057] Updated weights for policy 0, policy_version 430 (0.0030) +[2023-02-22 18:38:45,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3762.8). Total num frames: 1761280. Throughput: 0: 982.8. Samples: 440064. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:38:45,601][01245] Avg episode reward: [(0, '4.828')] +[2023-02-22 18:38:50,594][01245] Fps is (10 sec: 4507.3, 60 sec: 3822.9, 300 sec: 3790.5). Total num frames: 1781760. Throughput: 0: 973.1. Samples: 446654. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:38:50,597][01245] Avg episode reward: [(0, '4.828')] +[2023-02-22 18:38:55,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3790.5). Total num frames: 1798144. Throughput: 0: 944.5. Samples: 448902. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:38:55,597][01245] Avg episode reward: [(0, '5.184')] +[2023-02-22 18:38:55,603][15039] Saving new best policy, reward=5.184! +[2023-02-22 18:38:56,526][15057] Updated weights for policy 0, policy_version 440 (0.0015) +[2023-02-22 18:39:00,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3822.9, 300 sec: 3762.8). Total num frames: 1814528. Throughput: 0: 936.9. Samples: 453422. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:39:00,602][01245] Avg episode reward: [(0, '5.113')] +[2023-02-22 18:39:05,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3776.7). Total num frames: 1839104. Throughput: 0: 986.0. Samples: 460536. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:39:05,597][01245] Avg episode reward: [(0, '4.682')] +[2023-02-22 18:39:06,031][15057] Updated weights for policy 0, policy_version 450 (0.0022) +[2023-02-22 18:39:10,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3823.2, 300 sec: 3790.6). Total num frames: 1859584. Throughput: 0: 986.8. Samples: 464114. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:39:10,600][01245] Avg episode reward: [(0, '4.479')] +[2023-02-22 18:39:15,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3790.6). Total num frames: 1875968. Throughput: 0: 936.9. Samples: 468790. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:39:15,598][01245] Avg episode reward: [(0, '4.610')] +[2023-02-22 18:39:18,199][15057] Updated weights for policy 0, policy_version 460 (0.0018) +[2023-02-22 18:39:20,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3822.9, 300 sec: 3776.7). Total num frames: 1892352. Throughput: 0: 946.7. Samples: 473780. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:39:20,597][01245] Avg episode reward: [(0, '4.776')] +[2023-02-22 18:39:25,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3891.2, 300 sec: 3790.5). Total num frames: 1916928. Throughput: 0: 974.8. Samples: 477260. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:39:25,597][01245] Avg episode reward: [(0, '4.976')] +[2023-02-22 18:39:27,361][15057] Updated weights for policy 0, policy_version 470 (0.0024) +[2023-02-22 18:39:30,595][01245] Fps is (10 sec: 4505.1, 60 sec: 3822.9, 300 sec: 3804.5). Total num frames: 1937408. Throughput: 0: 979.4. Samples: 484136. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:39:30,602][01245] Avg episode reward: [(0, '4.735')] +[2023-02-22 18:39:35,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3822.9, 300 sec: 3790.6). Total num frames: 1949696. Throughput: 0: 926.8. Samples: 488358. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:39:35,601][01245] Avg episode reward: [(0, '4.805')] +[2023-02-22 18:39:39,846][15057] Updated weights for policy 0, policy_version 480 (0.0022) +[2023-02-22 18:39:40,594][01245] Fps is (10 sec: 2867.5, 60 sec: 3823.2, 300 sec: 3762.8). Total num frames: 1966080. Throughput: 0: 926.3. Samples: 490586. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:39:40,600][01245] Avg episode reward: [(0, '4.965')] +[2023-02-22 18:39:45,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3776.7). Total num frames: 1990656. Throughput: 0: 980.9. Samples: 497564. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:39:45,604][01245] Avg episode reward: [(0, '4.781')] +[2023-02-22 18:39:48,641][15057] Updated weights for policy 0, policy_version 490 (0.0015) +[2023-02-22 18:39:50,601][01245] Fps is (10 sec: 4502.2, 60 sec: 3822.4, 300 sec: 3790.4). Total num frames: 2011136. Throughput: 0: 960.7. Samples: 503776. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:39:50,606][01245] Avg episode reward: [(0, '4.883')] +[2023-02-22 18:39:55,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3790.5). Total num frames: 2027520. Throughput: 0: 931.3. Samples: 506024. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:39:55,600][01245] Avg episode reward: [(0, '4.950')] +[2023-02-22 18:40:00,594][01245] Fps is (10 sec: 3279.3, 60 sec: 3822.9, 300 sec: 3762.8). Total num frames: 2043904. Throughput: 0: 931.0. Samples: 510686. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:40:00,601][01245] Avg episode reward: [(0, '5.082')] +[2023-02-22 18:40:01,072][15057] Updated weights for policy 0, policy_version 500 (0.0021) +[2023-02-22 18:40:05,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3776.7). Total num frames: 2068480. Throughput: 0: 974.2. Samples: 517620. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:40:05,600][01245] Avg episode reward: [(0, '4.937')] +[2023-02-22 18:40:10,212][15057] Updated weights for policy 0, policy_version 510 (0.0011) +[2023-02-22 18:40:10,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3822.9, 300 sec: 3790.5). Total num frames: 2088960. Throughput: 0: 973.7. Samples: 521078. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:40:10,595][01245] Avg episode reward: [(0, '4.708')] +[2023-02-22 18:40:10,611][15039] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000510_2088960.pth... +[2023-02-22 18:40:10,758][15039] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000285_1167360.pth +[2023-02-22 18:40:15,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3754.7, 300 sec: 3776.7). Total num frames: 2101248. Throughput: 0: 922.1. Samples: 525628. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:40:15,596][01245] Avg episode reward: [(0, '4.711')] +[2023-02-22 18:40:20,595][01245] Fps is (10 sec: 2457.2, 60 sec: 3686.3, 300 sec: 3735.0). Total num frames: 2113536. Throughput: 0: 912.2. Samples: 529408. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:40:20,609][01245] Avg episode reward: [(0, '4.459')] +[2023-02-22 18:40:25,399][15057] Updated weights for policy 0, policy_version 520 (0.0012) +[2023-02-22 18:40:25,594][01245] Fps is (10 sec: 2867.2, 60 sec: 3549.9, 300 sec: 3721.2). Total num frames: 2129920. Throughput: 0: 909.2. Samples: 531498. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:40:25,599][01245] Avg episode reward: [(0, '4.419')] +[2023-02-22 18:40:30,594][01245] Fps is (10 sec: 3277.4, 60 sec: 3481.7, 300 sec: 3735.0). Total num frames: 2146304. Throughput: 0: 870.5. Samples: 536736. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:40:30,601][01245] Avg episode reward: [(0, '4.662')] +[2023-02-22 18:40:35,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3721.1). Total num frames: 2162688. Throughput: 0: 837.7. Samples: 541468. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:40:35,598][01245] Avg episode reward: [(0, '4.908')] +[2023-02-22 18:40:37,803][15057] Updated weights for policy 0, policy_version 530 (0.0012) +[2023-02-22 18:40:40,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3735.1). Total num frames: 2179072. Throughput: 0: 837.1. Samples: 543692. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:40:40,600][01245] Avg episode reward: [(0, '4.898')] +[2023-02-22 18:40:45,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3481.6, 300 sec: 3748.9). Total num frames: 2199552. Throughput: 0: 874.0. Samples: 550016. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:40:45,598][01245] Avg episode reward: [(0, '4.726')] +[2023-02-22 18:40:47,385][15057] Updated weights for policy 0, policy_version 540 (0.0031) +[2023-02-22 18:40:50,598][01245] Fps is (10 sec: 4503.5, 60 sec: 3550.0, 300 sec: 3790.5). Total num frames: 2224128. Throughput: 0: 871.6. Samples: 556848. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:40:50,600][01245] Avg episode reward: [(0, '4.896')] +[2023-02-22 18:40:55,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3481.6, 300 sec: 3776.7). Total num frames: 2236416. Throughput: 0: 844.1. Samples: 559062. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:40:55,598][01245] Avg episode reward: [(0, '4.866')] +[2023-02-22 18:40:59,887][15057] Updated weights for policy 0, policy_version 550 (0.0024) +[2023-02-22 18:41:00,594][01245] Fps is (10 sec: 2868.5, 60 sec: 3481.6, 300 sec: 3748.9). Total num frames: 2252800. Throughput: 0: 839.4. Samples: 563402. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:41:00,600][01245] Avg episode reward: [(0, '5.080')] +[2023-02-22 18:41:05,594][01245] Fps is (10 sec: 4095.9, 60 sec: 3481.6, 300 sec: 3762.8). Total num frames: 2277376. Throughput: 0: 903.5. Samples: 570062. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:41:05,597][01245] Avg episode reward: [(0, '5.148')] +[2023-02-22 18:41:08,651][15057] Updated weights for policy 0, policy_version 560 (0.0013) +[2023-02-22 18:41:10,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3481.6, 300 sec: 3776.7). Total num frames: 2297856. Throughput: 0: 936.2. Samples: 573626. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:41:10,604][01245] Avg episode reward: [(0, '5.176')] +[2023-02-22 18:41:15,597][01245] Fps is (10 sec: 3685.1, 60 sec: 3549.6, 300 sec: 3776.6). Total num frames: 2314240. Throughput: 0: 934.3. Samples: 578784. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:41:15,599][01245] Avg episode reward: [(0, '5.084')] +[2023-02-22 18:41:20,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3618.2, 300 sec: 3762.8). Total num frames: 2330624. Throughput: 0: 928.5. Samples: 583252. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:41:20,596][01245] Avg episode reward: [(0, '5.064')] +[2023-02-22 18:41:21,302][15057] Updated weights for policy 0, policy_version 570 (0.0012) +[2023-02-22 18:41:25,594][01245] Fps is (10 sec: 4097.5, 60 sec: 3754.7, 300 sec: 3762.8). Total num frames: 2355200. Throughput: 0: 955.3. Samples: 586680. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:41:25,596][01245] Avg episode reward: [(0, '4.888')] +[2023-02-22 18:41:30,003][15057] Updated weights for policy 0, policy_version 580 (0.0018) +[2023-02-22 18:41:30,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3822.9, 300 sec: 3776.7). Total num frames: 2375680. Throughput: 0: 969.7. Samples: 593652. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:41:30,596][01245] Avg episode reward: [(0, '4.805')] +[2023-02-22 18:41:35,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3776.7). Total num frames: 2392064. Throughput: 0: 925.1. Samples: 598472. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:41:35,601][01245] Avg episode reward: [(0, '4.819')] +[2023-02-22 18:41:40,594][01245] Fps is (10 sec: 2867.0, 60 sec: 3754.6, 300 sec: 3748.9). Total num frames: 2404352. Throughput: 0: 924.3. Samples: 600658. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:41:40,597][01245] Avg episode reward: [(0, '4.841')] +[2023-02-22 18:41:42,443][15057] Updated weights for policy 0, policy_version 590 (0.0013) +[2023-02-22 18:41:45,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3748.9). Total num frames: 2428928. Throughput: 0: 966.8. Samples: 606908. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:41:45,602][01245] Avg episode reward: [(0, '4.926')] +[2023-02-22 18:41:50,600][01245] Fps is (10 sec: 4912.3, 60 sec: 3822.8, 300 sec: 3776.6). Total num frames: 2453504. Throughput: 0: 973.8. Samples: 613888. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:41:50,610][01245] Avg episode reward: [(0, '4.961')] +[2023-02-22 18:41:51,940][15057] Updated weights for policy 0, policy_version 600 (0.0014) +[2023-02-22 18:41:55,598][01245] Fps is (10 sec: 3684.6, 60 sec: 3822.6, 300 sec: 3762.7). Total num frames: 2465792. Throughput: 0: 942.7. Samples: 616054. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:41:55,607][01245] Avg episode reward: [(0, '5.074')] +[2023-02-22 18:42:00,594][01245] Fps is (10 sec: 2869.1, 60 sec: 3822.9, 300 sec: 3748.9). Total num frames: 2482176. Throughput: 0: 925.8. Samples: 620442. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:42:00,600][01245] Avg episode reward: [(0, '5.041')] +[2023-02-22 18:42:03,585][15057] Updated weights for policy 0, policy_version 610 (0.0018) +[2023-02-22 18:42:05,594][01245] Fps is (10 sec: 4098.0, 60 sec: 3822.9, 300 sec: 3748.9). Total num frames: 2506752. Throughput: 0: 977.1. Samples: 627222. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:42:05,599][01245] Avg episode reward: [(0, '4.935')] +[2023-02-22 18:42:10,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3822.9, 300 sec: 3776.7). Total num frames: 2527232. Throughput: 0: 977.7. Samples: 630678. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:42:10,601][01245] Avg episode reward: [(0, '4.765')] +[2023-02-22 18:42:10,618][15039] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000617_2527232.pth... +[2023-02-22 18:42:10,784][15039] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000398_1630208.pth +[2023-02-22 18:42:13,694][15057] Updated weights for policy 0, policy_version 620 (0.0018) +[2023-02-22 18:42:15,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3823.2, 300 sec: 3776.7). Total num frames: 2543616. Throughput: 0: 938.5. Samples: 635884. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:42:15,604][01245] Avg episode reward: [(0, '4.801')] +[2023-02-22 18:42:20,594][01245] Fps is (10 sec: 3276.7, 60 sec: 3822.9, 300 sec: 3748.9). Total num frames: 2560000. Throughput: 0: 934.6. Samples: 640528. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:42:20,597][01245] Avg episode reward: [(0, '5.031')] +[2023-02-22 18:42:24,615][15057] Updated weights for policy 0, policy_version 630 (0.0021) +[2023-02-22 18:42:25,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3762.8). Total num frames: 2584576. Throughput: 0: 964.9. Samples: 644080. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:42:25,596][01245] Avg episode reward: [(0, '5.379')] +[2023-02-22 18:42:25,600][15039] Saving new best policy, reward=5.379! +[2023-02-22 18:42:30,594][01245] Fps is (10 sec: 4505.7, 60 sec: 3822.9, 300 sec: 3776.7). Total num frames: 2605056. Throughput: 0: 982.5. Samples: 651122. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:42:30,601][01245] Avg episode reward: [(0, '5.374')] +[2023-02-22 18:42:35,366][15057] Updated weights for policy 0, policy_version 640 (0.0024) +[2023-02-22 18:42:35,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3776.7). Total num frames: 2621440. Throughput: 0: 935.6. Samples: 655984. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:42:35,599][01245] Avg episode reward: [(0, '5.186')] +[2023-02-22 18:42:40,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3891.2, 300 sec: 3748.9). Total num frames: 2637824. Throughput: 0: 937.0. Samples: 658214. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:42:40,596][01245] Avg episode reward: [(0, '4.934')] +[2023-02-22 18:42:45,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3748.9). Total num frames: 2658304. Throughput: 0: 980.2. Samples: 664550. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:42:45,601][01245] Avg episode reward: [(0, '4.900')] +[2023-02-22 18:42:45,698][15057] Updated weights for policy 0, policy_version 650 (0.0014) +[2023-02-22 18:42:50,597][01245] Fps is (10 sec: 4503.9, 60 sec: 3823.1, 300 sec: 3776.6). Total num frames: 2682880. Throughput: 0: 982.7. Samples: 671446. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:42:50,601][01245] Avg episode reward: [(0, '5.090')] +[2023-02-22 18:42:55,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3823.2, 300 sec: 3762.8). Total num frames: 2695168. Throughput: 0: 955.1. Samples: 673656. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:42:55,601][01245] Avg episode reward: [(0, '5.365')] +[2023-02-22 18:42:57,173][15057] Updated weights for policy 0, policy_version 660 (0.0020) +[2023-02-22 18:43:00,594][01245] Fps is (10 sec: 2868.3, 60 sec: 3822.9, 300 sec: 3735.0). Total num frames: 2711552. Throughput: 0: 940.4. Samples: 678200. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:43:00,599][01245] Avg episode reward: [(0, '5.134')] +[2023-02-22 18:43:05,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3748.9). Total num frames: 2736128. Throughput: 0: 991.8. Samples: 685160. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:43:05,599][01245] Avg episode reward: [(0, '5.119')] +[2023-02-22 18:43:06,652][15057] Updated weights for policy 0, policy_version 670 (0.0018) +[2023-02-22 18:43:10,594][01245] Fps is (10 sec: 4915.2, 60 sec: 3891.2, 300 sec: 3776.7). Total num frames: 2760704. Throughput: 0: 992.4. Samples: 688738. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:43:10,604][01245] Avg episode reward: [(0, '5.301')] +[2023-02-22 18:43:15,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3762.8). Total num frames: 2772992. Throughput: 0: 949.2. Samples: 693834. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:43:15,601][01245] Avg episode reward: [(0, '5.671')] +[2023-02-22 18:43:15,638][15039] Saving new best policy, reward=5.671! +[2023-02-22 18:43:18,415][15057] Updated weights for policy 0, policy_version 680 (0.0023) +[2023-02-22 18:43:20,594][01245] Fps is (10 sec: 2867.2, 60 sec: 3822.9, 300 sec: 3748.9). Total num frames: 2789376. Throughput: 0: 944.9. Samples: 698504. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:43:20,601][01245] Avg episode reward: [(0, '5.757')] +[2023-02-22 18:43:20,615][15039] Saving new best policy, reward=5.757! +[2023-02-22 18:43:25,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3748.9). Total num frames: 2813952. Throughput: 0: 974.1. Samples: 702050. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:43:25,596][01245] Avg episode reward: [(0, '5.167')] +[2023-02-22 18:43:27,477][15057] Updated weights for policy 0, policy_version 690 (0.0032) +[2023-02-22 18:43:30,594][01245] Fps is (10 sec: 4915.2, 60 sec: 3891.2, 300 sec: 3790.5). Total num frames: 2838528. Throughput: 0: 996.4. Samples: 709386. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:43:30,597][01245] Avg episode reward: [(0, '5.187')] +[2023-02-22 18:43:35,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3891.2, 300 sec: 3790.6). Total num frames: 2854912. Throughput: 0: 951.0. Samples: 714236. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:43:35,602][01245] Avg episode reward: [(0, '5.155')] +[2023-02-22 18:43:39,403][15057] Updated weights for policy 0, policy_version 700 (0.0014) +[2023-02-22 18:43:40,594][01245] Fps is (10 sec: 3276.7, 60 sec: 3891.2, 300 sec: 3762.8). Total num frames: 2871296. Throughput: 0: 952.5. Samples: 716518. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:43:40,596][01245] Avg episode reward: [(0, '4.910')] +[2023-02-22 18:43:45,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3959.5, 300 sec: 3776.7). Total num frames: 2895872. Throughput: 0: 997.6. Samples: 723092. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:43:45,598][01245] Avg episode reward: [(0, '4.841')] +[2023-02-22 18:43:48,115][15057] Updated weights for policy 0, policy_version 710 (0.0034) +[2023-02-22 18:43:50,594][01245] Fps is (10 sec: 4505.8, 60 sec: 3891.4, 300 sec: 3790.5). Total num frames: 2916352. Throughput: 0: 993.4. Samples: 729862. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:43:50,596][01245] Avg episode reward: [(0, '5.354')] +[2023-02-22 18:43:55,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3959.5, 300 sec: 3790.5). Total num frames: 2932736. Throughput: 0: 965.0. Samples: 732164. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:43:55,602][01245] Avg episode reward: [(0, '5.491')] +[2023-02-22 18:44:00,215][15057] Updated weights for policy 0, policy_version 720 (0.0025) +[2023-02-22 18:44:00,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3959.5, 300 sec: 3762.8). Total num frames: 2949120. Throughput: 0: 956.5. Samples: 736878. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:44:00,598][01245] Avg episode reward: [(0, '5.624')] +[2023-02-22 18:44:05,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3891.2, 300 sec: 3762.8). Total num frames: 2969600. Throughput: 0: 996.2. Samples: 743334. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:44:05,596][01245] Avg episode reward: [(0, '5.642')] +[2023-02-22 18:44:09,325][15057] Updated weights for policy 0, policy_version 730 (0.0015) +[2023-02-22 18:44:10,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3776.7). Total num frames: 2990080. Throughput: 0: 995.0. Samples: 746826. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:44:10,596][01245] Avg episode reward: [(0, '6.050')] +[2023-02-22 18:44:10,665][15039] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000731_2994176.pth... +[2023-02-22 18:44:10,806][15039] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000510_2088960.pth +[2023-02-22 18:44:10,825][15039] Saving new best policy, reward=6.050! +[2023-02-22 18:44:15,597][01245] Fps is (10 sec: 3275.6, 60 sec: 3822.7, 300 sec: 3762.7). Total num frames: 3002368. Throughput: 0: 928.1. Samples: 751152. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:44:15,604][01245] Avg episode reward: [(0, '6.124')] +[2023-02-22 18:44:15,656][15039] Saving new best policy, reward=6.124! +[2023-02-22 18:44:20,594][01245] Fps is (10 sec: 2867.2, 60 sec: 3822.9, 300 sec: 3735.0). Total num frames: 3018752. Throughput: 0: 921.9. Samples: 755720. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:44:20,602][01245] Avg episode reward: [(0, '6.291')] +[2023-02-22 18:44:20,612][15039] Saving new best policy, reward=6.291! +[2023-02-22 18:44:22,629][15057] Updated weights for policy 0, policy_version 740 (0.0017) +[2023-02-22 18:44:25,594][01245] Fps is (10 sec: 4097.5, 60 sec: 3822.9, 300 sec: 3748.9). Total num frames: 3043328. Throughput: 0: 948.1. Samples: 759180. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:44:25,598][01245] Avg episode reward: [(0, '6.194')] +[2023-02-22 18:44:30,594][01245] Fps is (10 sec: 4915.2, 60 sec: 3822.9, 300 sec: 3790.5). Total num frames: 3067904. Throughput: 0: 960.9. Samples: 766334. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:44:30,599][01245] Avg episode reward: [(0, '6.597')] +[2023-02-22 18:44:30,609][15039] Saving new best policy, reward=6.597! +[2023-02-22 18:44:32,069][15057] Updated weights for policy 0, policy_version 750 (0.0024) +[2023-02-22 18:44:35,597][01245] Fps is (10 sec: 3685.0, 60 sec: 3754.4, 300 sec: 3776.6). Total num frames: 3080192. Throughput: 0: 913.9. Samples: 770992. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:44:35,600][01245] Avg episode reward: [(0, '6.842')] +[2023-02-22 18:44:35,609][15039] Saving new best policy, reward=6.842! +[2023-02-22 18:44:40,594][01245] Fps is (10 sec: 2867.2, 60 sec: 3754.7, 300 sec: 3748.9). Total num frames: 3096576. Throughput: 0: 908.2. Samples: 773032. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:44:40,601][01245] Avg episode reward: [(0, '6.904')] +[2023-02-22 18:44:40,613][15039] Saving new best policy, reward=6.904! +[2023-02-22 18:44:44,160][15057] Updated weights for policy 0, policy_version 760 (0.0024) +[2023-02-22 18:44:45,594][01245] Fps is (10 sec: 3687.6, 60 sec: 3686.4, 300 sec: 3749.0). Total num frames: 3117056. Throughput: 0: 936.3. Samples: 779010. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:44:45,603][01245] Avg episode reward: [(0, '7.115')] +[2023-02-22 18:44:45,608][15039] Saving new best policy, reward=7.115! +[2023-02-22 18:44:50,596][01245] Fps is (10 sec: 4094.9, 60 sec: 3686.2, 300 sec: 3762.7). Total num frames: 3137536. Throughput: 0: 937.9. Samples: 785540. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:44:50,600][01245] Avg episode reward: [(0, '6.755')] +[2023-02-22 18:44:55,595][01245] Fps is (10 sec: 3276.3, 60 sec: 3618.0, 300 sec: 3748.9). Total num frames: 3149824. Throughput: 0: 902.6. Samples: 787444. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:44:55,600][01245] Avg episode reward: [(0, '7.164')] +[2023-02-22 18:44:55,607][15039] Saving new best policy, reward=7.164! +[2023-02-22 18:44:56,755][15057] Updated weights for policy 0, policy_version 770 (0.0023) +[2023-02-22 18:45:00,595][01245] Fps is (10 sec: 2048.3, 60 sec: 3481.5, 300 sec: 3693.3). Total num frames: 3158016. Throughput: 0: 872.8. Samples: 790428. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:45:00,603][01245] Avg episode reward: [(0, '7.473')] +[2023-02-22 18:45:00,614][15039] Saving new best policy, reward=7.473! +[2023-02-22 18:45:05,594][01245] Fps is (10 sec: 2048.3, 60 sec: 3345.1, 300 sec: 3665.6). Total num frames: 3170304. Throughput: 0: 836.5. Samples: 793362. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:45:05,597][01245] Avg episode reward: [(0, '7.367')] +[2023-02-22 18:45:10,594][01245] Fps is (10 sec: 2867.6, 60 sec: 3276.8, 300 sec: 3679.5). Total num frames: 3186688. Throughput: 0: 810.3. Samples: 795644. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:45:10,597][01245] Avg episode reward: [(0, '8.313')] +[2023-02-22 18:45:10,699][15039] Saving new best policy, reward=8.313! +[2023-02-22 18:45:11,689][15057] Updated weights for policy 0, policy_version 780 (0.0031) +[2023-02-22 18:45:15,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3481.8, 300 sec: 3721.1). Total num frames: 3211264. Throughput: 0: 798.0. Samples: 802244. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:45:15,602][01245] Avg episode reward: [(0, '8.678')] +[2023-02-22 18:45:15,604][15039] Saving new best policy, reward=8.678! +[2023-02-22 18:45:20,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3413.3, 300 sec: 3707.2). Total num frames: 3223552. Throughput: 0: 797.6. Samples: 806882. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:45:20,601][01245] Avg episode reward: [(0, '8.683')] +[2023-02-22 18:45:20,611][15039] Saving new best policy, reward=8.683! +[2023-02-22 18:45:23,863][15057] Updated weights for policy 0, policy_version 790 (0.0026) +[2023-02-22 18:45:25,594][01245] Fps is (10 sec: 2867.2, 60 sec: 3276.8, 300 sec: 3707.2). Total num frames: 3239936. Throughput: 0: 800.7. Samples: 809062. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:45:25,596][01245] Avg episode reward: [(0, '8.073')] +[2023-02-22 18:45:30,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3276.8, 300 sec: 3735.0). Total num frames: 3264512. Throughput: 0: 807.9. Samples: 815366. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:45:30,602][01245] Avg episode reward: [(0, '7.897')] +[2023-02-22 18:45:33,194][15057] Updated weights for policy 0, policy_version 800 (0.0012) +[2023-02-22 18:45:35,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3413.5, 300 sec: 3748.9). Total num frames: 3284992. Throughput: 0: 814.8. Samples: 822202. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:45:35,598][01245] Avg episode reward: [(0, '8.396')] +[2023-02-22 18:45:40,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3413.3, 300 sec: 3735.0). Total num frames: 3301376. Throughput: 0: 821.4. Samples: 824404. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:45:40,600][01245] Avg episode reward: [(0, '9.292')] +[2023-02-22 18:45:40,616][15039] Saving new best policy, reward=9.292! +[2023-02-22 18:45:45,551][15057] Updated weights for policy 0, policy_version 810 (0.0025) +[2023-02-22 18:45:45,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3345.1, 300 sec: 3707.3). Total num frames: 3317760. Throughput: 0: 851.5. Samples: 828744. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:45:45,597][01245] Avg episode reward: [(0, '9.482')] +[2023-02-22 18:45:45,604][15039] Saving new best policy, reward=9.482! +[2023-02-22 18:45:50,594][01245] Fps is (10 sec: 3686.3, 60 sec: 3345.2, 300 sec: 3735.0). Total num frames: 3338240. Throughput: 0: 938.5. Samples: 835594. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:45:50,599][01245] Avg episode reward: [(0, '9.601')] +[2023-02-22 18:45:50,613][15039] Saving new best policy, reward=9.601! +[2023-02-22 18:45:54,327][15057] Updated weights for policy 0, policy_version 820 (0.0012) +[2023-02-22 18:45:55,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3550.0, 300 sec: 3762.8). Total num frames: 3362816. Throughput: 0: 964.2. Samples: 839034. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:45:55,596][01245] Avg episode reward: [(0, '8.585')] +[2023-02-22 18:46:00,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3618.2, 300 sec: 3721.1). Total num frames: 3375104. Throughput: 0: 933.2. Samples: 844238. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:46:00,603][01245] Avg episode reward: [(0, '8.321')] +[2023-02-22 18:46:05,594][01245] Fps is (10 sec: 2867.2, 60 sec: 3686.4, 300 sec: 3707.2). Total num frames: 3391488. Throughput: 0: 929.8. Samples: 848724. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:46:05,596][01245] Avg episode reward: [(0, '8.667')] +[2023-02-22 18:46:06,875][15057] Updated weights for policy 0, policy_version 830 (0.0016) +[2023-02-22 18:46:10,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3735.0). Total num frames: 3416064. Throughput: 0: 960.0. Samples: 852262. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:46:10,596][01245] Avg episode reward: [(0, '8.954')] +[2023-02-22 18:46:10,607][15039] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000834_3416064.pth... +[2023-02-22 18:46:10,740][15039] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000617_2527232.pth +[2023-02-22 18:46:15,596][01245] Fps is (10 sec: 4504.4, 60 sec: 3754.5, 300 sec: 3748.8). Total num frames: 3436544. Throughput: 0: 974.3. Samples: 859212. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:46:15,601][01245] Avg episode reward: [(0, '9.443')] +[2023-02-22 18:46:15,839][15057] Updated weights for policy 0, policy_version 840 (0.0023) +[2023-02-22 18:46:20,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3721.1). Total num frames: 3452928. Throughput: 0: 928.0. Samples: 863960. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:46:20,596][01245] Avg episode reward: [(0, '9.179')] +[2023-02-22 18:46:25,594][01245] Fps is (10 sec: 3277.7, 60 sec: 3822.9, 300 sec: 3707.2). Total num frames: 3469312. Throughput: 0: 928.6. Samples: 866190. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:46:25,596][01245] Avg episode reward: [(0, '8.944')] +[2023-02-22 18:46:27,872][15057] Updated weights for policy 0, policy_version 850 (0.0019) +[2023-02-22 18:46:30,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3735.0). Total num frames: 3493888. Throughput: 0: 976.0. Samples: 872664. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:46:30,597][01245] Avg episode reward: [(0, '8.765')] +[2023-02-22 18:46:35,598][01245] Fps is (10 sec: 4503.5, 60 sec: 3822.6, 300 sec: 3762.7). Total num frames: 3514368. Throughput: 0: 977.1. Samples: 879570. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:46:35,603][01245] Avg episode reward: [(0, '9.319')] +[2023-02-22 18:46:37,367][15057] Updated weights for policy 0, policy_version 860 (0.0025) +[2023-02-22 18:46:40,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3735.0). Total num frames: 3530752. Throughput: 0: 951.1. Samples: 881832. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:46:40,603][01245] Avg episode reward: [(0, '9.505')] +[2023-02-22 18:46:45,594][01245] Fps is (10 sec: 3278.3, 60 sec: 3822.9, 300 sec: 3707.3). Total num frames: 3547136. Throughput: 0: 932.5. Samples: 886202. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:46:45,600][01245] Avg episode reward: [(0, '10.274')] +[2023-02-22 18:46:45,607][15039] Saving new best policy, reward=10.274! +[2023-02-22 18:46:49,069][15057] Updated weights for policy 0, policy_version 870 (0.0018) +[2023-02-22 18:46:50,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3735.1). Total num frames: 3567616. Throughput: 0: 979.1. Samples: 892784. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:46:50,602][01245] Avg episode reward: [(0, '10.256')] +[2023-02-22 18:46:55,595][01245] Fps is (10 sec: 4504.8, 60 sec: 3822.8, 300 sec: 3762.7). Total num frames: 3592192. Throughput: 0: 978.7. Samples: 896304. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:46:55,602][01245] Avg episode reward: [(0, '10.309')] +[2023-02-22 18:46:55,606][15039] Saving new best policy, reward=10.309! +[2023-02-22 18:46:59,481][15057] Updated weights for policy 0, policy_version 880 (0.0026) +[2023-02-22 18:47:00,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3721.1). Total num frames: 3604480. Throughput: 0: 943.6. Samples: 901670. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:47:00,598][01245] Avg episode reward: [(0, '10.353')] +[2023-02-22 18:47:00,612][15039] Saving new best policy, reward=10.353! +[2023-02-22 18:47:05,594][01245] Fps is (10 sec: 2867.7, 60 sec: 3822.9, 300 sec: 3707.2). Total num frames: 3620864. Throughput: 0: 938.4. Samples: 906190. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:47:05,601][01245] Avg episode reward: [(0, '10.590')] +[2023-02-22 18:47:05,607][15039] Saving new best policy, reward=10.590! +[2023-02-22 18:47:10,218][15057] Updated weights for policy 0, policy_version 890 (0.0025) +[2023-02-22 18:47:10,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3735.0). Total num frames: 3645440. Throughput: 0: 967.8. Samples: 909740. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:47:10,596][01245] Avg episode reward: [(0, '11.236')] +[2023-02-22 18:47:10,604][15039] Saving new best policy, reward=11.236! +[2023-02-22 18:47:15,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3823.1, 300 sec: 3748.9). Total num frames: 3665920. Throughput: 0: 980.9. Samples: 916806. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:47:15,606][01245] Avg episode reward: [(0, '11.866')] +[2023-02-22 18:47:15,610][15039] Saving new best policy, reward=11.866! +[2023-02-22 18:47:20,594][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3721.1). Total num frames: 3682304. Throughput: 0: 932.8. Samples: 921540. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) +[2023-02-22 18:47:20,599][01245] Avg episode reward: [(0, '12.075')] +[2023-02-22 18:47:20,611][15039] Saving new best policy, reward=12.075! +[2023-02-22 18:47:21,354][15057] Updated weights for policy 0, policy_version 900 (0.0013) +[2023-02-22 18:47:25,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3822.9, 300 sec: 3707.2). Total num frames: 3698688. Throughput: 0: 931.6. Samples: 923752. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:47:25,600][01245] Avg episode reward: [(0, '11.108')] +[2023-02-22 18:47:30,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3735.0). Total num frames: 3723264. Throughput: 0: 977.6. Samples: 930194. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:47:30,602][01245] Avg episode reward: [(0, '11.198')] +[2023-02-22 18:47:31,125][15057] Updated weights for policy 0, policy_version 910 (0.0020) +[2023-02-22 18:47:35,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3823.2, 300 sec: 3748.9). Total num frames: 3743744. Throughput: 0: 988.9. Samples: 937284. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:47:35,599][01245] Avg episode reward: [(0, '10.621')] +[2023-02-22 18:47:40,594][01245] Fps is (10 sec: 3686.3, 60 sec: 3822.9, 300 sec: 3735.0). Total num frames: 3760128. Throughput: 0: 960.5. Samples: 939526. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:47:40,599][01245] Avg episode reward: [(0, '10.889')] +[2023-02-22 18:47:42,561][15057] Updated weights for policy 0, policy_version 920 (0.0022) +[2023-02-22 18:47:45,595][01245] Fps is (10 sec: 3276.3, 60 sec: 3822.8, 300 sec: 3707.3). Total num frames: 3776512. Throughput: 0: 941.7. Samples: 944046. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:47:45,602][01245] Avg episode reward: [(0, '11.553')] +[2023-02-22 18:47:50,594][01245] Fps is (10 sec: 4096.1, 60 sec: 3891.2, 300 sec: 3748.9). Total num frames: 3801088. Throughput: 0: 993.6. Samples: 950900. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:47:50,596][01245] Avg episode reward: [(0, '11.358')] +[2023-02-22 18:47:52,061][15057] Updated weights for policy 0, policy_version 930 (0.0034) +[2023-02-22 18:47:55,598][01245] Fps is (10 sec: 4913.7, 60 sec: 3891.0, 300 sec: 3776.6). Total num frames: 3825664. Throughput: 0: 993.9. Samples: 954472. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:47:55,602][01245] Avg episode reward: [(0, '11.286')] +[2023-02-22 18:48:00,594][01245] Fps is (10 sec: 3686.3, 60 sec: 3891.2, 300 sec: 3735.0). Total num frames: 3837952. Throughput: 0: 956.5. Samples: 959850. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:48:00,596][01245] Avg episode reward: [(0, '11.653')] +[2023-02-22 18:48:03,797][15057] Updated weights for policy 0, policy_version 940 (0.0021) +[2023-02-22 18:48:05,594][01245] Fps is (10 sec: 2868.5, 60 sec: 3891.2, 300 sec: 3707.2). Total num frames: 3854336. Throughput: 0: 951.3. Samples: 964348. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:48:05,600][01245] Avg episode reward: [(0, '11.339')] +[2023-02-22 18:48:10,594][01245] Fps is (10 sec: 4096.1, 60 sec: 3891.2, 300 sec: 3748.9). Total num frames: 3878912. Throughput: 0: 980.8. Samples: 967890. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:48:10,596][01245] Avg episode reward: [(0, '11.033')] +[2023-02-22 18:48:10,604][15039] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000947_3878912.pth... +[2023-02-22 18:48:10,736][15039] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000731_2994176.pth +[2023-02-22 18:48:13,020][15057] Updated weights for policy 0, policy_version 950 (0.0014) +[2023-02-22 18:48:15,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3891.2, 300 sec: 3762.8). Total num frames: 3899392. Throughput: 0: 991.7. Samples: 974822. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:48:15,599][01245] Avg episode reward: [(0, '11.520')] +[2023-02-22 18:48:20,596][01245] Fps is (10 sec: 3685.4, 60 sec: 3891.0, 300 sec: 3735.0). Total num frames: 3915776. Throughput: 0: 940.2. Samples: 979594. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:48:20,601][01245] Avg episode reward: [(0, '11.486')] +[2023-02-22 18:48:25,573][15057] Updated weights for policy 0, policy_version 960 (0.0024) +[2023-02-22 18:48:25,594][01245] Fps is (10 sec: 3276.8, 60 sec: 3891.2, 300 sec: 3707.2). Total num frames: 3932160. Throughput: 0: 940.5. Samples: 981850. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:48:25,596][01245] Avg episode reward: [(0, '11.949')] +[2023-02-22 18:48:30,594][01245] Fps is (10 sec: 3687.4, 60 sec: 3822.9, 300 sec: 3721.1). Total num frames: 3952640. Throughput: 0: 982.3. Samples: 988248. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:48:30,597][01245] Avg episode reward: [(0, '12.353')] +[2023-02-22 18:48:30,667][15039] Saving new best policy, reward=12.353! +[2023-02-22 18:48:34,027][15057] Updated weights for policy 0, policy_version 970 (0.0012) +[2023-02-22 18:48:35,594][01245] Fps is (10 sec: 4505.6, 60 sec: 3891.2, 300 sec: 3748.9). Total num frames: 3977216. Throughput: 0: 987.7. Samples: 995348. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:48:35,605][01245] Avg episode reward: [(0, '13.230')] +[2023-02-22 18:48:35,613][15039] Saving new best policy, reward=13.230! +[2023-02-22 18:48:40,594][01245] Fps is (10 sec: 4096.0, 60 sec: 3891.2, 300 sec: 3721.1). Total num frames: 3993600. Throughput: 0: 955.7. Samples: 997472. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:48:40,596][01245] Avg episode reward: [(0, '13.192')] +[2023-02-22 18:48:44,303][15039] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... +[2023-02-22 18:48:44,311][01245] Component Batcher_0 stopped! +[2023-02-22 18:48:44,310][15039] Stopping Batcher_0... +[2023-02-22 18:48:44,319][15039] Loop batcher_evt_loop terminating... +[2023-02-22 18:48:44,371][15057] Weights refcount: 2 0 +[2023-02-22 18:48:44,383][01245] Component InferenceWorker_p0-w0 stopped! +[2023-02-22 18:48:44,390][15057] Stopping InferenceWorker_p0-w0... +[2023-02-22 18:48:44,390][15057] Loop inference_proc0-0_evt_loop terminating... +[2023-02-22 18:48:44,394][01245] Component RolloutWorker_w6 stopped! +[2023-02-22 18:48:44,398][15064] Stopping RolloutWorker_w6... +[2023-02-22 18:48:44,399][15064] Loop rollout_proc6_evt_loop terminating... +[2023-02-22 18:48:44,412][15060] Stopping RolloutWorker_w2... +[2023-02-22 18:48:44,413][15060] Loop rollout_proc2_evt_loop terminating... +[2023-02-22 18:48:44,409][01245] Component RolloutWorker_w2 stopped! +[2023-02-22 18:48:44,457][15063] Stopping RolloutWorker_w3... +[2023-02-22 18:48:44,458][01245] Component RolloutWorker_w3 stopped! +[2023-02-22 18:48:44,469][15061] Stopping RolloutWorker_w5... +[2023-02-22 18:48:44,470][01245] Component RolloutWorker_w5 stopped! +[2023-02-22 18:48:44,476][01245] Component RolloutWorker_w4 stopped! +[2023-02-22 18:48:44,480][15062] Stopping RolloutWorker_w4... +[2023-02-22 18:48:44,480][15062] Loop rollout_proc4_evt_loop terminating... +[2023-02-22 18:48:44,479][15059] Stopping RolloutWorker_w1... +[2023-02-22 18:48:44,485][15059] Loop rollout_proc1_evt_loop terminating... +[2023-02-22 18:48:44,484][01245] Component RolloutWorker_w1 stopped! +[2023-02-22 18:48:44,470][15061] Loop rollout_proc5_evt_loop terminating... +[2023-02-22 18:48:44,500][15058] Stopping RolloutWorker_w0... +[2023-02-22 18:48:44,500][15058] Loop rollout_proc0_evt_loop terminating... +[2023-02-22 18:48:44,458][15063] Loop rollout_proc3_evt_loop terminating... +[2023-02-22 18:48:44,503][01245] Component RolloutWorker_w0 stopped! +[2023-02-22 18:48:44,552][15039] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000834_3416064.pth +[2023-02-22 18:48:44,573][15039] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... +[2023-02-22 18:48:44,604][01245] Component RolloutWorker_w7 stopped! +[2023-02-22 18:48:44,613][15065] Stopping RolloutWorker_w7... +[2023-02-22 18:48:44,620][15065] Loop rollout_proc7_evt_loop terminating... +[2023-02-22 18:48:44,795][01245] Component LearnerWorker_p0 stopped! +[2023-02-22 18:48:44,799][01245] Waiting for process learner_proc0 to stop... +[2023-02-22 18:48:44,802][15039] Stopping LearnerWorker_p0... +[2023-02-22 18:48:44,803][15039] Loop learner_proc0_evt_loop terminating... +[2023-02-22 18:48:46,695][01245] Waiting for process inference_proc0-0 to join... +[2023-02-22 18:48:47,062][01245] Waiting for process rollout_proc0 to join... +[2023-02-22 18:48:47,498][01245] Waiting for process rollout_proc1 to join... +[2023-02-22 18:48:47,500][01245] Waiting for process rollout_proc2 to join... +[2023-02-22 18:48:47,501][01245] Waiting for process rollout_proc3 to join... +[2023-02-22 18:48:47,503][01245] Waiting for process rollout_proc4 to join... +[2023-02-22 18:48:47,504][01245] Waiting for process rollout_proc5 to join... +[2023-02-22 18:48:47,505][01245] Waiting for process rollout_proc6 to join... +[2023-02-22 18:48:47,511][01245] Waiting for process rollout_proc7 to join... +[2023-02-22 18:48:47,512][01245] Batcher 0 profile tree view: +batching: 24.8797, releasing_batches: 0.0222 +[2023-02-22 18:48:47,514][01245] InferenceWorker_p0-w0 profile tree view: +wait_policy: 0.0138 + wait_policy_total: 524.3641 +update_model: 7.7211 + weight_update: 0.0016 +one_step: 0.0040 + handle_policy_step: 511.1373 + deserialize: 14.4846, stack: 2.9721, obs_to_device_normalize: 114.4620, forward: 244.7572, send_messages: 26.4607 + prepare_outputs: 81.9385 + to_cpu: 50.7951 +[2023-02-22 18:48:47,515][01245] Learner 0 profile tree view: +misc: 0.0060, prepare_batch: 16.4633 +train: 76.3437 + epoch_init: 0.0106, minibatch_init: 0.0184, losses_postprocess: 0.5721, kl_divergence: 0.6301, after_optimizer: 32.9371 + calculate_losses: 27.4845 + losses_init: 0.0065, forward_head: 1.7713, bptt_initial: 18.0713, tail: 1.0499, advantages_returns: 0.2806, losses: 3.6938 + bptt: 2.3116 + bptt_forward_core: 2.1994 + update: 14.0724 + clip: 1.3864 +[2023-02-22 18:48:47,516][01245] RolloutWorker_w0 profile tree view: +wait_for_trajectories: 0.3414, enqueue_policy_requests: 140.1474, env_step: 818.3112, overhead: 20.0871, complete_rollouts: 6.6500 +save_policy_outputs: 19.8322 + split_output_tensors: 9.6844 +[2023-02-22 18:48:47,518][01245] RolloutWorker_w7 profile tree view: +wait_for_trajectories: 0.3328, enqueue_policy_requests: 140.2637, env_step: 817.0888, overhead: 20.3039, complete_rollouts: 7.3417 +save_policy_outputs: 19.3221 + split_output_tensors: 9.2830 +[2023-02-22 18:48:47,520][01245] Loop Runner_EvtLoop terminating... +[2023-02-22 18:48:47,522][01245] Runner profile tree view: +main_loop: 1112.2975 +[2023-02-22 18:48:47,523][01245] Collected {0: 4005888}, FPS: 3601.5 +[2023-02-22 18:48:47,656][01245] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json +[2023-02-22 18:48:47,658][01245] Overriding arg 'num_workers' with value 1 passed from command line +[2023-02-22 18:48:47,659][01245] Adding new argument 'no_render'=True that is not in the saved config file! +[2023-02-22 18:48:47,661][01245] Adding new argument 'save_video'=True that is not in the saved config file! +[2023-02-22 18:48:47,662][01245] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! +[2023-02-22 18:48:47,663][01245] Adding new argument 'video_name'=None that is not in the saved config file! +[2023-02-22 18:48:47,664][01245] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file! +[2023-02-22 18:48:47,666][01245] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! +[2023-02-22 18:48:47,667][01245] Adding new argument 'push_to_hub'=False that is not in the saved config file! +[2023-02-22 18:48:47,669][01245] Adding new argument 'hf_repository'=None that is not in the saved config file! +[2023-02-22 18:48:47,670][01245] Adding new argument 'policy_index'=0 that is not in the saved config file! +[2023-02-22 18:48:47,671][01245] Adding new argument 'eval_deterministic'=False that is not in the saved config file! +[2023-02-22 18:48:47,672][01245] Adding new argument 'train_script'=None that is not in the saved config file! +[2023-02-22 18:48:47,673][01245] Adding new argument 'enjoy_script'=None that is not in the saved config file! +[2023-02-22 18:48:47,674][01245] Using frameskip 1 and render_action_repeat=4 for evaluation +[2023-02-22 18:48:47,705][01245] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:48:47,711][01245] RunningMeanStd input shape: (3, 72, 128) +[2023-02-22 18:48:47,713][01245] RunningMeanStd input shape: (1,) +[2023-02-22 18:48:47,730][01245] ConvEncoder: input_channels=3 +[2023-02-22 18:48:48,421][01245] Conv encoder output size: 512 +[2023-02-22 18:48:48,422][01245] Policy head output size: 512 +[2023-02-22 18:48:50,843][01245] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... +[2023-02-22 18:48:52,110][01245] Num frames 100... +[2023-02-22 18:48:52,223][01245] Num frames 200... +[2023-02-22 18:48:52,337][01245] Num frames 300... +[2023-02-22 18:48:52,458][01245] Num frames 400... +[2023-02-22 18:48:52,578][01245] Num frames 500... +[2023-02-22 18:48:52,693][01245] Num frames 600... +[2023-02-22 18:48:52,812][01245] Num frames 700... +[2023-02-22 18:48:52,931][01245] Num frames 800... +[2023-02-22 18:48:53,094][01245] Avg episode rewards: #0: 16.960, true rewards: #0: 8.960 +[2023-02-22 18:48:53,097][01245] Avg episode reward: 16.960, avg true_objective: 8.960 +[2023-02-22 18:48:53,106][01245] Num frames 900... +[2023-02-22 18:48:53,217][01245] Num frames 1000... +[2023-02-22 18:48:53,331][01245] Num frames 1100... +[2023-02-22 18:48:53,450][01245] Num frames 1200... +[2023-02-22 18:48:53,569][01245] Num frames 1300... +[2023-02-22 18:48:53,682][01245] Num frames 1400... +[2023-02-22 18:48:53,796][01245] Num frames 1500... +[2023-02-22 18:48:53,914][01245] Num frames 1600... +[2023-02-22 18:48:54,031][01245] Num frames 1700... +[2023-02-22 18:48:54,118][01245] Avg episode rewards: #0: 17.140, true rewards: #0: 8.640 +[2023-02-22 18:48:54,119][01245] Avg episode reward: 17.140, avg true_objective: 8.640 +[2023-02-22 18:48:54,202][01245] Num frames 1800... +[2023-02-22 18:48:54,314][01245] Num frames 1900... +[2023-02-22 18:48:54,431][01245] Num frames 2000... +[2023-02-22 18:48:54,548][01245] Num frames 2100... +[2023-02-22 18:48:54,664][01245] Num frames 2200... +[2023-02-22 18:48:54,737][01245] Avg episode rewards: #0: 14.710, true rewards: #0: 7.377 +[2023-02-22 18:48:54,738][01245] Avg episode reward: 14.710, avg true_objective: 7.377 +[2023-02-22 18:48:54,846][01245] Num frames 2300... +[2023-02-22 18:48:54,961][01245] Num frames 2400... +[2023-02-22 18:48:55,071][01245] Num frames 2500... +[2023-02-22 18:48:55,233][01245] Num frames 2600... +[2023-02-22 18:48:55,412][01245] Avg episode rewards: #0: 13.438, true rewards: #0: 6.687 +[2023-02-22 18:48:55,415][01245] Avg episode reward: 13.438, avg true_objective: 6.687 +[2023-02-22 18:48:55,458][01245] Num frames 2700... +[2023-02-22 18:48:55,631][01245] Num frames 2800... +[2023-02-22 18:48:55,790][01245] Num frames 2900... +[2023-02-22 18:48:55,948][01245] Num frames 3000... +[2023-02-22 18:48:56,111][01245] Num frames 3100... +[2023-02-22 18:48:56,272][01245] Num frames 3200... +[2023-02-22 18:48:56,427][01245] Num frames 3300... +[2023-02-22 18:48:56,588][01245] Num frames 3400... +[2023-02-22 18:48:56,761][01245] Num frames 3500... +[2023-02-22 18:48:56,883][01245] Avg episode rewards: #0: 14.278, true rewards: #0: 7.078 +[2023-02-22 18:48:56,885][01245] Avg episode reward: 14.278, avg true_objective: 7.078 +[2023-02-22 18:48:56,989][01245] Num frames 3600... +[2023-02-22 18:48:57,151][01245] Num frames 3700... +[2023-02-22 18:48:57,325][01245] Num frames 3800... +[2023-02-22 18:48:57,488][01245] Num frames 3900... +[2023-02-22 18:48:57,669][01245] Num frames 4000... +[2023-02-22 18:48:57,837][01245] Num frames 4100... +[2023-02-22 18:48:58,003][01245] Num frames 4200... +[2023-02-22 18:48:58,168][01245] Num frames 4300... +[2023-02-22 18:48:58,336][01245] Num frames 4400... +[2023-02-22 18:48:58,506][01245] Num frames 4500... +[2023-02-22 18:48:58,661][01245] Avg episode rewards: #0: 14.938, true rewards: #0: 7.605 +[2023-02-22 18:48:58,663][01245] Avg episode reward: 14.938, avg true_objective: 7.605 +[2023-02-22 18:48:58,713][01245] Num frames 4600... +[2023-02-22 18:48:58,824][01245] Num frames 4700... +[2023-02-22 18:48:58,939][01245] Num frames 4800... +[2023-02-22 18:48:59,060][01245] Num frames 4900... +[2023-02-22 18:48:59,175][01245] Num frames 5000... +[2023-02-22 18:48:59,288][01245] Num frames 5100... +[2023-02-22 18:48:59,403][01245] Num frames 5200... +[2023-02-22 18:48:59,517][01245] Avg episode rewards: #0: 14.350, true rewards: #0: 7.493 +[2023-02-22 18:48:59,519][01245] Avg episode reward: 14.350, avg true_objective: 7.493 +[2023-02-22 18:48:59,586][01245] Num frames 5300... +[2023-02-22 18:48:59,711][01245] Num frames 5400... +[2023-02-22 18:48:59,826][01245] Num frames 5500... +[2023-02-22 18:48:59,956][01245] Avg episode rewards: #0: 13.331, true rewards: #0: 6.956 +[2023-02-22 18:48:59,958][01245] Avg episode reward: 13.331, avg true_objective: 6.956 +[2023-02-22 18:49:00,004][01245] Num frames 5600... +[2023-02-22 18:49:00,119][01245] Num frames 5700... +[2023-02-22 18:49:00,232][01245] Num frames 5800... +[2023-02-22 18:49:00,347][01245] Num frames 5900... +[2023-02-22 18:49:00,469][01245] Num frames 6000... +[2023-02-22 18:49:00,585][01245] Num frames 6100... +[2023-02-22 18:49:00,700][01245] Num frames 6200... +[2023-02-22 18:49:00,821][01245] Num frames 6300... +[2023-02-22 18:49:00,939][01245] Num frames 6400... +[2023-02-22 18:49:01,054][01245] Num frames 6500... +[2023-02-22 18:49:01,181][01245] Num frames 6600... +[2023-02-22 18:49:01,306][01245] Num frames 6700... +[2023-02-22 18:49:01,422][01245] Num frames 6800... +[2023-02-22 18:49:01,537][01245] Num frames 6900... +[2023-02-22 18:49:01,655][01245] Num frames 7000... +[2023-02-22 18:49:01,776][01245] Num frames 7100... +[2023-02-22 18:49:01,866][01245] Avg episode rewards: #0: 15.588, true rewards: #0: 7.921 +[2023-02-22 18:49:01,868][01245] Avg episode reward: 15.588, avg true_objective: 7.921 +[2023-02-22 18:49:01,950][01245] Num frames 7200... +[2023-02-22 18:49:02,064][01245] Num frames 7300... +[2023-02-22 18:49:02,183][01245] Num frames 7400... +[2023-02-22 18:49:02,297][01245] Num frames 7500... +[2023-02-22 18:49:02,436][01245] Avg episode rewards: #0: 14.577, true rewards: #0: 7.577 +[2023-02-22 18:49:02,438][01245] Avg episode reward: 14.577, avg true_objective: 7.577 +[2023-02-22 18:49:49,879][01245] Replay video saved to /content/train_dir/default_experiment/replay.mp4! +[2023-02-22 18:50:45,670][01245] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json +[2023-02-22 18:50:45,672][01245] Overriding arg 'num_workers' with value 1 passed from command line +[2023-02-22 18:50:45,674][01245] Adding new argument 'no_render'=True that is not in the saved config file! +[2023-02-22 18:50:45,676][01245] Adding new argument 'save_video'=True that is not in the saved config file! +[2023-02-22 18:50:45,679][01245] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! +[2023-02-22 18:50:45,681][01245] Adding new argument 'video_name'=None that is not in the saved config file! +[2023-02-22 18:50:45,682][01245] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! +[2023-02-22 18:50:45,684][01245] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! +[2023-02-22 18:50:45,687][01245] Adding new argument 'push_to_hub'=True that is not in the saved config file! +[2023-02-22 18:50:45,688][01245] Adding new argument 'hf_repository'='NoNameFound/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! +[2023-02-22 18:50:45,690][01245] Adding new argument 'policy_index'=0 that is not in the saved config file! +[2023-02-22 18:50:45,691][01245] Adding new argument 'eval_deterministic'=False that is not in the saved config file! +[2023-02-22 18:50:45,692][01245] Adding new argument 'train_script'=None that is not in the saved config file! +[2023-02-22 18:50:45,694][01245] Adding new argument 'enjoy_script'=None that is not in the saved config file! +[2023-02-22 18:50:45,695][01245] Using frameskip 1 and render_action_repeat=4 for evaluation +[2023-02-22 18:50:45,725][01245] RunningMeanStd input shape: (3, 72, 128) +[2023-02-22 18:50:45,727][01245] RunningMeanStd input shape: (1,) +[2023-02-22 18:50:45,742][01245] ConvEncoder: input_channels=3 +[2023-02-22 18:50:45,781][01245] Conv encoder output size: 512 +[2023-02-22 18:50:45,783][01245] Policy head output size: 512 +[2023-02-22 18:50:45,804][01245] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... +[2023-02-22 18:50:46,261][01245] Num frames 100... +[2023-02-22 18:50:46,379][01245] Num frames 200... +[2023-02-22 18:50:46,507][01245] Num frames 300... +[2023-02-22 18:50:46,630][01245] Num frames 400... +[2023-02-22 18:50:46,753][01245] Num frames 500... +[2023-02-22 18:50:46,877][01245] Num frames 600... +[2023-02-22 18:50:47,000][01245] Num frames 700... +[2023-02-22 18:50:47,062][01245] Avg episode rewards: #0: 13.040, true rewards: #0: 7.040 +[2023-02-22 18:50:47,063][01245] Avg episode reward: 13.040, avg true_objective: 7.040 +[2023-02-22 18:50:47,176][01245] Num frames 800... +[2023-02-22 18:50:47,292][01245] Num frames 900... +[2023-02-22 18:50:47,450][01245] Avg episode rewards: #0: 8.960, true rewards: #0: 4.960 +[2023-02-22 18:50:47,452][01245] Avg episode reward: 8.960, avg true_objective: 4.960 +[2023-02-22 18:50:47,466][01245] Num frames 1000... +[2023-02-22 18:50:47,584][01245] Num frames 1100... +[2023-02-22 18:50:47,718][01245] Num frames 1200... +[2023-02-22 18:50:47,840][01245] Num frames 1300... +[2023-02-22 18:50:47,978][01245] Num frames 1400... +[2023-02-22 18:50:48,107][01245] Num frames 1500... +[2023-02-22 18:50:48,205][01245] Avg episode rewards: #0: 9.120, true rewards: #0: 5.120 +[2023-02-22 18:50:48,207][01245] Avg episode reward: 9.120, avg true_objective: 5.120 +[2023-02-22 18:50:48,282][01245] Num frames 1600... +[2023-02-22 18:50:48,399][01245] Num frames 1700... +[2023-02-22 18:50:48,514][01245] Num frames 1800... +[2023-02-22 18:50:48,636][01245] Num frames 1900... +[2023-02-22 18:50:48,751][01245] Num frames 2000... +[2023-02-22 18:50:48,865][01245] Num frames 2100... +[2023-02-22 18:50:48,946][01245] Avg episode rewards: #0: 10.053, true rewards: #0: 5.302 +[2023-02-22 18:50:48,947][01245] Avg episode reward: 10.053, avg true_objective: 5.302 +[2023-02-22 18:50:49,045][01245] Num frames 2200... +[2023-02-22 18:50:49,161][01245] Num frames 2300... +[2023-02-22 18:50:49,277][01245] Num frames 2400... +[2023-02-22 18:50:49,391][01245] Num frames 2500... +[2023-02-22 18:50:49,510][01245] Num frames 2600... +[2023-02-22 18:50:49,624][01245] Num frames 2700... +[2023-02-22 18:50:49,735][01245] Num frames 2800... +[2023-02-22 18:50:49,820][01245] Avg episode rewards: #0: 10.250, true rewards: #0: 5.650 +[2023-02-22 18:50:49,821][01245] Avg episode reward: 10.250, avg true_objective: 5.650 +[2023-02-22 18:50:49,911][01245] Num frames 2900... +[2023-02-22 18:50:50,031][01245] Num frames 3000... +[2023-02-22 18:50:50,153][01245] Num frames 3100... +[2023-02-22 18:50:50,266][01245] Num frames 3200... +[2023-02-22 18:50:50,390][01245] Num frames 3300... +[2023-02-22 18:50:50,507][01245] Num frames 3400... +[2023-02-22 18:50:50,621][01245] Num frames 3500... +[2023-02-22 18:50:50,733][01245] Num frames 3600... +[2023-02-22 18:50:50,857][01245] Num frames 3700... +[2023-02-22 18:50:50,980][01245] Num frames 3800... +[2023-02-22 18:50:51,098][01245] Num frames 3900... +[2023-02-22 18:50:51,220][01245] Num frames 4000... +[2023-02-22 18:50:51,297][01245] Avg episode rewards: #0: 12.687, true rewards: #0: 6.687 +[2023-02-22 18:50:51,299][01245] Avg episode reward: 12.687, avg true_objective: 6.687 +[2023-02-22 18:50:51,409][01245] Num frames 4100... +[2023-02-22 18:50:51,520][01245] Num frames 4200... +[2023-02-22 18:50:51,634][01245] Num frames 4300... +[2023-02-22 18:50:51,747][01245] Num frames 4400... +[2023-02-22 18:50:51,861][01245] Num frames 4500... +[2023-02-22 18:50:51,978][01245] Num frames 4600... +[2023-02-22 18:50:52,102][01245] Num frames 4700... +[2023-02-22 18:50:52,222][01245] Num frames 4800... +[2023-02-22 18:50:52,329][01245] Avg episode rewards: #0: 13.063, true rewards: #0: 6.920 +[2023-02-22 18:50:52,331][01245] Avg episode reward: 13.063, avg true_objective: 6.920 +[2023-02-22 18:50:52,398][01245] Num frames 4900... +[2023-02-22 18:50:52,512][01245] Num frames 5000... +[2023-02-22 18:50:52,627][01245] Num frames 5100... +[2023-02-22 18:50:52,748][01245] Num frames 5200... +[2023-02-22 18:50:52,861][01245] Num frames 5300... +[2023-02-22 18:50:52,976][01245] Num frames 5400... +[2023-02-22 18:50:53,096][01245] Avg episode rewards: #0: 12.565, true rewards: #0: 6.815 +[2023-02-22 18:50:53,098][01245] Avg episode reward: 12.565, avg true_objective: 6.815 +[2023-02-22 18:50:53,154][01245] Num frames 5500... +[2023-02-22 18:50:53,266][01245] Num frames 5600... +[2023-02-22 18:50:53,382][01245] Num frames 5700... +[2023-02-22 18:50:53,500][01245] Num frames 5800... +[2023-02-22 18:50:53,620][01245] Num frames 5900... +[2023-02-22 18:50:53,739][01245] Num frames 6000... +[2023-02-22 18:50:53,861][01245] Num frames 6100... +[2023-02-22 18:50:53,980][01245] Num frames 6200... +[2023-02-22 18:50:54,111][01245] Num frames 6300... +[2023-02-22 18:50:54,223][01245] Avg episode rewards: #0: 13.498, true rewards: #0: 7.053 +[2023-02-22 18:50:54,225][01245] Avg episode reward: 13.498, avg true_objective: 7.053 +[2023-02-22 18:50:54,300][01245] Num frames 6400... +[2023-02-22 18:50:54,474][01245] Num frames 6500... +[2023-02-22 18:50:54,648][01245] Num frames 6600... +[2023-02-22 18:50:54,810][01245] Num frames 6700... +[2023-02-22 18:50:54,978][01245] Num frames 6800... +[2023-02-22 18:50:55,145][01245] Num frames 6900... +[2023-02-22 18:50:55,311][01245] Num frames 7000... +[2023-02-22 18:50:55,491][01245] Num frames 7100... +[2023-02-22 18:50:55,667][01245] Num frames 7200... +[2023-02-22 18:50:55,745][01245] Avg episode rewards: #0: 13.812, true rewards: #0: 7.212 +[2023-02-22 18:50:55,751][01245] Avg episode reward: 13.812, avg true_objective: 7.212 +[2023-02-22 18:51:41,114][01245] Replay video saved to /content/train_dir/default_experiment/replay.mp4! +[2023-02-22 18:51:46,385][01245] The model has been pushed to https://huggingface.co/NoNameFound/rl_course_vizdoom_health_gathering_supreme +[2023-02-22 18:53:27,528][01245] Environment doom_basic already registered, overwriting... +[2023-02-22 18:53:27,531][01245] Environment doom_two_colors_easy already registered, overwriting... +[2023-02-22 18:53:27,534][01245] Environment doom_two_colors_hard already registered, overwriting... +[2023-02-22 18:53:27,535][01245] Environment doom_dm already registered, overwriting... +[2023-02-22 18:53:27,537][01245] Environment doom_dwango5 already registered, overwriting... +[2023-02-22 18:53:27,539][01245] Environment doom_my_way_home_flat_actions already registered, overwriting... +[2023-02-22 18:53:27,542][01245] Environment doom_defend_the_center_flat_actions already registered, overwriting... +[2023-02-22 18:53:27,543][01245] Environment doom_my_way_home already registered, overwriting... +[2023-02-22 18:53:27,545][01245] Environment doom_deadly_corridor already registered, overwriting... +[2023-02-22 18:53:27,546][01245] Environment doom_defend_the_center already registered, overwriting... +[2023-02-22 18:53:27,548][01245] Environment doom_defend_the_line already registered, overwriting... +[2023-02-22 18:53:27,549][01245] Environment doom_health_gathering already registered, overwriting... +[2023-02-22 18:53:27,551][01245] Environment doom_health_gathering_supreme already registered, overwriting... +[2023-02-22 18:53:27,552][01245] Environment doom_battle already registered, overwriting... +[2023-02-22 18:53:27,554][01245] Environment doom_battle2 already registered, overwriting... +[2023-02-22 18:53:27,555][01245] Environment doom_duel_bots already registered, overwriting... +[2023-02-22 18:53:27,556][01245] Environment doom_deathmatch_bots already registered, overwriting... +[2023-02-22 18:53:27,558][01245] Environment doom_duel already registered, overwriting... +[2023-02-22 18:53:27,559][01245] Environment doom_deathmatch_full already registered, overwriting... +[2023-02-22 18:53:27,561][01245] Environment doom_benchmark already registered, overwriting... +[2023-02-22 18:53:27,562][01245] register_encoder_factory: +[2023-02-22 18:53:27,603][01245] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json +[2023-02-22 18:53:27,605][01245] Overriding arg 'train_for_env_steps' with value 8000000 passed from command line +[2023-02-22 18:53:27,611][01245] Experiment dir /content/train_dir/default_experiment already exists! +[2023-02-22 18:53:27,613][01245] Resuming existing experiment from /content/train_dir/default_experiment... +[2023-02-22 18:53:27,615][01245] Weights and Biases integration disabled +[2023-02-22 18:53:27,622][01245] Environment var CUDA_VISIBLE_DEVICES is 0 + +[2023-02-22 18:53:29,196][01245] Starting experiment with the following configuration: +help=False +algo=APPO +env=doom_health_gathering_supreme +experiment=default_experiment +train_dir=/content/train_dir +restart_behavior=resume +device=gpu +seed=None +num_policies=1 +async_rl=True +serial_mode=False +batched_sampling=False +num_batches_to_accumulate=2 +worker_num_splits=2 +policy_workers_per_policy=1 +max_policy_lag=1000 +num_workers=8 +num_envs_per_worker=4 +batch_size=1024 +num_batches_per_epoch=1 +num_epochs=1 +rollout=32 +recurrence=32 +shuffle_minibatches=False +gamma=0.99 +reward_scale=1.0 +reward_clip=1000.0 +value_bootstrap=False +normalize_returns=True +exploration_loss_coeff=0.001 +value_loss_coeff=0.5 +kl_loss_coeff=0.0 +exploration_loss=symmetric_kl +gae_lambda=0.95 +ppo_clip_ratio=0.1 +ppo_clip_value=0.2 +with_vtrace=False +vtrace_rho=1.0 +vtrace_c=1.0 +optimizer=adam +adam_eps=1e-06 +adam_beta1=0.9 +adam_beta2=0.999 +max_grad_norm=4.0 +learning_rate=0.0001 +lr_schedule=constant +lr_schedule_kl_threshold=0.008 +lr_adaptive_min=1e-06 +lr_adaptive_max=0.01 +obs_subtract_mean=0.0 +obs_scale=255.0 +normalize_input=True +normalize_input_keys=None +decorrelate_experience_max_seconds=0 +decorrelate_envs_on_one_worker=True +actor_worker_gpus=[] +set_workers_cpu_affinity=True +force_envs_single_thread=False +default_niceness=0 +log_to_file=True +experiment_summaries_interval=10 +flush_summaries_interval=30 +stats_avg=100 +summaries_use_frameskip=True +heartbeat_interval=20 +heartbeat_reporting_interval=600 +train_for_env_steps=8000000 +train_for_seconds=10000000000 +save_every_sec=120 +keep_checkpoints=2 +load_checkpoint_kind=latest +save_milestones_sec=-1 +save_best_every_sec=5 +save_best_metric=reward +save_best_after=100000 +benchmark=False +encoder_mlp_layers=[512, 512] +encoder_conv_architecture=convnet_simple +encoder_conv_mlp_layers=[512] +use_rnn=True +rnn_size=512 +rnn_type=gru +rnn_num_layers=1 +decoder_mlp_layers=[] +nonlinearity=elu +policy_initialization=orthogonal +policy_init_gain=1.0 +actor_critic_share_weights=True +adaptive_stddev=True +continuous_tanh_scale=0.0 +initial_stddev=1.0 +use_env_info_cache=False +env_gpu_actions=False +env_gpu_observations=True +env_frameskip=4 +env_framestack=1 +pixel_format=CHW +use_record_episode_statistics=False +with_wandb=False +wandb_user=None +wandb_project=sample_factory +wandb_group=None +wandb_job_type=SF +wandb_tags=[] +with_pbt=False +pbt_mix_policies_in_one_env=True +pbt_period_env_steps=5000000 +pbt_start_mutation=20000000 +pbt_replace_fraction=0.3 +pbt_mutation_rate=0.15 +pbt_replace_reward_gap=0.1 +pbt_replace_reward_gap_absolute=1e-06 +pbt_optimize_gamma=False +pbt_target_objective=true_objective +pbt_perturb_min=1.1 +pbt_perturb_max=1.5 +num_agents=-1 +num_humans=0 +num_bots=-1 +start_bot_difficulty=None +timelimit=None +res_w=128 +res_h=72 +wide_aspect_ratio=False +eval_env_frameskip=1 +fps=35 +command_line=--env=doom_health_gathering_supreme --num_workers=8 --num_envs_per_worker=4 --train_for_env_steps=4000000 +cli_args={'env': 'doom_health_gathering_supreme', 'num_workers': 8, 'num_envs_per_worker': 4, 'train_for_env_steps': 4000000} +git_hash=unknown +git_repo_name=not a git repository +[2023-02-22 18:53:29,201][01245] Saving configuration to /content/train_dir/default_experiment/config.json... +[2023-02-22 18:53:29,204][01245] Rollout worker 0 uses device cpu +[2023-02-22 18:53:29,206][01245] Rollout worker 1 uses device cpu +[2023-02-22 18:53:29,211][01245] Rollout worker 2 uses device cpu +[2023-02-22 18:53:29,213][01245] Rollout worker 3 uses device cpu +[2023-02-22 18:53:29,214][01245] Rollout worker 4 uses device cpu +[2023-02-22 18:53:29,216][01245] Rollout worker 5 uses device cpu +[2023-02-22 18:53:29,217][01245] Rollout worker 6 uses device cpu +[2023-02-22 18:53:29,219][01245] Rollout worker 7 uses device cpu +[2023-02-22 18:53:29,357][01245] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-22 18:53:29,361][01245] InferenceWorker_p0-w0: min num requests: 2 +[2023-02-22 18:53:29,394][01245] Starting all processes... +[2023-02-22 18:53:29,396][01245] Starting process learner_proc0 +[2023-02-22 18:53:29,526][01245] Starting all processes... +[2023-02-22 18:53:29,538][01245] Starting process inference_proc0-0 +[2023-02-22 18:53:29,539][01245] Starting process rollout_proc0 +[2023-02-22 18:53:29,539][01245] Starting process rollout_proc1 +[2023-02-22 18:53:29,539][01245] Starting process rollout_proc2 +[2023-02-22 18:53:29,539][01245] Starting process rollout_proc3 +[2023-02-22 18:53:29,539][01245] Starting process rollout_proc4 +[2023-02-22 18:53:29,539][01245] Starting process rollout_proc5 +[2023-02-22 18:53:29,539][01245] Starting process rollout_proc6 +[2023-02-22 18:53:29,539][01245] Starting process rollout_proc7 +[2023-02-22 18:53:40,546][23673] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-22 18:53:40,547][23673] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0 +[2023-02-22 18:53:40,612][23673] Num visible devices: 1 +[2023-02-22 18:53:40,650][23673] Starting seed is not provided +[2023-02-22 18:53:40,652][23673] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-22 18:53:40,653][23673] Initializing actor-critic model on device cuda:0 +[2023-02-22 18:53:40,654][23673] RunningMeanStd input shape: (3, 72, 128) +[2023-02-22 18:53:40,655][23673] RunningMeanStd input shape: (1,) +[2023-02-22 18:53:40,768][23673] ConvEncoder: input_channels=3 +[2023-02-22 18:53:41,721][23687] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-22 18:53:41,722][23687] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0 +[2023-02-22 18:53:41,772][23673] Conv encoder output size: 512 +[2023-02-22 18:53:41,772][23673] Policy head output size: 512 +[2023-02-22 18:53:41,785][23687] Num visible devices: 1 +[2023-02-22 18:53:41,889][23673] Created Actor Critic model with architecture: +[2023-02-22 18:53:41,893][23673] 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) + ) +) +[2023-02-22 18:53:41,922][23688] Worker 1 uses CPU cores [1] +[2023-02-22 18:53:42,395][23691] Worker 4 uses CPU cores [0] +[2023-02-22 18:53:42,472][23689] Worker 0 uses CPU cores [0] +[2023-02-22 18:53:42,650][23694] Worker 2 uses CPU cores [0] +[2023-02-22 18:53:42,972][23706] Worker 6 uses CPU cores [0] +[2023-02-22 18:53:42,990][23701] Worker 5 uses CPU cores [1] +[2023-02-22 18:53:43,001][23700] Worker 3 uses CPU cores [1] +[2023-02-22 18:53:43,054][23703] Worker 7 uses CPU cores [1] +[2023-02-22 18:53:45,263][23673] Using optimizer +[2023-02-22 18:53:45,264][23673] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... +[2023-02-22 18:53:45,299][23673] Loading model from checkpoint +[2023-02-22 18:53:45,304][23673] Loaded experiment state at self.train_step=978, self.env_steps=4005888 +[2023-02-22 18:53:45,304][23673] Initialized policy 0 weights for model version 978 +[2023-02-22 18:53:45,307][23673] LearnerWorker_p0 finished initialization! +[2023-02-22 18:53:45,309][23673] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-22 18:53:45,521][23687] RunningMeanStd input shape: (3, 72, 128) +[2023-02-22 18:53:45,522][23687] RunningMeanStd input shape: (1,) +[2023-02-22 18:53:45,535][23687] ConvEncoder: input_channels=3 +[2023-02-22 18:53:45,650][23687] Conv encoder output size: 512 +[2023-02-22 18:53:45,650][23687] Policy head output size: 512 +[2023-02-22 18:53:47,622][01245] 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) +[2023-02-22 18:53:48,168][01245] Inference worker 0-0 is ready! +[2023-02-22 18:53:48,171][01245] All inference workers are ready! Signal rollout workers to start! +[2023-02-22 18:53:48,271][23688] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:53:48,270][23701] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:53:48,273][23700] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:53:48,279][23703] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:53:48,302][23689] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:53:48,314][23706] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:53:48,315][23691] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:53:48,337][23694] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-22 18:53:49,352][01245] Heartbeat connected on Batcher_0 +[2023-02-22 18:53:49,353][01245] Heartbeat connected on LearnerWorker_p0 +[2023-02-22 18:53:49,401][01245] Heartbeat connected on InferenceWorker_p0-w0 +[2023-02-22 18:53:50,049][23701] Decorrelating experience for 0 frames... +[2023-02-22 18:53:50,054][23688] Decorrelating experience for 0 frames... +[2023-02-22 18:53:50,056][23703] Decorrelating experience for 0 frames... +[2023-02-22 18:53:50,669][23691] Decorrelating experience for 0 frames... +[2023-02-22 18:53:50,677][23689] Decorrelating experience for 0 frames... +[2023-02-22 18:53:50,683][23706] Decorrelating experience for 0 frames... +[2023-02-22 18:53:50,702][23694] Decorrelating experience for 0 frames... +[2023-02-22 18:53:51,456][23688] Decorrelating experience for 32 frames... +[2023-02-22 18:53:51,498][23700] Decorrelating experience for 0 frames... +[2023-02-22 18:53:52,299][23689] Decorrelating experience for 32 frames... +[2023-02-22 18:53:52,305][23706] Decorrelating experience for 32 frames... +[2023-02-22 18:53:52,401][23694] Decorrelating experience for 32 frames... +[2023-02-22 18:53:52,622][01245] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 4005888. Throughput: 0: 0.0. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) +[2023-02-22 18:53:53,136][23691] Decorrelating experience for 32 frames... +[2023-02-22 18:53:53,437][23700] Decorrelating experience for 32 frames... +[2023-02-22 18:53:53,611][23688] Decorrelating experience for 64 frames... +[2023-02-22 18:53:53,803][23689] Decorrelating experience for 64 frames... +[2023-02-22 18:53:53,970][23701] Decorrelating experience for 32 frames... +[2023-02-22 18:53:55,008][23700] Decorrelating experience for 64 frames... +[2023-02-22 18:53:55,078][23688] Decorrelating experience for 96 frames... +[2023-02-22 18:53:55,341][23694] Decorrelating experience for 64 frames... +[2023-02-22 18:53:55,362][01245] Heartbeat connected on RolloutWorker_w1 +[2023-02-22 18:53:55,738][23701] Decorrelating experience for 64 frames... +[2023-02-22 18:53:55,777][23706] Decorrelating experience for 64 frames... +[2023-02-22 18:53:57,622][01245] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 4005888. Throughput: 0: 0.0. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) +[2023-02-22 18:53:57,748][23691] Decorrelating experience for 64 frames... +[2023-02-22 18:53:57,859][23694] Decorrelating experience for 96 frames... +[2023-02-22 18:53:57,989][23700] Decorrelating experience for 96 frames... +[2023-02-22 18:53:58,047][01245] Heartbeat connected on RolloutWorker_w2 +[2023-02-22 18:53:58,063][23706] Decorrelating experience for 96 frames... +[2023-02-22 18:53:58,323][01245] Heartbeat connected on RolloutWorker_w6 +[2023-02-22 18:53:58,330][01245] Heartbeat connected on RolloutWorker_w3 +[2023-02-22 18:53:58,641][23703] Decorrelating experience for 32 frames... +[2023-02-22 18:54:01,557][23691] Decorrelating experience for 96 frames... +[2023-02-22 18:54:01,596][23701] Decorrelating experience for 96 frames... +[2023-02-22 18:54:02,622][01245] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 4005888. Throughput: 0: 81.6. Samples: 1224. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) +[2023-02-22 18:54:02,637][01245] Avg episode reward: [(0, '3.681')] +[2023-02-22 18:54:03,128][01245] Heartbeat connected on RolloutWorker_w4 +[2023-02-22 18:54:03,269][01245] Heartbeat connected on RolloutWorker_w5 +[2023-02-22 18:54:04,963][23689] Decorrelating experience for 96 frames... +[2023-02-22 18:54:05,277][01245] Heartbeat connected on RolloutWorker_w0 +[2023-02-22 18:54:05,563][23673] Signal inference workers to stop experience collection... +[2023-02-22 18:54:05,585][23687] InferenceWorker_p0-w0: stopping experience collection +[2023-02-22 18:54:05,557][23703] Decorrelating experience for 64 frames... +[2023-02-22 18:54:06,446][23703] Decorrelating experience for 96 frames... +[2023-02-22 18:54:06,596][01245] Heartbeat connected on RolloutWorker_w7 +[2023-02-22 18:54:07,622][01245] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 4005888. Throughput: 0: 127.1. Samples: 2542. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) +[2023-02-22 18:54:07,625][01245] Avg episode reward: [(0, '4.923')] +[2023-02-22 18:54:07,839][23673] Signal inference workers to resume experience collection... +[2023-02-22 18:54:07,842][23687] InferenceWorker_p0-w0: resuming experience collection +[2023-02-22 18:54:12,624][01245] Fps is (10 sec: 1638.0, 60 sec: 655.3, 300 sec: 655.3). Total num frames: 4022272. Throughput: 0: 177.0. Samples: 4426. Policy #0 lag: (min: 1.0, avg: 1.0, max: 1.0) +[2023-02-22 18:54:12,629][01245] Avg episode reward: [(0, '6.192')] +[2023-02-22 18:54:17,623][01245] Fps is (10 sec: 2866.9, 60 sec: 955.7, 300 sec: 955.7). Total num frames: 4034560. Throughput: 0: 214.9. Samples: 6448. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:54:17,631][01245] Avg episode reward: [(0, '9.494')] +[2023-02-22 18:54:21,148][23687] Updated weights for policy 0, policy_version 988 (0.0776) +[2023-02-22 18:54:22,622][01245] Fps is (10 sec: 2867.8, 60 sec: 1287.3, 300 sec: 1287.3). Total num frames: 4050944. Throughput: 0: 311.7. Samples: 10908. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:54:22,629][01245] Avg episode reward: [(0, '10.492')] +[2023-02-22 18:54:27,622][01245] Fps is (10 sec: 3686.8, 60 sec: 1638.4, 300 sec: 1638.4). Total num frames: 4071424. Throughput: 0: 435.0. Samples: 17398. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:54:27,630][01245] Avg episode reward: [(0, '11.435')] +[2023-02-22 18:54:31,272][23687] Updated weights for policy 0, policy_version 998 (0.0025) +[2023-02-22 18:54:32,622][01245] Fps is (10 sec: 3686.5, 60 sec: 1820.4, 300 sec: 1820.4). Total num frames: 4087808. Throughput: 0: 453.8. Samples: 20420. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:54:32,627][01245] Avg episode reward: [(0, '12.610')] +[2023-02-22 18:54:37,622][01245] Fps is (10 sec: 3276.8, 60 sec: 1966.1, 300 sec: 1966.1). Total num frames: 4104192. Throughput: 0: 543.7. Samples: 24468. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:54:37,626][01245] Avg episode reward: [(0, '13.222')] +[2023-02-22 18:54:42,622][01245] Fps is (10 sec: 3276.8, 60 sec: 2085.2, 300 sec: 2085.2). Total num frames: 4120576. Throughput: 0: 650.3. Samples: 29262. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:54:42,628][01245] Avg episode reward: [(0, '13.478')] +[2023-02-22 18:54:42,633][23673] Saving new best policy, reward=13.478! +[2023-02-22 18:54:44,061][23687] Updated weights for policy 0, policy_version 1008 (0.0032) +[2023-02-22 18:54:47,622][01245] Fps is (10 sec: 3686.3, 60 sec: 2252.8, 300 sec: 2252.8). Total num frames: 4141056. Throughput: 0: 691.7. Samples: 32352. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:54:47,629][01245] Avg episode reward: [(0, '13.641')] +[2023-02-22 18:54:47,644][23673] Saving new best policy, reward=13.641! +[2023-02-22 18:54:52,622][01245] Fps is (10 sec: 4096.0, 60 sec: 2594.1, 300 sec: 2394.6). Total num frames: 4161536. Throughput: 0: 798.3. Samples: 38466. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:54:52,624][01245] Avg episode reward: [(0, '13.890')] +[2023-02-22 18:54:52,626][23673] Saving new best policy, reward=13.890! +[2023-02-22 18:54:55,821][23687] Updated weights for policy 0, policy_version 1018 (0.0023) +[2023-02-22 18:54:57,622][01245] Fps is (10 sec: 3276.8, 60 sec: 2798.9, 300 sec: 2399.1). Total num frames: 4173824. Throughput: 0: 845.2. Samples: 42460. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:54:57,627][01245] Avg episode reward: [(0, '13.337')] +[2023-02-22 18:55:02,622][01245] Fps is (10 sec: 2867.1, 60 sec: 3072.0, 300 sec: 2457.6). Total num frames: 4190208. Throughput: 0: 845.7. Samples: 44506. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:55:02,625][01245] Avg episode reward: [(0, '12.629')] +[2023-02-22 18:55:07,317][23687] Updated weights for policy 0, policy_version 1028 (0.0025) +[2023-02-22 18:55:07,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3413.3, 300 sec: 2560.0). Total num frames: 4210688. Throughput: 0: 879.1. Samples: 50468. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:55:07,629][01245] Avg episode reward: [(0, '13.967')] +[2023-02-22 18:55:07,641][23673] Saving new best policy, reward=13.967! +[2023-02-22 18:55:12,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3413.4, 300 sec: 2602.2). Total num frames: 4227072. Throughput: 0: 869.3. Samples: 56516. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:55:12,627][01245] Avg episode reward: [(0, '12.857')] +[2023-02-22 18:55:17,627][01245] Fps is (10 sec: 2865.9, 60 sec: 3413.1, 300 sec: 2594.0). Total num frames: 4239360. Throughput: 0: 846.5. Samples: 58518. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:55:17,641][01245] Avg episode reward: [(0, '13.155')] +[2023-02-22 18:55:20,931][23687] Updated weights for policy 0, policy_version 1038 (0.0042) +[2023-02-22 18:55:22,622][01245] Fps is (10 sec: 2867.3, 60 sec: 3413.3, 300 sec: 2630.1). Total num frames: 4255744. Throughput: 0: 836.7. Samples: 62120. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:55:22,625][01245] Avg episode reward: [(0, '14.301')] +[2023-02-22 18:55:22,629][23673] Saving new best policy, reward=14.301! +[2023-02-22 18:55:27,622][01245] Fps is (10 sec: 3688.1, 60 sec: 3413.3, 300 sec: 2703.4). Total num frames: 4276224. Throughput: 0: 856.8. Samples: 67816. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:55:27,625][01245] Avg episode reward: [(0, '14.735')] +[2023-02-22 18:55:27,637][23673] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001044_4276224.pth... +[2023-02-22 18:55:27,847][23673] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000947_3878912.pth +[2023-02-22 18:55:27,865][23673] Saving new best policy, reward=14.735! +[2023-02-22 18:55:32,624][01245] Fps is (10 sec: 3276.0, 60 sec: 3344.9, 300 sec: 2691.6). Total num frames: 4288512. Throughput: 0: 849.1. Samples: 70564. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:55:32,634][01245] Avg episode reward: [(0, '13.846')] +[2023-02-22 18:55:33,072][23687] Updated weights for policy 0, policy_version 1048 (0.0026) +[2023-02-22 18:55:37,622][01245] Fps is (10 sec: 2457.6, 60 sec: 3276.8, 300 sec: 2681.0). Total num frames: 4300800. Throughput: 0: 790.0. Samples: 74018. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:55:37,625][01245] Avg episode reward: [(0, '13.101')] +[2023-02-22 18:55:42,622][01245] Fps is (10 sec: 2458.2, 60 sec: 3208.5, 300 sec: 2671.3). Total num frames: 4313088. Throughput: 0: 788.0. Samples: 77918. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:55:42,627][01245] Avg episode reward: [(0, '12.064')] +[2023-02-22 18:55:46,853][23687] Updated weights for policy 0, policy_version 1058 (0.0015) +[2023-02-22 18:55:47,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3208.5, 300 sec: 2730.7). Total num frames: 4333568. Throughput: 0: 807.3. Samples: 80834. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:55:47,624][01245] Avg episode reward: [(0, '11.928')] +[2023-02-22 18:55:52,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3208.5, 300 sec: 2785.3). Total num frames: 4354048. Throughput: 0: 815.6. Samples: 87168. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:55:52,628][01245] Avg episode reward: [(0, '12.222')] +[2023-02-22 18:55:57,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3276.8, 300 sec: 2804.2). Total num frames: 4370432. Throughput: 0: 784.9. Samples: 91834. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) +[2023-02-22 18:55:57,628][01245] Avg episode reward: [(0, '12.544')] +[2023-02-22 18:55:58,767][23687] Updated weights for policy 0, policy_version 1068 (0.0017) +[2023-02-22 18:56:02,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3208.5, 300 sec: 2791.4). Total num frames: 4382720. Throughput: 0: 784.5. Samples: 93816. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:56:02,625][01245] Avg episode reward: [(0, '13.893')] +[2023-02-22 18:56:07,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3208.5, 300 sec: 2837.9). Total num frames: 4403200. Throughput: 0: 818.4. Samples: 98950. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:56:07,630][01245] Avg episode reward: [(0, '15.557')] +[2023-02-22 18:56:07,640][23673] Saving new best policy, reward=15.557! +[2023-02-22 18:56:10,248][23687] Updated weights for policy 0, policy_version 1078 (0.0014) +[2023-02-22 18:56:12,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3276.8, 300 sec: 2881.3). Total num frames: 4423680. Throughput: 0: 832.0. Samples: 105258. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) +[2023-02-22 18:56:12,629][01245] Avg episode reward: [(0, '17.085')] +[2023-02-22 18:56:12,636][23673] Saving new best policy, reward=17.085! +[2023-02-22 18:56:17,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3345.3, 300 sec: 2894.5). Total num frames: 4440064. Throughput: 0: 828.5. Samples: 107844. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:56:17,626][01245] Avg episode reward: [(0, '17.355')] +[2023-02-22 18:56:17,636][23673] Saving new best policy, reward=17.355! +[2023-02-22 18:56:22,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3276.8, 300 sec: 2880.4). Total num frames: 4452352. Throughput: 0: 840.0. Samples: 111820. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:56:22,626][01245] Avg episode reward: [(0, '18.295')] +[2023-02-22 18:56:22,628][23673] Saving new best policy, reward=18.295! +[2023-02-22 18:56:23,330][23687] Updated weights for policy 0, policy_version 1088 (0.0025) +[2023-02-22 18:56:27,622][01245] Fps is (10 sec: 2867.1, 60 sec: 3208.5, 300 sec: 2892.8). Total num frames: 4468736. Throughput: 0: 868.3. Samples: 116992. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) +[2023-02-22 18:56:27,631][01245] Avg episode reward: [(0, '16.820')] +[2023-02-22 18:56:32,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3413.5, 300 sec: 2954.1). Total num frames: 4493312. Throughput: 0: 874.4. Samples: 120180. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:56:32,632][01245] Avg episode reward: [(0, '16.132')] +[2023-02-22 18:56:33,469][23687] Updated weights for policy 0, policy_version 1098 (0.0022) +[2023-02-22 18:56:37,622][01245] Fps is (10 sec: 4096.1, 60 sec: 3481.6, 300 sec: 2963.6). Total num frames: 4509696. Throughput: 0: 862.4. Samples: 125976. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) +[2023-02-22 18:56:37,627][01245] Avg episode reward: [(0, '17.119')] +[2023-02-22 18:56:42,629][01245] Fps is (10 sec: 2865.1, 60 sec: 3481.2, 300 sec: 2949.0). Total num frames: 4521984. Throughput: 0: 846.6. Samples: 129936. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:56:42,635][01245] Avg episode reward: [(0, '16.765')] +[2023-02-22 18:56:46,761][23687] Updated weights for policy 0, policy_version 1108 (0.0022) +[2023-02-22 18:56:47,622][01245] Fps is (10 sec: 2867.1, 60 sec: 3413.3, 300 sec: 2958.2). Total num frames: 4538368. Throughput: 0: 848.2. Samples: 131984. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:56:47,628][01245] Avg episode reward: [(0, '16.581')] +[2023-02-22 18:56:52,622][01245] Fps is (10 sec: 3689.1, 60 sec: 3413.3, 300 sec: 2989.0). Total num frames: 4558848. Throughput: 0: 876.8. Samples: 138404. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:56:52,625][01245] Avg episode reward: [(0, '17.825')] +[2023-02-22 18:56:57,255][23687] Updated weights for policy 0, policy_version 1118 (0.0027) +[2023-02-22 18:56:57,622][01245] Fps is (10 sec: 4096.1, 60 sec: 3481.6, 300 sec: 3018.1). Total num frames: 4579328. Throughput: 0: 862.5. Samples: 144072. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:56:57,626][01245] Avg episode reward: [(0, '17.694')] +[2023-02-22 18:57:02,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3481.6, 300 sec: 3003.7). Total num frames: 4591616. Throughput: 0: 850.0. Samples: 146094. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:57:02,630][01245] Avg episode reward: [(0, '16.247')] +[2023-02-22 18:57:07,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3413.3, 300 sec: 3010.6). Total num frames: 4608000. Throughput: 0: 853.6. Samples: 150232. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:57:07,629][01245] Avg episode reward: [(0, '16.285')] +[2023-02-22 18:57:09,883][23687] Updated weights for policy 0, policy_version 1128 (0.0016) +[2023-02-22 18:57:12,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3413.3, 300 sec: 3037.0). Total num frames: 4628480. Throughput: 0: 886.8. Samples: 156896. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:57:12,628][01245] Avg episode reward: [(0, '14.952')] +[2023-02-22 18:57:17,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3481.6, 300 sec: 3062.2). Total num frames: 4648960. Throughput: 0: 888.9. Samples: 160182. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:57:17,625][01245] Avg episode reward: [(0, '14.650')] +[2023-02-22 18:57:21,046][23687] Updated weights for policy 0, policy_version 1138 (0.0025) +[2023-02-22 18:57:22,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3067.2). Total num frames: 4665344. Throughput: 0: 860.1. Samples: 164682. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:57:22,631][01245] Avg episode reward: [(0, '15.659')] +[2023-02-22 18:57:27,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3481.6, 300 sec: 3053.4). Total num frames: 4677632. Throughput: 0: 870.1. Samples: 169086. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:57:27,629][01245] Avg episode reward: [(0, '15.175')] +[2023-02-22 18:57:27,644][23673] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001142_4677632.pth... +[2023-02-22 18:57:27,817][23673] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth +[2023-02-22 18:57:32,105][23687] Updated weights for policy 0, policy_version 1148 (0.0012) +[2023-02-22 18:57:32,622][01245] Fps is (10 sec: 3686.3, 60 sec: 3481.6, 300 sec: 3094.8). Total num frames: 4702208. Throughput: 0: 899.8. Samples: 172474. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:57:32,628][01245] Avg episode reward: [(0, '16.371')] +[2023-02-22 18:57:37,622][01245] Fps is (10 sec: 4505.6, 60 sec: 3549.9, 300 sec: 3116.5). Total num frames: 4722688. Throughput: 0: 911.0. Samples: 179398. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:57:37,628][01245] Avg episode reward: [(0, '16.308')] +[2023-02-22 18:57:42,622][01245] Fps is (10 sec: 3686.5, 60 sec: 3618.6, 300 sec: 3119.9). Total num frames: 4739072. Throughput: 0: 884.4. Samples: 183870. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 18:57:42,628][01245] Avg episode reward: [(0, '16.524')] +[2023-02-22 18:57:43,429][23687] Updated weights for policy 0, policy_version 1158 (0.0014) +[2023-02-22 18:57:47,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3549.9, 300 sec: 3106.1). Total num frames: 4751360. Throughput: 0: 883.2. Samples: 185840. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:57:47,624][01245] Avg episode reward: [(0, '16.484')] +[2023-02-22 18:57:52,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3126.3). Total num frames: 4771840. Throughput: 0: 905.6. Samples: 190984. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:57:52,625][01245] Avg episode reward: [(0, '17.797')] +[2023-02-22 18:57:55,599][23687] Updated weights for policy 0, policy_version 1168 (0.0017) +[2023-02-22 18:57:57,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3549.9, 300 sec: 3145.7). Total num frames: 4792320. Throughput: 0: 886.5. Samples: 196788. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:57:57,628][01245] Avg episode reward: [(0, '17.586')] +[2023-02-22 18:58:02,625][01245] Fps is (10 sec: 3276.0, 60 sec: 3549.7, 300 sec: 3132.2). Total num frames: 4804608. Throughput: 0: 859.2. Samples: 198846. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:58:02,629][01245] Avg episode reward: [(0, '18.662')] +[2023-02-22 18:58:02,631][23673] Saving new best policy, reward=18.662! +[2023-02-22 18:58:07,623][01245] Fps is (10 sec: 2457.4, 60 sec: 3481.6, 300 sec: 3119.3). Total num frames: 4816896. Throughput: 0: 852.0. Samples: 203022. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:58:07,631][01245] Avg episode reward: [(0, '18.198')] +[2023-02-22 18:58:09,081][23687] Updated weights for policy 0, policy_version 1178 (0.0045) +[2023-02-22 18:58:12,622][01245] Fps is (10 sec: 3277.6, 60 sec: 3481.6, 300 sec: 3137.7). Total num frames: 4837376. Throughput: 0: 884.4. Samples: 208886. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:58:12,628][01245] Avg episode reward: [(0, '18.582')] +[2023-02-22 18:58:17,622][01245] Fps is (10 sec: 4505.9, 60 sec: 3549.9, 300 sec: 3170.6). Total num frames: 4861952. Throughput: 0: 883.9. Samples: 212250. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:58:17,628][01245] Avg episode reward: [(0, '18.192')] +[2023-02-22 18:58:18,400][23687] Updated weights for policy 0, policy_version 1188 (0.0016) +[2023-02-22 18:58:22,624][01245] Fps is (10 sec: 3685.8, 60 sec: 3481.5, 300 sec: 3157.6). Total num frames: 4874240. Throughput: 0: 848.1. Samples: 217566. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:58:22,629][01245] Avg episode reward: [(0, '16.704')] +[2023-02-22 18:58:27,622][01245] Fps is (10 sec: 2457.6, 60 sec: 3481.6, 300 sec: 3145.1). Total num frames: 4886528. Throughput: 0: 833.1. Samples: 221358. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:58:27,627][01245] Avg episode reward: [(0, '16.094')] +[2023-02-22 18:58:32,478][23687] Updated weights for policy 0, policy_version 1198 (0.0018) +[2023-02-22 18:58:32,626][01245] Fps is (10 sec: 3276.1, 60 sec: 3413.1, 300 sec: 3161.8). Total num frames: 4907008. Throughput: 0: 841.1. Samples: 223692. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:58:32,630][01245] Avg episode reward: [(0, '15.365')] +[2023-02-22 18:58:37,627][01245] Fps is (10 sec: 3275.4, 60 sec: 3276.6, 300 sec: 3149.6). Total num frames: 4919296. Throughput: 0: 829.9. Samples: 228332. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:58:37,629][01245] Avg episode reward: [(0, '14.900')] +[2023-02-22 18:58:42,622][01245] Fps is (10 sec: 2458.5, 60 sec: 3208.5, 300 sec: 3138.0). Total num frames: 4931584. Throughput: 0: 784.6. Samples: 232094. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 18:58:42,625][01245] Avg episode reward: [(0, '14.420')] +[2023-02-22 18:58:47,622][01245] Fps is (10 sec: 2458.7, 60 sec: 3208.5, 300 sec: 3179.6). Total num frames: 4943872. Throughput: 0: 780.4. Samples: 233962. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:58:47,626][01245] Avg episode reward: [(0, '14.356')] +[2023-02-22 18:58:48,294][23687] Updated weights for policy 0, policy_version 1208 (0.0021) +[2023-02-22 18:58:52,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3140.3, 300 sec: 3235.1). Total num frames: 4960256. Throughput: 0: 779.5. Samples: 238098. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:58:52,625][01245] Avg episode reward: [(0, '14.878')] +[2023-02-22 18:58:57,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3140.3, 300 sec: 3304.6). Total num frames: 4980736. Throughput: 0: 788.6. Samples: 244374. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:58:57,625][01245] Avg episode reward: [(0, '16.011')] +[2023-02-22 18:58:59,063][23687] Updated weights for policy 0, policy_version 1218 (0.0021) +[2023-02-22 18:59:02,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3276.9, 300 sec: 3374.0). Total num frames: 5001216. Throughput: 0: 785.6. Samples: 247600. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 18:59:02,629][01245] Avg episode reward: [(0, '16.420')] +[2023-02-22 18:59:07,625][01245] Fps is (10 sec: 3275.7, 60 sec: 3276.7, 300 sec: 3360.1). Total num frames: 5013504. Throughput: 0: 770.1. Samples: 252224. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:59:07,628][01245] Avg episode reward: [(0, '16.866')] +[2023-02-22 18:59:12,181][23687] Updated weights for policy 0, policy_version 1228 (0.0020) +[2023-02-22 18:59:12,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3208.5, 300 sec: 3374.0). Total num frames: 5029888. Throughput: 0: 780.8. Samples: 256494. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:59:12,625][01245] Avg episode reward: [(0, '17.693')] +[2023-02-22 18:59:17,622][01245] Fps is (10 sec: 3687.6, 60 sec: 3140.3, 300 sec: 3387.9). Total num frames: 5050368. Throughput: 0: 797.7. Samples: 259586. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:59:17,630][01245] Avg episode reward: [(0, '18.712')] +[2023-02-22 18:59:17,641][23673] Saving new best policy, reward=18.712! +[2023-02-22 18:59:21,898][23687] Updated weights for policy 0, policy_version 1238 (0.0021) +[2023-02-22 18:59:22,624][01245] Fps is (10 sec: 4095.3, 60 sec: 3276.8, 300 sec: 3387.9). Total num frames: 5070848. Throughput: 0: 837.7. Samples: 266028. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:59:22,636][01245] Avg episode reward: [(0, '18.860')] +[2023-02-22 18:59:22,638][23673] Saving new best policy, reward=18.860! +[2023-02-22 18:59:27,624][01245] Fps is (10 sec: 3276.2, 60 sec: 3276.7, 300 sec: 3374.0). Total num frames: 5083136. Throughput: 0: 847.9. Samples: 270250. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 18:59:27,626][01245] Avg episode reward: [(0, '18.566')] +[2023-02-22 18:59:27,652][23673] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001241_5083136.pth... +[2023-02-22 18:59:27,814][23673] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001044_4276224.pth +[2023-02-22 18:59:32,622][01245] Fps is (10 sec: 2867.7, 60 sec: 3208.7, 300 sec: 3374.0). Total num frames: 5099520. Throughput: 0: 850.1. Samples: 272218. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 18:59:32,630][01245] Avg episode reward: [(0, '18.343')] +[2023-02-22 18:59:35,412][23687] Updated weights for policy 0, policy_version 1248 (0.0012) +[2023-02-22 18:59:37,622][01245] Fps is (10 sec: 3687.1, 60 sec: 3345.3, 300 sec: 3387.9). Total num frames: 5120000. Throughput: 0: 881.6. Samples: 277772. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:59:37,624][01245] Avg episode reward: [(0, '18.261')] +[2023-02-22 18:59:42,622][01245] Fps is (10 sec: 4095.9, 60 sec: 3481.6, 300 sec: 3387.9). Total num frames: 5140480. Throughput: 0: 877.5. Samples: 283862. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 18:59:42,630][01245] Avg episode reward: [(0, '16.137')] +[2023-02-22 18:59:46,914][23687] Updated weights for policy 0, policy_version 1258 (0.0013) +[2023-02-22 18:59:47,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3481.6, 300 sec: 3360.1). Total num frames: 5152768. Throughput: 0: 854.0. Samples: 286028. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:59:47,625][01245] Avg episode reward: [(0, '14.584')] +[2023-02-22 18:59:52,622][01245] Fps is (10 sec: 2457.7, 60 sec: 3413.3, 300 sec: 3360.1). Total num frames: 5165056. Throughput: 0: 843.7. Samples: 290188. Policy #0 lag: (min: 0.0, avg: 0.3, max: 2.0) +[2023-02-22 18:59:52,624][01245] Avg episode reward: [(0, '14.569')] +[2023-02-22 18:59:57,622][01245] Fps is (10 sec: 3276.9, 60 sec: 3413.3, 300 sec: 3374.0). Total num frames: 5185536. Throughput: 0: 873.8. Samples: 295816. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 18:59:57,627][01245] Avg episode reward: [(0, '14.572')] +[2023-02-22 18:59:58,775][23687] Updated weights for policy 0, policy_version 1268 (0.0030) +[2023-02-22 19:00:02,622][01245] Fps is (10 sec: 4505.6, 60 sec: 3481.6, 300 sec: 3387.9). Total num frames: 5210112. Throughput: 0: 877.5. Samples: 299072. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:00:02,625][01245] Avg episode reward: [(0, '14.714')] +[2023-02-22 19:00:07,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3481.8, 300 sec: 3374.0). Total num frames: 5222400. Throughput: 0: 855.9. Samples: 304540. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:00:07,628][01245] Avg episode reward: [(0, '14.489')] +[2023-02-22 19:00:10,877][23687] Updated weights for policy 0, policy_version 1278 (0.0012) +[2023-02-22 19:00:12,625][01245] Fps is (10 sec: 2866.3, 60 sec: 3481.4, 300 sec: 3387.9). Total num frames: 5238784. Throughput: 0: 853.0. Samples: 308634. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:00:12,629][01245] Avg episode reward: [(0, '14.858')] +[2023-02-22 19:00:17,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3413.3, 300 sec: 3387.9). Total num frames: 5255168. Throughput: 0: 862.8. Samples: 311042. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:00:17,625][01245] Avg episode reward: [(0, '16.745')] +[2023-02-22 19:00:21,619][23687] Updated weights for policy 0, policy_version 1288 (0.0020) +[2023-02-22 19:00:22,622][01245] Fps is (10 sec: 4097.3, 60 sec: 3481.7, 300 sec: 3401.8). Total num frames: 5279744. Throughput: 0: 885.0. Samples: 317596. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 19:00:22,625][01245] Avg episode reward: [(0, '16.098')] +[2023-02-22 19:00:27,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3550.0, 300 sec: 3415.7). Total num frames: 5296128. Throughput: 0: 869.6. Samples: 322994. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:00:27,625][01245] Avg episode reward: [(0, '17.049')] +[2023-02-22 19:00:32,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3481.6, 300 sec: 3415.6). Total num frames: 5308416. Throughput: 0: 866.1. Samples: 325004. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) +[2023-02-22 19:00:32,632][01245] Avg episode reward: [(0, '17.174')] +[2023-02-22 19:00:34,906][23687] Updated weights for policy 0, policy_version 1298 (0.0013) +[2023-02-22 19:00:37,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3413.3, 300 sec: 3429.5). Total num frames: 5324800. Throughput: 0: 874.7. Samples: 329550. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:00:37,630][01245] Avg episode reward: [(0, '16.772')] +[2023-02-22 19:00:42,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3481.6, 300 sec: 3443.4). Total num frames: 5349376. Throughput: 0: 895.3. Samples: 336104. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 19:00:42,629][01245] Avg episode reward: [(0, '16.310')] +[2023-02-22 19:00:44,315][23687] Updated weights for policy 0, policy_version 1308 (0.0018) +[2023-02-22 19:00:47,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3549.9, 300 sec: 3429.5). Total num frames: 5365760. Throughput: 0: 895.6. Samples: 339372. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:00:47,626][01245] Avg episode reward: [(0, '15.922')] +[2023-02-22 19:00:52,625][01245] Fps is (10 sec: 2866.2, 60 sec: 3549.7, 300 sec: 3415.6). Total num frames: 5378048. Throughput: 0: 864.8. Samples: 343458. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:00:52,628][01245] Avg episode reward: [(0, '16.103')] +[2023-02-22 19:00:57,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3481.6, 300 sec: 3429.5). Total num frames: 5394432. Throughput: 0: 879.6. Samples: 348212. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:00:57,625][01245] Avg episode reward: [(0, '16.822')] +[2023-02-22 19:00:57,811][23687] Updated weights for policy 0, policy_version 1318 (0.0028) +[2023-02-22 19:01:02,622][01245] Fps is (10 sec: 4097.4, 60 sec: 3481.6, 300 sec: 3443.4). Total num frames: 5419008. Throughput: 0: 898.6. Samples: 351480. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) +[2023-02-22 19:01:02,625][01245] Avg episode reward: [(0, '17.576')] +[2023-02-22 19:01:07,623][01245] Fps is (10 sec: 4095.8, 60 sec: 3549.8, 300 sec: 3429.5). Total num frames: 5435392. Throughput: 0: 891.5. Samples: 357714. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:01:07,628][01245] Avg episode reward: [(0, '17.673')] +[2023-02-22 19:01:08,138][23687] Updated weights for policy 0, policy_version 1328 (0.0013) +[2023-02-22 19:01:12,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3550.1, 300 sec: 3429.5). Total num frames: 5451776. Throughput: 0: 863.8. Samples: 361864. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 19:01:12,629][01245] Avg episode reward: [(0, '18.268')] +[2023-02-22 19:01:17,622][01245] Fps is (10 sec: 3276.9, 60 sec: 3549.9, 300 sec: 3443.4). Total num frames: 5468160. Throughput: 0: 865.0. Samples: 363930. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:01:17,628][01245] Avg episode reward: [(0, '18.193')] +[2023-02-22 19:01:20,357][23687] Updated weights for policy 0, policy_version 1338 (0.0017) +[2023-02-22 19:01:22,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3481.6, 300 sec: 3457.3). Total num frames: 5488640. Throughput: 0: 899.9. Samples: 370044. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:01:22,625][01245] Avg episode reward: [(0, '18.586')] +[2023-02-22 19:01:27,623][01245] Fps is (10 sec: 4095.8, 60 sec: 3549.8, 300 sec: 3443.4). Total num frames: 5509120. Throughput: 0: 889.7. Samples: 376142. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) +[2023-02-22 19:01:27,625][01245] Avg episode reward: [(0, '18.314')] +[2023-02-22 19:01:27,648][23673] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001345_5509120.pth... +[2023-02-22 19:01:27,875][23673] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001142_4677632.pth +[2023-02-22 19:01:31,763][23687] Updated weights for policy 0, policy_version 1348 (0.0014) +[2023-02-22 19:01:32,629][01245] Fps is (10 sec: 3274.4, 60 sec: 3549.4, 300 sec: 3429.4). Total num frames: 5521408. Throughput: 0: 861.5. Samples: 378148. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:01:32,632][01245] Avg episode reward: [(0, '18.959')] +[2023-02-22 19:01:32,634][23673] Saving new best policy, reward=18.959! +[2023-02-22 19:01:37,622][01245] Fps is (10 sec: 2457.7, 60 sec: 3481.6, 300 sec: 3429.6). Total num frames: 5533696. Throughput: 0: 863.9. Samples: 382332. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:01:37,624][01245] Avg episode reward: [(0, '19.547')] +[2023-02-22 19:01:37,683][23673] Saving new best policy, reward=19.547! +[2023-02-22 19:01:42,622][01245] Fps is (10 sec: 3689.1, 60 sec: 3481.6, 300 sec: 3457.3). Total num frames: 5558272. Throughput: 0: 897.2. Samples: 388586. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:01:42,624][01245] Avg episode reward: [(0, '19.118')] +[2023-02-22 19:01:43,234][23687] Updated weights for policy 0, policy_version 1358 (0.0021) +[2023-02-22 19:01:47,622][01245] Fps is (10 sec: 4505.6, 60 sec: 3549.9, 300 sec: 3457.3). Total num frames: 5578752. Throughput: 0: 900.4. Samples: 392000. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:01:47,626][01245] Avg episode reward: [(0, '20.131')] +[2023-02-22 19:01:47,637][23673] Saving new best policy, reward=20.131! +[2023-02-22 19:01:52,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3618.3, 300 sec: 3443.4). Total num frames: 5595136. Throughput: 0: 875.7. Samples: 397120. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:01:52,632][01245] Avg episode reward: [(0, '20.542')] +[2023-02-22 19:01:52,636][23673] Saving new best policy, reward=20.542! +[2023-02-22 19:01:55,012][23687] Updated weights for policy 0, policy_version 1368 (0.0023) +[2023-02-22 19:01:57,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3549.9, 300 sec: 3443.4). Total num frames: 5607424. Throughput: 0: 882.8. Samples: 401592. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:01:57,629][01245] Avg episode reward: [(0, '21.238')] +[2023-02-22 19:01:57,660][23673] Saving new best policy, reward=21.238! +[2023-02-22 19:02:02,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3471.2). Total num frames: 5632000. Throughput: 0: 909.5. Samples: 404858. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:02:02,625][01245] Avg episode reward: [(0, '20.626')] +[2023-02-22 19:02:04,793][23687] Updated weights for policy 0, policy_version 1378 (0.0014) +[2023-02-22 19:02:07,624][01245] Fps is (10 sec: 4914.1, 60 sec: 3686.3, 300 sec: 3485.0). Total num frames: 5656576. Throughput: 0: 926.5. Samples: 411740. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:02:07,627][01245] Avg episode reward: [(0, '21.225')] +[2023-02-22 19:02:12,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3618.1, 300 sec: 3457.3). Total num frames: 5668864. Throughput: 0: 903.6. Samples: 416804. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:02:12,626][01245] Avg episode reward: [(0, '20.656')] +[2023-02-22 19:02:17,023][23687] Updated weights for policy 0, policy_version 1388 (0.0028) +[2023-02-22 19:02:17,622][01245] Fps is (10 sec: 2867.8, 60 sec: 3618.1, 300 sec: 3457.3). Total num frames: 5685248. Throughput: 0: 908.5. Samples: 419026. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:02:17,626][01245] Avg episode reward: [(0, '20.274')] +[2023-02-22 19:02:22,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3618.1, 300 sec: 3485.1). Total num frames: 5705728. Throughput: 0: 944.8. Samples: 424850. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 19:02:22,625][01245] Avg episode reward: [(0, '19.932')] +[2023-02-22 19:02:25,902][23687] Updated weights for policy 0, policy_version 1398 (0.0019) +[2023-02-22 19:02:27,622][01245] Fps is (10 sec: 4915.3, 60 sec: 3754.7, 300 sec: 3499.0). Total num frames: 5734400. Throughput: 0: 966.9. Samples: 432098. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 19:02:27,625][01245] Avg episode reward: [(0, '20.148')] +[2023-02-22 19:02:32,629][01245] Fps is (10 sec: 4093.0, 60 sec: 3754.7, 300 sec: 3471.1). Total num frames: 5746688. Throughput: 0: 948.6. Samples: 434696. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) +[2023-02-22 19:02:32,631][01245] Avg episode reward: [(0, '20.074')] +[2023-02-22 19:02:37,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3822.9, 300 sec: 3471.2). Total num frames: 5763072. Throughput: 0: 936.5. Samples: 439264. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:02:37,626][01245] Avg episode reward: [(0, '20.666')] +[2023-02-22 19:02:38,046][23687] Updated weights for policy 0, policy_version 1408 (0.0030) +[2023-02-22 19:02:42,622][01245] Fps is (10 sec: 4099.0, 60 sec: 3822.9, 300 sec: 3512.8). Total num frames: 5787648. Throughput: 0: 979.0. Samples: 445648. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 19:02:42,626][01245] Avg episode reward: [(0, '20.593')] +[2023-02-22 19:02:46,764][23687] Updated weights for policy 0, policy_version 1418 (0.0015) +[2023-02-22 19:02:47,622][01245] Fps is (10 sec: 4505.6, 60 sec: 3822.9, 300 sec: 3512.8). Total num frames: 5808128. Throughput: 0: 983.6. Samples: 449118. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:02:47,624][01245] Avg episode reward: [(0, '19.143')] +[2023-02-22 19:02:52,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3499.0). Total num frames: 5824512. Throughput: 0: 954.9. Samples: 454710. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:02:52,628][01245] Avg episode reward: [(0, '19.065')] +[2023-02-22 19:02:57,623][01245] Fps is (10 sec: 3276.4, 60 sec: 3891.1, 300 sec: 3512.9). Total num frames: 5840896. Throughput: 0: 942.2. Samples: 459204. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:02:57,627][01245] Avg episode reward: [(0, '19.622')] +[2023-02-22 19:02:59,410][23687] Updated weights for policy 0, policy_version 1428 (0.0020) +[2023-02-22 19:03:02,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3540.6). Total num frames: 5861376. Throughput: 0: 963.8. Samples: 462396. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:03:02,628][01245] Avg episode reward: [(0, '20.028')] +[2023-02-22 19:03:07,622][01245] Fps is (10 sec: 4506.2, 60 sec: 3823.1, 300 sec: 3554.5). Total num frames: 5885952. Throughput: 0: 989.9. Samples: 469394. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:03:07,626][01245] Avg episode reward: [(0, '19.900')] +[2023-02-22 19:03:08,848][23687] Updated weights for policy 0, policy_version 1438 (0.0024) +[2023-02-22 19:03:12,624][01245] Fps is (10 sec: 3685.6, 60 sec: 3822.8, 300 sec: 3512.8). Total num frames: 5898240. Throughput: 0: 923.1. Samples: 473638. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:03:12,633][01245] Avg episode reward: [(0, '19.406')] +[2023-02-22 19:03:17,622][01245] Fps is (10 sec: 2457.6, 60 sec: 3754.7, 300 sec: 3512.9). Total num frames: 5910528. Throughput: 0: 904.7. Samples: 475400. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:03:17,627][01245] Avg episode reward: [(0, '18.568')] +[2023-02-22 19:03:22,622][01245] Fps is (10 sec: 2458.2, 60 sec: 3618.1, 300 sec: 3512.8). Total num frames: 5922816. Throughput: 0: 885.8. Samples: 479126. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:03:22,629][01245] Avg episode reward: [(0, '18.448')] +[2023-02-22 19:03:24,218][23687] Updated weights for policy 0, policy_version 1448 (0.0024) +[2023-02-22 19:03:27,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3526.8). Total num frames: 5947392. Throughput: 0: 886.9. Samples: 485560. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:03:27,629][01245] Avg episode reward: [(0, '17.609')] +[2023-02-22 19:03:27,641][23673] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001452_5947392.pth... +[2023-02-22 19:03:27,777][23673] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001241_5083136.pth +[2023-02-22 19:03:32,622][01245] Fps is (10 sec: 4505.6, 60 sec: 3686.9, 300 sec: 3554.5). Total num frames: 5967872. Throughput: 0: 887.7. Samples: 489066. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:03:32,624][01245] Avg episode reward: [(0, '18.054')] +[2023-02-22 19:03:32,780][23687] Updated weights for policy 0, policy_version 1458 (0.0016) +[2023-02-22 19:03:37,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3686.4, 300 sec: 3568.4). Total num frames: 5984256. Throughput: 0: 892.1. Samples: 494856. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:03:37,625][01245] Avg episode reward: [(0, '17.516')] +[2023-02-22 19:03:42,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3582.3). Total num frames: 6000640. Throughput: 0: 893.4. Samples: 499404. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:03:42,625][01245] Avg episode reward: [(0, '18.419')] +[2023-02-22 19:03:44,974][23687] Updated weights for policy 0, policy_version 1468 (0.0019) +[2023-02-22 19:03:47,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3596.1). Total num frames: 6021120. Throughput: 0: 892.4. Samples: 502552. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:03:47,624][01245] Avg episode reward: [(0, '18.836')] +[2023-02-22 19:03:52,622][01245] Fps is (10 sec: 4505.6, 60 sec: 3686.4, 300 sec: 3610.0). Total num frames: 6045696. Throughput: 0: 883.5. Samples: 509150. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:03:52,630][01245] Avg episode reward: [(0, '18.159')] +[2023-02-22 19:03:54,954][23687] Updated weights for policy 0, policy_version 1478 (0.0022) +[2023-02-22 19:03:57,622][01245] Fps is (10 sec: 4095.9, 60 sec: 3686.5, 300 sec: 3596.1). Total num frames: 6062080. Throughput: 0: 901.8. Samples: 514216. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:03:57,629][01245] Avg episode reward: [(0, '18.390')] +[2023-02-22 19:04:02,623][01245] Fps is (10 sec: 2866.9, 60 sec: 3549.8, 300 sec: 3596.2). Total num frames: 6074368. Throughput: 0: 911.5. Samples: 516420. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 19:04:02,627][01245] Avg episode reward: [(0, '19.194')] +[2023-02-22 19:04:06,641][23687] Updated weights for policy 0, policy_version 1488 (0.0012) +[2023-02-22 19:04:07,622][01245] Fps is (10 sec: 3686.5, 60 sec: 3549.9, 300 sec: 3623.9). Total num frames: 6098944. Throughput: 0: 958.8. Samples: 522272. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:04:07,629][01245] Avg episode reward: [(0, '17.924')] +[2023-02-22 19:04:12,622][01245] Fps is (10 sec: 4506.0, 60 sec: 3686.5, 300 sec: 3623.9). Total num frames: 6119424. Throughput: 0: 967.4. Samples: 529092. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:04:12,630][01245] Avg episode reward: [(0, '19.083')] +[2023-02-22 19:04:16,901][23687] Updated weights for policy 0, policy_version 1498 (0.0017) +[2023-02-22 19:04:17,629][01245] Fps is (10 sec: 3684.0, 60 sec: 3754.3, 300 sec: 3610.0). Total num frames: 6135808. Throughput: 0: 947.8. Samples: 531724. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:04:17,637][01245] Avg episode reward: [(0, '18.410')] +[2023-02-22 19:04:22,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3754.7, 300 sec: 3610.1). Total num frames: 6148096. Throughput: 0: 913.5. Samples: 535962. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:04:22,624][01245] Avg episode reward: [(0, '17.753')] +[2023-02-22 19:04:27,622][01245] Fps is (10 sec: 3688.8, 60 sec: 3754.7, 300 sec: 3637.8). Total num frames: 6172672. Throughput: 0: 946.2. Samples: 541982. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:04:27,626][01245] Avg episode reward: [(0, '18.077')] +[2023-02-22 19:04:28,238][23687] Updated weights for policy 0, policy_version 1508 (0.0026) +[2023-02-22 19:04:32,622][01245] Fps is (10 sec: 4915.3, 60 sec: 3822.9, 300 sec: 3651.7). Total num frames: 6197248. Throughput: 0: 952.6. Samples: 545420. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 19:04:32,625][01245] Avg episode reward: [(0, '17.679')] +[2023-02-22 19:04:37,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3754.7, 300 sec: 3623.9). Total num frames: 6209536. Throughput: 0: 934.7. Samples: 551212. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:04:37,626][01245] Avg episode reward: [(0, '17.384')] +[2023-02-22 19:04:39,489][23687] Updated weights for policy 0, policy_version 1518 (0.0012) +[2023-02-22 19:04:42,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3754.7, 300 sec: 3637.8). Total num frames: 6225920. Throughput: 0: 916.2. Samples: 555446. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 19:04:42,627][01245] Avg episode reward: [(0, '18.617')] +[2023-02-22 19:04:47,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3754.7, 300 sec: 3665.6). Total num frames: 6246400. Throughput: 0: 929.9. Samples: 558264. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) +[2023-02-22 19:04:47,626][01245] Avg episode reward: [(0, '18.497')] +[2023-02-22 19:04:50,222][23687] Updated weights for policy 0, policy_version 1528 (0.0015) +[2023-02-22 19:04:52,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3686.4, 300 sec: 3665.6). Total num frames: 6266880. Throughput: 0: 948.9. Samples: 564972. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 19:04:52,625][01245] Avg episode reward: [(0, '19.467')] +[2023-02-22 19:04:57,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3686.4, 300 sec: 3637.8). Total num frames: 6283264. Throughput: 0: 918.2. Samples: 570410. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 19:04:57,625][01245] Avg episode reward: [(0, '20.402')] +[2023-02-22 19:05:01,984][23687] Updated weights for policy 0, policy_version 1538 (0.0014) +[2023-02-22 19:05:02,626][01245] Fps is (10 sec: 3275.6, 60 sec: 3754.5, 300 sec: 3651.6). Total num frames: 6299648. Throughput: 0: 908.5. Samples: 572604. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 19:05:02,630][01245] Avg episode reward: [(0, '20.726')] +[2023-02-22 19:05:07,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3686.4, 300 sec: 3665.6). Total num frames: 6320128. Throughput: 0: 935.7. Samples: 578070. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:05:07,625][01245] Avg episode reward: [(0, '20.306')] +[2023-02-22 19:05:11,548][23687] Updated weights for policy 0, policy_version 1548 (0.0013) +[2023-02-22 19:05:12,622][01245] Fps is (10 sec: 4507.2, 60 sec: 3754.7, 300 sec: 3693.3). Total num frames: 6344704. Throughput: 0: 956.8. Samples: 585036. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:05:12,630][01245] Avg episode reward: [(0, '21.531')] +[2023-02-22 19:05:12,633][23673] Saving new best policy, reward=21.531! +[2023-02-22 19:05:17,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3755.1, 300 sec: 3665.6). Total num frames: 6361088. Throughput: 0: 939.6. Samples: 587702. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:05:17,631][01245] Avg episode reward: [(0, '20.196')] +[2023-02-22 19:05:22,625][01245] Fps is (10 sec: 2866.2, 60 sec: 3754.5, 300 sec: 3651.6). Total num frames: 6373376. Throughput: 0: 905.4. Samples: 591956. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:05:22,630][01245] Avg episode reward: [(0, '20.033')] +[2023-02-22 19:05:24,448][23687] Updated weights for policy 0, policy_version 1558 (0.0016) +[2023-02-22 19:05:27,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3686.4, 300 sec: 3679.5). Total num frames: 6393856. Throughput: 0: 935.0. Samples: 597522. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:05:27,627][01245] Avg episode reward: [(0, '19.475')] +[2023-02-22 19:05:27,635][23673] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001561_6393856.pth... +[2023-02-22 19:05:27,803][23673] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001345_5509120.pth +[2023-02-22 19:05:32,622][01245] Fps is (10 sec: 4507.1, 60 sec: 3686.4, 300 sec: 3707.2). Total num frames: 6418432. Throughput: 0: 946.3. Samples: 600848. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:05:32,627][01245] Avg episode reward: [(0, '21.814')] +[2023-02-22 19:05:32,631][23673] Saving new best policy, reward=21.814! +[2023-02-22 19:05:33,504][23687] Updated weights for policy 0, policy_version 1568 (0.0018) +[2023-02-22 19:05:37,626][01245] Fps is (10 sec: 4094.3, 60 sec: 3754.4, 300 sec: 3679.4). Total num frames: 6434816. Throughput: 0: 927.3. Samples: 606704. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) +[2023-02-22 19:05:37,629][01245] Avg episode reward: [(0, '21.153')] +[2023-02-22 19:05:42,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3686.4, 300 sec: 3665.6). Total num frames: 6447104. Throughput: 0: 902.6. Samples: 611026. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 19:05:42,629][01245] Avg episode reward: [(0, '21.437')] +[2023-02-22 19:05:46,554][23687] Updated weights for policy 0, policy_version 1578 (0.0016) +[2023-02-22 19:05:47,622][01245] Fps is (10 sec: 3278.2, 60 sec: 3686.4, 300 sec: 3693.4). Total num frames: 6467584. Throughput: 0: 911.1. Samples: 613600. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 19:05:47,628][01245] Avg episode reward: [(0, '21.165')] +[2023-02-22 19:05:52,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3686.4, 300 sec: 3707.2). Total num frames: 6488064. Throughput: 0: 935.9. Samples: 620184. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 19:05:52,627][01245] Avg episode reward: [(0, '21.039')] +[2023-02-22 19:05:56,692][23687] Updated weights for policy 0, policy_version 1588 (0.0017) +[2023-02-22 19:05:57,625][01245] Fps is (10 sec: 3685.5, 60 sec: 3686.3, 300 sec: 3679.4). Total num frames: 6504448. Throughput: 0: 902.8. Samples: 625666. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) +[2023-02-22 19:05:57,633][01245] Avg episode reward: [(0, '19.776')] +[2023-02-22 19:06:02,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3686.6, 300 sec: 3679.5). Total num frames: 6520832. Throughput: 0: 889.9. Samples: 627748. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:06:02,628][01245] Avg episode reward: [(0, '19.073')] +[2023-02-22 19:06:07,622][01245] Fps is (10 sec: 3277.6, 60 sec: 3618.1, 300 sec: 3679.5). Total num frames: 6537216. Throughput: 0: 908.3. Samples: 632826. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) +[2023-02-22 19:06:07,625][01245] Avg episode reward: [(0, '18.846')] +[2023-02-22 19:06:08,499][23687] Updated weights for policy 0, policy_version 1598 (0.0041) +[2023-02-22 19:06:12,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3618.1, 300 sec: 3707.2). Total num frames: 6561792. Throughput: 0: 934.1. Samples: 639558. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:06:12,629][01245] Avg episode reward: [(0, '20.517')] +[2023-02-22 19:06:17,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3618.1, 300 sec: 3693.3). Total num frames: 6578176. Throughput: 0: 921.2. Samples: 642304. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:06:17,625][01245] Avg episode reward: [(0, '21.103')] +[2023-02-22 19:06:20,406][23687] Updated weights for policy 0, policy_version 1608 (0.0021) +[2023-02-22 19:06:22,623][01245] Fps is (10 sec: 2866.8, 60 sec: 3618.3, 300 sec: 3665.6). Total num frames: 6590464. Throughput: 0: 876.4. Samples: 646138. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:06:22,627][01245] Avg episode reward: [(0, '21.056')] +[2023-02-22 19:06:27,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3549.9, 300 sec: 3679.5). Total num frames: 6606848. Throughput: 0: 876.0. Samples: 650444. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:06:27,631][01245] Avg episode reward: [(0, '20.307')] +[2023-02-22 19:06:32,456][23687] Updated weights for policy 0, policy_version 1618 (0.0019) +[2023-02-22 19:06:32,622][01245] Fps is (10 sec: 3686.9, 60 sec: 3481.6, 300 sec: 3707.2). Total num frames: 6627328. Throughput: 0: 885.9. Samples: 653466. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:06:32,631][01245] Avg episode reward: [(0, '21.004')] +[2023-02-22 19:06:37,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3481.9, 300 sec: 3679.5). Total num frames: 6643712. Throughput: 0: 879.0. Samples: 659740. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:06:37,625][01245] Avg episode reward: [(0, '20.840')] +[2023-02-22 19:06:42,622][01245] Fps is (10 sec: 3276.7, 60 sec: 3549.8, 300 sec: 3665.6). Total num frames: 6660096. Throughput: 0: 857.9. Samples: 664272. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) +[2023-02-22 19:06:42,630][01245] Avg episode reward: [(0, '19.415')] +[2023-02-22 19:06:44,752][23687] Updated weights for policy 0, policy_version 1628 (0.0023) +[2023-02-22 19:06:47,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3679.5). Total num frames: 6680576. Throughput: 0: 867.2. Samples: 666770. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:06:47,625][01245] Avg episode reward: [(0, '18.251')] +[2023-02-22 19:06:52,622][01245] Fps is (10 sec: 4505.8, 60 sec: 3618.1, 300 sec: 3721.1). Total num frames: 6705152. Throughput: 0: 913.4. Samples: 673928. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 19:06:52,624][01245] Avg episode reward: [(0, '18.232')] +[2023-02-22 19:06:53,452][23687] Updated weights for policy 0, policy_version 1638 (0.0013) +[2023-02-22 19:06:57,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3618.3, 300 sec: 3693.3). Total num frames: 6721536. Throughput: 0: 895.4. Samples: 679850. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 19:06:57,628][01245] Avg episode reward: [(0, '17.615')] +[2023-02-22 19:07:02,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3618.1, 300 sec: 3665.6). Total num frames: 6737920. Throughput: 0: 885.1. Samples: 682132. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 19:07:02,624][01245] Avg episode reward: [(0, '16.921')] +[2023-02-22 19:07:05,565][23687] Updated weights for policy 0, policy_version 1648 (0.0011) +[2023-02-22 19:07:07,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3686.4, 300 sec: 3693.3). Total num frames: 6758400. Throughput: 0: 912.7. Samples: 687210. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:07:07,630][01245] Avg episode reward: [(0, '18.643')] +[2023-02-22 19:07:12,622][01245] Fps is (10 sec: 4505.6, 60 sec: 3686.4, 300 sec: 3721.1). Total num frames: 6782976. Throughput: 0: 977.3. Samples: 694424. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:07:12,627][01245] Avg episode reward: [(0, '19.696')] +[2023-02-22 19:07:14,251][23687] Updated weights for policy 0, policy_version 1658 (0.0013) +[2023-02-22 19:07:17,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3686.4, 300 sec: 3707.2). Total num frames: 6799360. Throughput: 0: 982.5. Samples: 697680. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:07:17,627][01245] Avg episode reward: [(0, '21.289')] +[2023-02-22 19:07:22,622][01245] Fps is (10 sec: 3276.7, 60 sec: 3754.7, 300 sec: 3665.6). Total num frames: 6815744. Throughput: 0: 939.5. Samples: 702018. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:07:22,628][01245] Avg episode reward: [(0, '21.535')] +[2023-02-22 19:07:26,835][23687] Updated weights for policy 0, policy_version 1668 (0.0034) +[2023-02-22 19:07:27,623][01245] Fps is (10 sec: 3276.8, 60 sec: 3754.7, 300 sec: 3679.5). Total num frames: 6832128. Throughput: 0: 962.7. Samples: 707594. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:07:27,627][01245] Avg episode reward: [(0, '21.724')] +[2023-02-22 19:07:27,726][23673] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001669_6836224.pth... +[2023-02-22 19:07:27,847][23673] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001452_5947392.pth +[2023-02-22 19:07:32,622][01245] Fps is (10 sec: 4096.1, 60 sec: 3822.9, 300 sec: 3707.2). Total num frames: 6856704. Throughput: 0: 985.3. Samples: 711108. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:07:32,624][01245] Avg episode reward: [(0, '22.433')] +[2023-02-22 19:07:32,627][23673] Saving new best policy, reward=22.433! +[2023-02-22 19:07:36,534][23687] Updated weights for policy 0, policy_version 1678 (0.0017) +[2023-02-22 19:07:37,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3679.5). Total num frames: 6873088. Throughput: 0: 964.3. Samples: 717322. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 19:07:37,625][01245] Avg episode reward: [(0, '22.034')] +[2023-02-22 19:07:42,623][01245] Fps is (10 sec: 2867.0, 60 sec: 3754.6, 300 sec: 3651.7). Total num frames: 6885376. Throughput: 0: 912.6. Samples: 720916. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 19:07:42,627][01245] Avg episode reward: [(0, '22.118')] +[2023-02-22 19:07:47,622][01245] Fps is (10 sec: 2457.6, 60 sec: 3618.1, 300 sec: 3637.8). Total num frames: 6897664. Throughput: 0: 899.6. Samples: 722616. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:07:47,626][01245] Avg episode reward: [(0, '22.141')] +[2023-02-22 19:07:51,396][23687] Updated weights for policy 0, policy_version 1688 (0.0011) +[2023-02-22 19:07:52,622][01245] Fps is (10 sec: 3277.0, 60 sec: 3549.9, 300 sec: 3651.7). Total num frames: 6918144. Throughput: 0: 893.8. Samples: 727432. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:07:52,625][01245] Avg episode reward: [(0, '21.608')] +[2023-02-22 19:07:57,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3618.1, 300 sec: 3651.7). Total num frames: 6938624. Throughput: 0: 884.6. Samples: 734232. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:07:57,628][01245] Avg episode reward: [(0, '20.782')] +[2023-02-22 19:08:02,039][23687] Updated weights for policy 0, policy_version 1698 (0.0021) +[2023-02-22 19:08:02,628][01245] Fps is (10 sec: 3684.2, 60 sec: 3617.8, 300 sec: 3623.8). Total num frames: 6955008. Throughput: 0: 866.6. Samples: 736680. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:08:02,639][01245] Avg episode reward: [(0, '20.446')] +[2023-02-22 19:08:07,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3481.6, 300 sec: 3623.9). Total num frames: 6967296. Throughput: 0: 848.7. Samples: 740210. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:08:07,627][01245] Avg episode reward: [(0, '20.462')] +[2023-02-22 19:08:12,622][01245] Fps is (10 sec: 2868.9, 60 sec: 3345.1, 300 sec: 3637.8). Total num frames: 6983680. Throughput: 0: 824.4. Samples: 744694. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:08:12,626][01245] Avg episode reward: [(0, '19.896')] +[2023-02-22 19:08:15,906][23687] Updated weights for policy 0, policy_version 1708 (0.0021) +[2023-02-22 19:08:17,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3345.1, 300 sec: 3651.7). Total num frames: 7000064. Throughput: 0: 807.3. Samples: 747436. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:08:17,625][01245] Avg episode reward: [(0, '19.813')] +[2023-02-22 19:08:22,623][01245] Fps is (10 sec: 3276.4, 60 sec: 3345.0, 300 sec: 3623.9). Total num frames: 7016448. Throughput: 0: 780.2. Samples: 752430. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:08:22,629][01245] Avg episode reward: [(0, '19.111')] +[2023-02-22 19:08:27,622][01245] Fps is (10 sec: 2457.6, 60 sec: 3208.5, 300 sec: 3582.3). Total num frames: 7024640. Throughput: 0: 777.5. Samples: 755904. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 19:08:27,632][01245] Avg episode reward: [(0, '19.952')] +[2023-02-22 19:08:31,314][23687] Updated weights for policy 0, policy_version 1718 (0.0022) +[2023-02-22 19:08:32,622][01245] Fps is (10 sec: 2457.9, 60 sec: 3072.0, 300 sec: 3582.3). Total num frames: 7041024. Throughput: 0: 777.9. Samples: 757620. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 19:08:32,625][01245] Avg episode reward: [(0, '20.552')] +[2023-02-22 19:08:37,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3140.3, 300 sec: 3596.1). Total num frames: 7061504. Throughput: 0: 797.3. Samples: 763312. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 19:08:37,625][01245] Avg episode reward: [(0, '20.761')] +[2023-02-22 19:08:41,857][23687] Updated weights for policy 0, policy_version 1728 (0.0043) +[2023-02-22 19:08:42,627][01245] Fps is (10 sec: 3684.5, 60 sec: 3208.3, 300 sec: 3582.2). Total num frames: 7077888. Throughput: 0: 765.6. Samples: 768690. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:08:42,631][01245] Avg episode reward: [(0, '21.568')] +[2023-02-22 19:08:47,627][01245] Fps is (10 sec: 2865.7, 60 sec: 3208.2, 300 sec: 3540.5). Total num frames: 7090176. Throughput: 0: 752.9. Samples: 770558. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:08:47,634][01245] Avg episode reward: [(0, '21.411')] +[2023-02-22 19:08:52,622][01245] Fps is (10 sec: 2458.9, 60 sec: 3072.0, 300 sec: 3526.7). Total num frames: 7102464. Throughput: 0: 756.8. Samples: 774266. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:08:52,625][01245] Avg episode reward: [(0, '21.555')] +[2023-02-22 19:08:55,793][23687] Updated weights for policy 0, policy_version 1738 (0.0025) +[2023-02-22 19:08:57,622][01245] Fps is (10 sec: 3278.5, 60 sec: 3072.0, 300 sec: 3554.5). Total num frames: 7122944. Throughput: 0: 792.5. Samples: 780356. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:08:57,627][01245] Avg episode reward: [(0, '22.522')] +[2023-02-22 19:08:57,697][23673] Saving new best policy, reward=22.522! +[2023-02-22 19:09:02,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3140.6, 300 sec: 3540.6). Total num frames: 7143424. Throughput: 0: 798.0. Samples: 783344. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 19:09:02,625][01245] Avg episode reward: [(0, '21.533')] +[2023-02-22 19:09:07,627][01245] Fps is (10 sec: 3275.4, 60 sec: 3140.0, 300 sec: 3512.8). Total num frames: 7155712. Throughput: 0: 786.9. Samples: 787842. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:09:07,629][01245] Avg episode reward: [(0, '21.111')] +[2023-02-22 19:09:07,733][23687] Updated weights for policy 0, policy_version 1748 (0.0013) +[2023-02-22 19:09:12,626][01245] Fps is (10 sec: 2456.6, 60 sec: 3071.8, 300 sec: 3499.0). Total num frames: 7168000. Throughput: 0: 787.7. Samples: 791354. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) +[2023-02-22 19:09:12,629][01245] Avg episode reward: [(0, '21.087')] +[2023-02-22 19:09:17,622][01245] Fps is (10 sec: 3278.2, 60 sec: 3140.3, 300 sec: 3526.7). Total num frames: 7188480. Throughput: 0: 809.4. Samples: 794044. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 19:09:17,628][01245] Avg episode reward: [(0, '21.990')] +[2023-02-22 19:09:20,324][23687] Updated weights for policy 0, policy_version 1758 (0.0016) +[2023-02-22 19:09:22,622][01245] Fps is (10 sec: 4097.6, 60 sec: 3208.6, 300 sec: 3512.8). Total num frames: 7208960. Throughput: 0: 819.2. Samples: 800174. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:09:22,631][01245] Avg episode reward: [(0, '20.928')] +[2023-02-22 19:09:27,625][01245] Fps is (10 sec: 3276.0, 60 sec: 3276.7, 300 sec: 3471.2). Total num frames: 7221248. Throughput: 0: 800.4. Samples: 804704. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:09:27,628][01245] Avg episode reward: [(0, '21.023')] +[2023-02-22 19:09:27,643][23673] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001763_7221248.pth... +[2023-02-22 19:09:27,859][23673] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001561_6393856.pth +[2023-02-22 19:09:32,622][01245] Fps is (10 sec: 2457.6, 60 sec: 3208.5, 300 sec: 3471.2). Total num frames: 7233536. Throughput: 0: 801.0. Samples: 806600. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:09:32,625][01245] Avg episode reward: [(0, '22.182')] +[2023-02-22 19:09:34,270][23687] Updated weights for policy 0, policy_version 1768 (0.0020) +[2023-02-22 19:09:37,622][01245] Fps is (10 sec: 3277.6, 60 sec: 3208.5, 300 sec: 3485.1). Total num frames: 7254016. Throughput: 0: 828.3. Samples: 811540. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:09:37,625][01245] Avg episode reward: [(0, '22.534')] +[2023-02-22 19:09:37,636][23673] Saving new best policy, reward=22.534! +[2023-02-22 19:09:42,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3277.1, 300 sec: 3485.1). Total num frames: 7274496. Throughput: 0: 821.5. Samples: 817324. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:09:42,630][01245] Avg episode reward: [(0, '22.378')] +[2023-02-22 19:09:45,062][23687] Updated weights for policy 0, policy_version 1778 (0.0017) +[2023-02-22 19:09:47,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3277.1, 300 sec: 3457.3). Total num frames: 7286784. Throughput: 0: 810.2. Samples: 819804. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:09:47,626][01245] Avg episode reward: [(0, '22.209')] +[2023-02-22 19:09:52,622][01245] Fps is (10 sec: 2457.6, 60 sec: 3276.8, 300 sec: 3443.4). Total num frames: 7299072. Throughput: 0: 798.7. Samples: 823782. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:09:52,629][01245] Avg episode reward: [(0, '22.723')] +[2023-02-22 19:09:52,634][23673] Saving new best policy, reward=22.723! +[2023-02-22 19:09:57,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3208.5, 300 sec: 3443.5). Total num frames: 7315456. Throughput: 0: 828.1. Samples: 828614. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:09:57,626][01245] Avg episode reward: [(0, '23.626')] +[2023-02-22 19:09:57,736][23673] Saving new best policy, reward=23.626! +[2023-02-22 19:09:58,795][23687] Updated weights for policy 0, policy_version 1788 (0.0028) +[2023-02-22 19:10:02,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3208.5, 300 sec: 3443.4). Total num frames: 7335936. Throughput: 0: 832.0. Samples: 831486. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 19:10:02,631][01245] Avg episode reward: [(0, '22.484')] +[2023-02-22 19:10:07,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3277.0, 300 sec: 3415.6). Total num frames: 7352320. Throughput: 0: 811.8. Samples: 836706. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:10:07,625][01245] Avg episode reward: [(0, '21.993')] +[2023-02-22 19:10:11,597][23687] Updated weights for policy 0, policy_version 1798 (0.0033) +[2023-02-22 19:10:12,622][01245] Fps is (10 sec: 2867.1, 60 sec: 3277.0, 300 sec: 3401.8). Total num frames: 7364608. Throughput: 0: 795.1. Samples: 840482. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:10:12,629][01245] Avg episode reward: [(0, '22.051')] +[2023-02-22 19:10:17,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3208.5, 300 sec: 3415.7). Total num frames: 7380992. Throughput: 0: 790.9. Samples: 842190. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:10:17,624][01245] Avg episode reward: [(0, '21.398')] +[2023-02-22 19:10:22,623][01245] Fps is (10 sec: 3686.3, 60 sec: 3208.5, 300 sec: 3415.6). Total num frames: 7401472. Throughput: 0: 812.7. Samples: 848110. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:10:22,627][01245] Avg episode reward: [(0, '21.749')] +[2023-02-22 19:10:23,411][23687] Updated weights for policy 0, policy_version 1808 (0.0025) +[2023-02-22 19:10:27,623][01245] Fps is (10 sec: 3686.2, 60 sec: 3276.9, 300 sec: 3387.9). Total num frames: 7417856. Throughput: 0: 810.7. Samples: 853806. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:10:27,629][01245] Avg episode reward: [(0, '21.139')] +[2023-02-22 19:10:32,622][01245] Fps is (10 sec: 2867.3, 60 sec: 3276.8, 300 sec: 3374.0). Total num frames: 7430144. Throughput: 0: 802.7. Samples: 855928. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:10:32,630][01245] Avg episode reward: [(0, '20.798')] +[2023-02-22 19:10:36,839][23687] Updated weights for policy 0, policy_version 1818 (0.0029) +[2023-02-22 19:10:37,622][01245] Fps is (10 sec: 2867.4, 60 sec: 3208.5, 300 sec: 3387.9). Total num frames: 7446528. Throughput: 0: 804.1. Samples: 859968. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:10:37,630][01245] Avg episode reward: [(0, '21.175')] +[2023-02-22 19:10:42,624][01245] Fps is (10 sec: 3685.7, 60 sec: 3208.4, 300 sec: 3387.9). Total num frames: 7467008. Throughput: 0: 834.0. Samples: 866148. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:10:42,629][01245] Avg episode reward: [(0, '20.609')] +[2023-02-22 19:10:47,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3276.8, 300 sec: 3374.0). Total num frames: 7483392. Throughput: 0: 838.5. Samples: 869220. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 19:10:47,625][01245] Avg episode reward: [(0, '21.277')] +[2023-02-22 19:10:47,652][23687] Updated weights for policy 0, policy_version 1828 (0.0014) +[2023-02-22 19:10:52,622][01245] Fps is (10 sec: 3277.5, 60 sec: 3345.1, 300 sec: 3374.0). Total num frames: 7499776. Throughput: 0: 816.0. Samples: 873424. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) +[2023-02-22 19:10:52,628][01245] Avg episode reward: [(0, '21.226')] +[2023-02-22 19:10:57,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3276.8, 300 sec: 3360.1). Total num frames: 7512064. Throughput: 0: 818.5. Samples: 877314. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 19:10:57,627][01245] Avg episode reward: [(0, '21.364')] +[2023-02-22 19:11:00,869][23687] Updated weights for policy 0, policy_version 1838 (0.0013) +[2023-02-22 19:11:02,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3276.8, 300 sec: 3374.0). Total num frames: 7532544. Throughput: 0: 850.9. Samples: 880482. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:11:02,627][01245] Avg episode reward: [(0, '21.409')] +[2023-02-22 19:11:07,622][01245] Fps is (10 sec: 4505.6, 60 sec: 3413.3, 300 sec: 3374.0). Total num frames: 7557120. Throughput: 0: 869.0. Samples: 887214. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:11:07,629][01245] Avg episode reward: [(0, '21.764')] +[2023-02-22 19:11:12,275][23687] Updated weights for policy 0, policy_version 1848 (0.0012) +[2023-02-22 19:11:12,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3413.4, 300 sec: 3360.1). Total num frames: 7569408. Throughput: 0: 836.7. Samples: 891458. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:11:12,624][01245] Avg episode reward: [(0, '22.855')] +[2023-02-22 19:11:17,622][01245] Fps is (10 sec: 2457.6, 60 sec: 3345.1, 300 sec: 3360.1). Total num frames: 7581696. Throughput: 0: 834.0. Samples: 893456. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:11:17,630][01245] Avg episode reward: [(0, '21.116')] +[2023-02-22 19:11:22,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3345.1, 300 sec: 3374.0). Total num frames: 7602176. Throughput: 0: 862.9. Samples: 898798. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:11:22,625][01245] Avg episode reward: [(0, '19.979')] +[2023-02-22 19:11:23,906][23687] Updated weights for policy 0, policy_version 1858 (0.0034) +[2023-02-22 19:11:27,622][01245] Fps is (10 sec: 4096.0, 60 sec: 3413.4, 300 sec: 3374.0). Total num frames: 7622656. Throughput: 0: 857.9. Samples: 904752. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:11:27,625][01245] Avg episode reward: [(0, '21.264')] +[2023-02-22 19:11:27,641][23673] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001861_7622656.pth... +[2023-02-22 19:11:27,851][23673] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001669_6836224.pth +[2023-02-22 19:11:32,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3413.3, 300 sec: 3360.1). Total num frames: 7634944. Throughput: 0: 833.8. Samples: 906742. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 19:11:32,625][01245] Avg episode reward: [(0, '20.515')] +[2023-02-22 19:11:37,622][01245] Fps is (10 sec: 2457.6, 60 sec: 3345.1, 300 sec: 3346.2). Total num frames: 7647232. Throughput: 0: 827.0. Samples: 910640. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:11:37,627][01245] Avg episode reward: [(0, '20.319')] +[2023-02-22 19:11:38,503][23687] Updated weights for policy 0, policy_version 1868 (0.0027) +[2023-02-22 19:11:42,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3345.2, 300 sec: 3346.2). Total num frames: 7667712. Throughput: 0: 856.4. Samples: 915852. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:11:42,624][01245] Avg episode reward: [(0, '20.675')] +[2023-02-22 19:11:47,622][01245] Fps is (10 sec: 4096.1, 60 sec: 3413.3, 300 sec: 3332.3). Total num frames: 7688192. Throughput: 0: 857.6. Samples: 919074. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:11:47,626][01245] Avg episode reward: [(0, '21.669')] +[2023-02-22 19:11:47,866][23687] Updated weights for policy 0, policy_version 1878 (0.0022) +[2023-02-22 19:11:52,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3413.3, 300 sec: 3332.3). Total num frames: 7704576. Throughput: 0: 829.1. Samples: 924522. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:11:52,626][01245] Avg episode reward: [(0, '21.786')] +[2023-02-22 19:11:57,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3413.3, 300 sec: 3318.5). Total num frames: 7716864. Throughput: 0: 824.9. Samples: 928578. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:11:57,627][01245] Avg episode reward: [(0, '21.583')] +[2023-02-22 19:12:01,546][23687] Updated weights for policy 0, policy_version 1888 (0.0014) +[2023-02-22 19:12:02,622][01245] Fps is (10 sec: 3276.7, 60 sec: 3413.3, 300 sec: 3318.5). Total num frames: 7737344. Throughput: 0: 831.8. Samples: 930888. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:12:02,627][01245] Avg episode reward: [(0, '21.565')] +[2023-02-22 19:12:07,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3208.5, 300 sec: 3276.8). Total num frames: 7749632. Throughput: 0: 825.1. Samples: 935926. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:12:07,629][01245] Avg episode reward: [(0, '22.627')] +[2023-02-22 19:12:12,625][01245] Fps is (10 sec: 2456.9, 60 sec: 3208.4, 300 sec: 3262.9). Total num frames: 7761920. Throughput: 0: 770.4. Samples: 939422. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:12:12,628][01245] Avg episode reward: [(0, '22.177')] +[2023-02-22 19:12:17,589][23687] Updated weights for policy 0, policy_version 1898 (0.0015) +[2023-02-22 19:12:17,622][01245] Fps is (10 sec: 2457.6, 60 sec: 3208.5, 300 sec: 3249.0). Total num frames: 7774208. Throughput: 0: 762.1. Samples: 941038. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:12:17,627][01245] Avg episode reward: [(0, '22.273')] +[2023-02-22 19:12:22,622][01245] Fps is (10 sec: 2458.4, 60 sec: 3072.0, 300 sec: 3235.1). Total num frames: 7786496. Throughput: 0: 758.6. Samples: 944778. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-22 19:12:22,624][01245] Avg episode reward: [(0, '21.856')] +[2023-02-22 19:12:27,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3072.0, 300 sec: 3221.3). Total num frames: 7806976. Throughput: 0: 765.5. Samples: 950300. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:12:27,624][01245] Avg episode reward: [(0, '21.342')] +[2023-02-22 19:12:29,726][23687] Updated weights for policy 0, policy_version 1908 (0.0014) +[2023-02-22 19:12:32,626][01245] Fps is (10 sec: 3685.0, 60 sec: 3140.1, 300 sec: 3221.2). Total num frames: 7823360. Throughput: 0: 759.4. Samples: 953250. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-22 19:12:32,630][01245] Avg episode reward: [(0, '20.725')] +[2023-02-22 19:12:37,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3208.5, 300 sec: 3235.2). Total num frames: 7839744. Throughput: 0: 746.5. Samples: 958114. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:12:37,627][01245] Avg episode reward: [(0, '20.625')] +[2023-02-22 19:12:42,622][01245] Fps is (10 sec: 2868.2, 60 sec: 3072.0, 300 sec: 3235.1). Total num frames: 7852032. Throughput: 0: 745.8. Samples: 962138. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-22 19:12:42,630][01245] Avg episode reward: [(0, '19.938')] +[2023-02-22 19:12:43,641][23687] Updated weights for policy 0, policy_version 1918 (0.0021) +[2023-02-22 19:12:47,622][01245] Fps is (10 sec: 3276.8, 60 sec: 3072.0, 300 sec: 3235.1). Total num frames: 7872512. Throughput: 0: 754.9. Samples: 964858. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:12:47,631][01245] Avg episode reward: [(0, '20.377')] +[2023-02-22 19:12:52,622][01245] Fps is (10 sec: 4096.1, 60 sec: 3140.3, 300 sec: 3235.1). Total num frames: 7892992. Throughput: 0: 787.2. Samples: 971350. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:12:52,628][01245] Avg episode reward: [(0, '21.290')] +[2023-02-22 19:12:53,128][23687] Updated weights for policy 0, policy_version 1928 (0.0035) +[2023-02-22 19:12:57,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3208.5, 300 sec: 3235.2). Total num frames: 7909376. Throughput: 0: 818.6. Samples: 976254. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:12:57,625][01245] Avg episode reward: [(0, '21.253')] +[2023-02-22 19:13:02,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3072.0, 300 sec: 3235.1). Total num frames: 7921664. Throughput: 0: 823.1. Samples: 978078. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 19:13:02,633][01245] Avg episode reward: [(0, '21.279')] +[2023-02-22 19:13:07,160][23687] Updated weights for policy 0, policy_version 1938 (0.0021) +[2023-02-22 19:13:07,622][01245] Fps is (10 sec: 2867.2, 60 sec: 3140.3, 300 sec: 3235.1). Total num frames: 7938048. Throughput: 0: 839.2. Samples: 982544. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-22 19:13:07,624][01245] Avg episode reward: [(0, '22.048')] +[2023-02-22 19:13:12,622][01245] Fps is (10 sec: 3686.5, 60 sec: 3277.0, 300 sec: 3249.0). Total num frames: 7958528. Throughput: 0: 862.8. Samples: 989126. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-22 19:13:12,631][01245] Avg episode reward: [(0, '21.679')] +[2023-02-22 19:13:17,622][01245] Fps is (10 sec: 3686.4, 60 sec: 3345.1, 300 sec: 3249.0). Total num frames: 7974912. Throughput: 0: 863.3. Samples: 992096. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-22 19:13:17,625][01245] Avg episode reward: [(0, '21.754')] +[2023-02-22 19:13:17,793][23687] Updated weights for policy 0, policy_version 1948 (0.0018) +[2023-02-22 19:13:22,628][01245] Fps is (10 sec: 3274.7, 60 sec: 3413.0, 300 sec: 3276.7). Total num frames: 7991296. Throughput: 0: 843.0. Samples: 996056. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-22 19:13:22,637][01245] Avg episode reward: [(0, '20.810')] +[2023-02-22 19:13:27,587][23673] Stopping Batcher_0... +[2023-02-22 19:13:27,593][23673] Loop batcher_evt_loop terminating... +[2023-02-22 19:13:27,598][23673] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001955_8007680.pth... +[2023-02-22 19:13:27,593][01245] Component Batcher_0 stopped! +[2023-02-22 19:13:27,647][23687] Weights refcount: 2 0 +[2023-02-22 19:13:27,653][01245] Component InferenceWorker_p0-w0 stopped! +[2023-02-22 19:13:27,661][23687] Stopping InferenceWorker_p0-w0... +[2023-02-22 19:13:27,661][23687] Loop inference_proc0-0_evt_loop terminating... +[2023-02-22 19:13:27,733][23673] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001763_7221248.pth +[2023-02-22 19:13:27,748][23673] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001955_8007680.pth... +[2023-02-22 19:13:27,820][23700] Stopping RolloutWorker_w3... +[2023-02-22 19:13:27,826][23688] Stopping RolloutWorker_w1... +[2023-02-22 19:13:27,821][01245] Component RolloutWorker_w3 stopped! +[2023-02-22 19:13:27,822][23700] Loop rollout_proc3_evt_loop terminating... +[2023-02-22 19:13:27,828][01245] Component RolloutWorker_w1 stopped! +[2023-02-22 19:13:27,827][23688] Loop rollout_proc1_evt_loop terminating... +[2023-02-22 19:13:27,853][23689] Stopping RolloutWorker_w0... +[2023-02-22 19:13:27,855][23689] Loop rollout_proc0_evt_loop terminating... +[2023-02-22 19:13:27,856][23691] Stopping RolloutWorker_w4... +[2023-02-22 19:13:27,857][23691] Loop rollout_proc4_evt_loop terminating... +[2023-02-22 19:13:27,860][23694] Stopping RolloutWorker_w2... +[2023-02-22 19:13:27,854][01245] Component RolloutWorker_w0 stopped! +[2023-02-22 19:13:27,861][23706] Stopping RolloutWorker_w6... +[2023-02-22 19:13:27,862][23706] Loop rollout_proc6_evt_loop terminating... +[2023-02-22 19:13:27,860][23694] Loop rollout_proc2_evt_loop terminating... +[2023-02-22 19:13:27,861][01245] Component RolloutWorker_w4 stopped! +[2023-02-22 19:13:27,864][01245] Component RolloutWorker_w2 stopped! +[2023-02-22 19:13:27,867][01245] Component RolloutWorker_w6 stopped! +[2023-02-22 19:13:27,901][01245] Component RolloutWorker_w5 stopped! +[2023-02-22 19:13:27,901][23701] Stopping RolloutWorker_w5... +[2023-02-22 19:13:27,912][01245] Component RolloutWorker_w7 stopped! +[2023-02-22 19:13:27,910][23703] Stopping RolloutWorker_w7... +[2023-02-22 19:13:27,908][23701] Loop rollout_proc5_evt_loop terminating... +[2023-02-22 19:13:27,920][23703] Loop rollout_proc7_evt_loop terminating... +[2023-02-22 19:13:28,003][01245] Component LearnerWorker_p0 stopped! +[2023-02-22 19:13:28,012][01245] Waiting for process learner_proc0 to stop... +[2023-02-22 19:13:28,018][23673] Stopping LearnerWorker_p0... +[2023-02-22 19:13:28,018][23673] Loop learner_proc0_evt_loop terminating... +[2023-02-22 19:13:30,842][01245] Waiting for process inference_proc0-0 to join... +[2023-02-22 19:13:30,922][01245] Waiting for process rollout_proc0 to join... +[2023-02-22 19:13:30,925][01245] Waiting for process rollout_proc1 to join... +[2023-02-22 19:13:30,928][01245] Waiting for process rollout_proc2 to join... +[2023-02-22 19:13:30,931][01245] Waiting for process rollout_proc3 to join... +[2023-02-22 19:13:30,932][01245] Waiting for process rollout_proc4 to join... +[2023-02-22 19:13:30,936][01245] Waiting for process rollout_proc5 to join... +[2023-02-22 19:13:30,937][01245] Waiting for process rollout_proc6 to join... +[2023-02-22 19:13:30,939][01245] Waiting for process rollout_proc7 to join... +[2023-02-22 19:13:30,940][01245] Batcher 0 profile tree view: +batching: 26.4957, releasing_batches: 0.0270 +[2023-02-22 19:13:30,942][01245] InferenceWorker_p0-w0 profile tree view: +wait_policy: 0.0000 + wait_policy_total: 568.1817 +update_model: 9.0231 + weight_update: 0.0023 +one_step: 0.0048 + handle_policy_step: 555.7939 + deserialize: 15.4881, stack: 3.2464, obs_to_device_normalize: 121.6912, forward: 270.8200, send_messages: 28.5871 + prepare_outputs: 88.1779 + to_cpu: 53.3961 +[2023-02-22 19:13:30,944][01245] Learner 0 profile tree view: +misc: 0.0061, prepare_batch: 20.5293 +train: 82.3298 + epoch_init: 0.0062, minibatch_init: 0.0073, losses_postprocess: 0.6499, kl_divergence: 0.5569, after_optimizer: 3.7222 + calculate_losses: 27.0915 + losses_init: 0.0037, forward_head: 1.8727, bptt_initial: 17.3837, tail: 1.2380, advantages_returns: 0.3463, losses: 3.4184 + bptt: 2.4672 + bptt_forward_core: 2.3984 + update: 49.5459 + clip: 1.4689 +[2023-02-22 19:13:30,945][01245] RolloutWorker_w0 profile tree view: +wait_for_trajectories: 0.4174, enqueue_policy_requests: 159.3009, env_step: 878.5273, overhead: 23.1882, complete_rollouts: 6.8349 +save_policy_outputs: 22.3895 + split_output_tensors: 10.4932 +[2023-02-22 19:13:30,947][01245] RolloutWorker_w7 profile tree view: +wait_for_trajectories: 0.4266, enqueue_policy_requests: 158.4046, env_step: 878.2349, overhead: 23.2540, complete_rollouts: 7.5983 +save_policy_outputs: 22.3752 + split_output_tensors: 11.0709 +[2023-02-22 19:13:30,949][01245] Loop Runner_EvtLoop terminating... +[2023-02-22 19:13:30,951][01245] Runner profile tree view: +main_loop: 1201.5573 +[2023-02-22 19:13:30,953][01245] Collected {0: 8007680}, FPS: 3330.5 +[2023-02-22 19:13:31,004][01245] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json +[2023-02-22 19:13:31,006][01245] Overriding arg 'num_workers' with value 1 passed from command line +[2023-02-22 19:13:31,007][01245] Adding new argument 'no_render'=True that is not in the saved config file! +[2023-02-22 19:13:31,013][01245] Adding new argument 'save_video'=True that is not in the saved config file! +[2023-02-22 19:13:31,014][01245] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! +[2023-02-22 19:13:31,017][01245] Adding new argument 'video_name'=None that is not in the saved config file! +[2023-02-22 19:13:31,020][01245] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file! +[2023-02-22 19:13:31,021][01245] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! +[2023-02-22 19:13:31,023][01245] Adding new argument 'push_to_hub'=False that is not in the saved config file! +[2023-02-22 19:13:31,026][01245] Adding new argument 'hf_repository'=None that is not in the saved config file! +[2023-02-22 19:13:31,031][01245] Adding new argument 'policy_index'=0 that is not in the saved config file! +[2023-02-22 19:13:31,033][01245] Adding new argument 'eval_deterministic'=False that is not in the saved config file! +[2023-02-22 19:13:31,036][01245] Adding new argument 'train_script'=None that is not in the saved config file! +[2023-02-22 19:13:31,038][01245] Adding new argument 'enjoy_script'=None that is not in the saved config file! +[2023-02-22 19:13:31,043][01245] Using frameskip 1 and render_action_repeat=4 for evaluation +[2023-02-22 19:13:31,063][01245] RunningMeanStd input shape: (3, 72, 128) +[2023-02-22 19:13:31,065][01245] RunningMeanStd input shape: (1,) +[2023-02-22 19:13:31,080][01245] ConvEncoder: input_channels=3 +[2023-02-22 19:13:31,142][01245] Conv encoder output size: 512 +[2023-02-22 19:13:31,145][01245] Policy head output size: 512 +[2023-02-22 19:13:31,185][01245] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001955_8007680.pth... +[2023-02-22 19:13:31,761][01245] Num frames 100... +[2023-02-22 19:13:31,887][01245] Num frames 200... +[2023-02-22 19:13:32,012][01245] Num frames 300... +[2023-02-22 19:13:32,135][01245] Num frames 400... +[2023-02-22 19:13:32,264][01245] Num frames 500... +[2023-02-22 19:13:32,394][01245] Num frames 600... +[2023-02-22 19:13:32,524][01245] Num frames 700... +[2023-02-22 19:13:32,637][01245] Avg episode rewards: #0: 17.400, true rewards: #0: 7.400 +[2023-02-22 19:13:32,640][01245] Avg episode reward: 17.400, avg true_objective: 7.400 +[2023-02-22 19:13:32,717][01245] Num frames 800... +[2023-02-22 19:13:32,853][01245] Num frames 900... +[2023-02-22 19:13:32,982][01245] Num frames 1000... +[2023-02-22 19:13:33,107][01245] Num frames 1100... +[2023-02-22 19:13:33,234][01245] Num frames 1200... +[2023-02-22 19:13:33,355][01245] Num frames 1300... +[2023-02-22 19:13:33,478][01245] Num frames 1400... +[2023-02-22 19:13:33,608][01245] Num frames 1500... +[2023-02-22 19:13:33,733][01245] Num frames 1600... +[2023-02-22 19:13:33,837][01245] Avg episode rewards: #0: 17.180, true rewards: #0: 8.180 +[2023-02-22 19:13:33,838][01245] Avg episode reward: 17.180, avg true_objective: 8.180 +[2023-02-22 19:13:33,918][01245] Num frames 1700... +[2023-02-22 19:13:34,041][01245] Num frames 1800... +[2023-02-22 19:13:34,161][01245] Num frames 1900... +[2023-02-22 19:13:34,294][01245] Num frames 2000... +[2023-02-22 19:13:34,416][01245] Num frames 2100... +[2023-02-22 19:13:34,548][01245] Num frames 2200... +[2023-02-22 19:13:34,673][01245] Num frames 2300... +[2023-02-22 19:13:34,798][01245] Num frames 2400... +[2023-02-22 19:13:34,924][01245] Num frames 2500... +[2023-02-22 19:13:35,062][01245] Num frames 2600... +[2023-02-22 19:13:35,245][01245] Num frames 2700... +[2023-02-22 19:13:35,426][01245] Num frames 2800... +[2023-02-22 19:13:35,602][01245] Num frames 2900... +[2023-02-22 19:13:35,775][01245] Num frames 3000... +[2023-02-22 19:13:35,957][01245] Num frames 3100... +[2023-02-22 19:13:36,095][01245] Avg episode rewards: #0: 24.147, true rewards: #0: 10.480 +[2023-02-22 19:13:36,097][01245] Avg episode reward: 24.147, avg true_objective: 10.480 +[2023-02-22 19:13:36,200][01245] Num frames 3200... +[2023-02-22 19:13:36,387][01245] Num frames 3300... +[2023-02-22 19:13:36,561][01245] Num frames 3400... +[2023-02-22 19:13:36,736][01245] Num frames 3500... +[2023-02-22 19:13:36,908][01245] Num frames 3600... +[2023-02-22 19:13:37,091][01245] Num frames 3700... +[2023-02-22 19:13:37,261][01245] Num frames 3800... +[2023-02-22 19:13:37,445][01245] Num frames 3900... +[2023-02-22 19:13:37,620][01245] Num frames 4000... +[2023-02-22 19:13:37,800][01245] Num frames 4100... +[2023-02-22 19:13:37,981][01245] Num frames 4200... +[2023-02-22 19:13:38,156][01245] Num frames 4300... +[2023-02-22 19:13:38,332][01245] Num frames 4400... +[2023-02-22 19:13:38,508][01245] Num frames 4500... +[2023-02-22 19:13:38,697][01245] Num frames 4600... +[2023-02-22 19:13:38,895][01245] Avg episode rewards: #0: 27.950, true rewards: #0: 11.700 +[2023-02-22 19:13:38,897][01245] Avg episode reward: 27.950, avg true_objective: 11.700 +[2023-02-22 19:13:38,936][01245] Num frames 4700... +[2023-02-22 19:13:39,063][01245] Num frames 4800... +[2023-02-22 19:13:39,183][01245] Num frames 4900... +[2023-02-22 19:13:39,303][01245] Num frames 5000... +[2023-02-22 19:13:39,440][01245] Num frames 5100... +[2023-02-22 19:13:39,563][01245] Num frames 5200... +[2023-02-22 19:13:39,690][01245] Num frames 5300... +[2023-02-22 19:13:39,809][01245] Num frames 5400... +[2023-02-22 19:13:39,931][01245] Num frames 5500... +[2023-02-22 19:13:40,056][01245] Num frames 5600... +[2023-02-22 19:13:40,206][01245] Avg episode rewards: #0: 26.144, true rewards: #0: 11.344 +[2023-02-22 19:13:40,207][01245] Avg episode reward: 26.144, avg true_objective: 11.344 +[2023-02-22 19:13:40,248][01245] Num frames 5700... +[2023-02-22 19:13:40,388][01245] Num frames 5800... +[2023-02-22 19:13:40,519][01245] Num frames 5900... +[2023-02-22 19:13:40,653][01245] Num frames 6000... +[2023-02-22 19:13:40,777][01245] Num frames 6100... +[2023-02-22 19:13:40,900][01245] Num frames 6200... +[2023-02-22 19:13:40,977][01245] Avg episode rewards: #0: 23.527, true rewards: #0: 10.360 +[2023-02-22 19:13:40,978][01245] Avg episode reward: 23.527, avg true_objective: 10.360 +[2023-02-22 19:13:41,090][01245] Num frames 6300... +[2023-02-22 19:13:41,208][01245] Num frames 6400... +[2023-02-22 19:13:41,326][01245] Num frames 6500... +[2023-02-22 19:13:41,462][01245] Avg episode rewards: #0: 20.811, true rewards: #0: 9.383 +[2023-02-22 19:13:41,464][01245] Avg episode reward: 20.811, avg true_objective: 9.383 +[2023-02-22 19:13:41,516][01245] Num frames 6600... +[2023-02-22 19:13:41,641][01245] Num frames 6700... +[2023-02-22 19:13:41,764][01245] Num frames 6800... +[2023-02-22 19:13:41,884][01245] Num frames 6900... +[2023-02-22 19:13:42,012][01245] Num frames 7000... +[2023-02-22 19:13:42,132][01245] Num frames 7100... +[2023-02-22 19:13:42,251][01245] Num frames 7200... +[2023-02-22 19:13:42,392][01245] Avg episode rewards: #0: 19.965, true rewards: #0: 9.090 +[2023-02-22 19:13:42,395][01245] Avg episode reward: 19.965, avg true_objective: 9.090 +[2023-02-22 19:13:42,437][01245] Num frames 7300... +[2023-02-22 19:13:42,564][01245] Num frames 7400... +[2023-02-22 19:13:42,686][01245] Num frames 7500... +[2023-02-22 19:13:42,811][01245] Num frames 7600... +[2023-02-22 19:13:42,980][01245] Avg episode rewards: #0: 18.320, true rewards: #0: 8.542 +[2023-02-22 19:13:42,983][01245] Avg episode reward: 18.320, avg true_objective: 8.542 +[2023-02-22 19:13:43,001][01245] Num frames 7700... +[2023-02-22 19:13:43,121][01245] Num frames 7800... +[2023-02-22 19:13:43,246][01245] Num frames 7900... +[2023-02-22 19:13:43,377][01245] Num frames 8000... +[2023-02-22 19:13:43,506][01245] Num frames 8100... +[2023-02-22 19:13:43,631][01245] Num frames 8200... +[2023-02-22 19:13:43,754][01245] Num frames 8300... +[2023-02-22 19:13:43,882][01245] Num frames 8400... +[2023-02-22 19:13:44,007][01245] Num frames 8500... +[2023-02-22 19:13:44,184][01245] Avg episode rewards: #0: 18.197, true rewards: #0: 8.597 +[2023-02-22 19:13:44,185][01245] Avg episode reward: 18.197, avg true_objective: 8.597 +[2023-02-22 19:13:44,193][01245] Num frames 8600... +[2023-02-22 19:14:42,843][01245] Replay video saved to /content/train_dir/default_experiment/replay.mp4! +[2023-02-22 19:14:43,199][01245] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json +[2023-02-22 19:14:43,202][01245] Overriding arg 'num_workers' with value 1 passed from command line +[2023-02-22 19:14:43,204][01245] Adding new argument 'no_render'=True that is not in the saved config file! +[2023-02-22 19:14:43,206][01245] Adding new argument 'save_video'=True that is not in the saved config file! +[2023-02-22 19:14:43,209][01245] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! +[2023-02-22 19:14:43,210][01245] Adding new argument 'video_name'=None that is not in the saved config file! +[2023-02-22 19:14:43,216][01245] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! +[2023-02-22 19:14:43,218][01245] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! +[2023-02-22 19:14:43,219][01245] Adding new argument 'push_to_hub'=True that is not in the saved config file! +[2023-02-22 19:14:43,220][01245] Adding new argument 'hf_repository'='NoNameFound/rl_course_vizdoom_health_gathering_supreme8mil' that is not in the saved config file! +[2023-02-22 19:14:43,221][01245] Adding new argument 'policy_index'=0 that is not in the saved config file! +[2023-02-22 19:14:43,222][01245] Adding new argument 'eval_deterministic'=False that is not in the saved config file! +[2023-02-22 19:14:43,224][01245] Adding new argument 'train_script'=None that is not in the saved config file! +[2023-02-22 19:14:43,225][01245] Adding new argument 'enjoy_script'=None that is not in the saved config file! +[2023-02-22 19:14:43,226][01245] Using frameskip 1 and render_action_repeat=4 for evaluation +[2023-02-22 19:14:43,254][01245] RunningMeanStd input shape: (3, 72, 128) +[2023-02-22 19:14:43,256][01245] RunningMeanStd input shape: (1,) +[2023-02-22 19:14:43,275][01245] ConvEncoder: input_channels=3 +[2023-02-22 19:14:43,338][01245] Conv encoder output size: 512 +[2023-02-22 19:14:43,341][01245] Policy head output size: 512 +[2023-02-22 19:14:43,371][01245] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001955_8007680.pth... +[2023-02-22 19:14:44,162][01245] Num frames 100... +[2023-02-22 19:14:44,340][01245] Num frames 200... +[2023-02-22 19:14:44,517][01245] Num frames 300... +[2023-02-22 19:14:44,700][01245] Num frames 400... +[2023-02-22 19:14:44,886][01245] Num frames 500... +[2023-02-22 19:14:45,062][01245] Num frames 600... +[2023-02-22 19:14:45,239][01245] Num frames 700... +[2023-02-22 19:14:45,411][01245] Num frames 800... +[2023-02-22 19:14:45,578][01245] Num frames 900... +[2023-02-22 19:14:45,754][01245] Num frames 1000... +[2023-02-22 19:14:45,931][01245] Num frames 1100... +[2023-02-22 19:14:46,109][01245] Num frames 1200... +[2023-02-22 19:14:46,298][01245] Num frames 1300... +[2023-02-22 19:14:46,489][01245] Num frames 1400... +[2023-02-22 19:14:46,718][01245] Num frames 1500... +[2023-02-22 19:14:46,925][01245] Num frames 1600... +[2023-02-22 19:14:46,979][01245] Avg episode rewards: #0: 35.000, true rewards: #0: 16.000 +[2023-02-22 19:14:46,982][01245] Avg episode reward: 35.000, avg true_objective: 16.000 +[2023-02-22 19:14:47,163][01245] Num frames 1700... +[2023-02-22 19:14:47,350][01245] Num frames 1800... +[2023-02-22 19:14:47,550][01245] Num frames 1900... +[2023-02-22 19:14:47,767][01245] Num frames 2000... +[2023-02-22 19:14:47,988][01245] Num frames 2100... +[2023-02-22 19:14:48,204][01245] Num frames 2200... +[2023-02-22 19:14:48,417][01245] Num frames 2300... +[2023-02-22 19:14:48,626][01245] Num frames 2400... +[2023-02-22 19:14:48,680][01245] Avg episode rewards: #0: 27.000, true rewards: #0: 12.000 +[2023-02-22 19:14:48,683][01245] Avg episode reward: 27.000, avg true_objective: 12.000 +[2023-02-22 19:14:48,890][01245] Num frames 2500... +[2023-02-22 19:14:49,108][01245] Num frames 2600... +[2023-02-22 19:14:49,285][01245] Num frames 2700... +[2023-02-22 19:14:49,460][01245] Num frames 2800... +[2023-02-22 19:14:49,638][01245] Num frames 2900... +[2023-02-22 19:14:49,811][01245] Num frames 3000... +[2023-02-22 19:14:49,993][01245] Num frames 3100... +[2023-02-22 19:14:50,167][01245] Num frames 3200... +[2023-02-22 19:14:50,346][01245] Num frames 3300... +[2023-02-22 19:14:50,527][01245] Num frames 3400... +[2023-02-22 19:14:50,631][01245] Avg episode rewards: #0: 25.747, true rewards: #0: 11.413 +[2023-02-22 19:14:50,633][01245] Avg episode reward: 25.747, avg true_objective: 11.413 +[2023-02-22 19:14:50,779][01245] Num frames 3500... +[2023-02-22 19:14:50,959][01245] Num frames 3600... +[2023-02-22 19:14:51,142][01245] Num frames 3700... +[2023-02-22 19:14:51,323][01245] Num frames 3800... +[2023-02-22 19:14:51,503][01245] Num frames 3900... +[2023-02-22 19:14:51,693][01245] Num frames 4000... +[2023-02-22 19:14:51,834][01245] Num frames 4100... +[2023-02-22 19:14:51,955][01245] Num frames 4200... +[2023-02-22 19:14:52,082][01245] Num frames 4300... +[2023-02-22 19:14:52,206][01245] Num frames 4400... +[2023-02-22 19:14:52,283][01245] Avg episode rewards: #0: 24.290, true rewards: #0: 11.040 +[2023-02-22 19:14:52,284][01245] Avg episode reward: 24.290, avg true_objective: 11.040 +[2023-02-22 19:14:52,390][01245] Num frames 4500... +[2023-02-22 19:14:52,513][01245] Num frames 4600... +[2023-02-22 19:14:52,647][01245] Num frames 4700... +[2023-02-22 19:14:52,771][01245] Num frames 4800... +[2023-02-22 19:14:52,897][01245] Num frames 4900... +[2023-02-22 19:14:53,026][01245] Num frames 5000... +[2023-02-22 19:14:53,148][01245] Num frames 5100... +[2023-02-22 19:14:53,270][01245] Num frames 5200... +[2023-02-22 19:14:53,397][01245] Num frames 5300... +[2023-02-22 19:14:53,521][01245] Num frames 5400... +[2023-02-22 19:14:53,651][01245] Num frames 5500... +[2023-02-22 19:14:53,775][01245] Num frames 5600... +[2023-02-22 19:14:53,901][01245] Num frames 5700... +[2023-02-22 19:14:54,082][01245] Avg episode rewards: #0: 26.196, true rewards: #0: 11.596 +[2023-02-22 19:14:54,083][01245] Avg episode reward: 26.196, avg true_objective: 11.596 +[2023-02-22 19:14:54,091][01245] Num frames 5800... +[2023-02-22 19:14:54,214][01245] Num frames 5900... +[2023-02-22 19:14:54,339][01245] Num frames 6000... +[2023-02-22 19:14:54,481][01245] Num frames 6100... +[2023-02-22 19:14:54,604][01245] Num frames 6200... +[2023-02-22 19:14:54,740][01245] Num frames 6300... +[2023-02-22 19:14:54,867][01245] Num frames 6400... +[2023-02-22 19:14:55,003][01245] Num frames 6500... +[2023-02-22 19:14:55,131][01245] Num frames 6600... +[2023-02-22 19:14:55,256][01245] Num frames 6700... +[2023-02-22 19:14:55,382][01245] Num frames 6800... +[2023-02-22 19:14:55,507][01245] Num frames 6900... +[2023-02-22 19:14:55,639][01245] Num frames 7000... +[2023-02-22 19:14:55,771][01245] Num frames 7100... +[2023-02-22 19:14:55,894][01245] Num frames 7200... +[2023-02-22 19:14:56,025][01245] Num frames 7300... +[2023-02-22 19:14:56,149][01245] Num frames 7400... +[2023-02-22 19:14:56,271][01245] Num frames 7500... +[2023-02-22 19:14:56,396][01245] Num frames 7600... +[2023-02-22 19:14:56,520][01245] Num frames 7700... +[2023-02-22 19:14:56,647][01245] Num frames 7800... +[2023-02-22 19:14:56,833][01245] Avg episode rewards: #0: 30.497, true rewards: #0: 13.163 +[2023-02-22 19:14:56,835][01245] Avg episode reward: 30.497, avg true_objective: 13.163 +[2023-02-22 19:14:56,844][01245] Num frames 7900... +[2023-02-22 19:14:56,973][01245] Num frames 8000... +[2023-02-22 19:14:57,098][01245] Num frames 8100... +[2023-02-22 19:14:57,220][01245] Num frames 8200... +[2023-02-22 19:14:57,342][01245] Num frames 8300... +[2023-02-22 19:14:57,455][01245] Avg episode rewards: #0: 26.923, true rewards: #0: 11.923 +[2023-02-22 19:14:57,456][01245] Avg episode reward: 26.923, avg true_objective: 11.923 +[2023-02-22 19:14:57,531][01245] Num frames 8400... +[2023-02-22 19:14:57,653][01245] Num frames 8500... +[2023-02-22 19:14:57,787][01245] Num frames 8600... +[2023-02-22 19:14:57,912][01245] Num frames 8700... +[2023-02-22 19:14:58,042][01245] Num frames 8800... +[2023-02-22 19:14:58,165][01245] Num frames 8900... +[2023-02-22 19:14:58,291][01245] Num frames 9000... +[2023-02-22 19:14:58,416][01245] Num frames 9100... +[2023-02-22 19:14:58,542][01245] Num frames 9200... +[2023-02-22 19:14:58,670][01245] Num frames 9300... +[2023-02-22 19:14:58,804][01245] Num frames 9400... +[2023-02-22 19:14:58,928][01245] Num frames 9500... +[2023-02-22 19:14:59,057][01245] Num frames 9600... +[2023-02-22 19:14:59,182][01245] Num frames 9700... +[2023-02-22 19:14:59,305][01245] Num frames 9800... +[2023-02-22 19:14:59,424][01245] Avg episode rewards: #0: 27.437, true rewards: #0: 12.312 +[2023-02-22 19:14:59,426][01245] Avg episode reward: 27.437, avg true_objective: 12.312 +[2023-02-22 19:14:59,494][01245] Num frames 9900... +[2023-02-22 19:14:59,620][01245] Num frames 10000... +[2023-02-22 19:14:59,751][01245] Num frames 10100... +[2023-02-22 19:14:59,880][01245] Num frames 10200... +[2023-02-22 19:15:00,010][01245] Num frames 10300... +[2023-02-22 19:15:00,137][01245] Num frames 10400... +[2023-02-22 19:15:00,263][01245] Num frames 10500... +[2023-02-22 19:15:00,386][01245] Num frames 10600... +[2023-02-22 19:15:00,510][01245] Num frames 10700... +[2023-02-22 19:15:00,637][01245] Num frames 10800... +[2023-02-22 19:15:00,773][01245] Num frames 10900... +[2023-02-22 19:15:00,921][01245] Avg episode rewards: #0: 27.522, true rewards: #0: 12.189 +[2023-02-22 19:15:00,923][01245] Avg episode reward: 27.522, avg true_objective: 12.189 +[2023-02-22 19:15:00,967][01245] Num frames 11000... +[2023-02-22 19:15:01,088][01245] Num frames 11100... +[2023-02-22 19:15:01,209][01245] Num frames 11200... +[2023-02-22 19:15:01,333][01245] Num frames 11300... +[2023-02-22 19:15:01,454][01245] Num frames 11400... +[2023-02-22 19:15:01,577][01245] Num frames 11500... +[2023-02-22 19:15:01,653][01245] Avg episode rewards: #0: 25.714, true rewards: #0: 11.514 +[2023-02-22 19:15:01,655][01245] Avg episode reward: 25.714, avg true_objective: 11.514 +[2023-02-22 19:16:21,236][01245] Replay video saved to /content/train_dir/default_experiment/replay.mp4!