[2025-08-21 19:28:00,745][02158] Saving configuration to /content/train_dir/default_experiment/config.json... [2025-08-21 19:28:00,747][02158] Rollout worker 0 uses device cpu [2025-08-21 19:28:00,748][02158] Rollout worker 1 uses device cpu [2025-08-21 19:28:00,750][02158] Rollout worker 2 uses device cpu [2025-08-21 19:28:00,750][02158] Rollout worker 3 uses device cpu [2025-08-21 19:28:00,752][02158] Rollout worker 4 uses device cpu [2025-08-21 19:28:00,752][02158] Rollout worker 5 uses device cpu [2025-08-21 19:28:00,753][02158] Rollout worker 6 uses device cpu [2025-08-21 19:28:00,754][02158] Rollout worker 7 uses device cpu [2025-08-21 19:28:00,898][02158] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-08-21 19:28:00,899][02158] InferenceWorker_p0-w0: min num requests: 2 [2025-08-21 19:28:00,936][02158] Starting all processes... [2025-08-21 19:28:00,937][02158] Starting process learner_proc0 [2025-08-21 19:28:00,999][02158] Starting all processes... [2025-08-21 19:28:01,005][02158] Starting process inference_proc0-0 [2025-08-21 19:28:01,007][02158] Starting process rollout_proc0 [2025-08-21 19:28:01,007][02158] Starting process rollout_proc1 [2025-08-21 19:28:01,007][02158] Starting process rollout_proc2 [2025-08-21 19:28:01,007][02158] Starting process rollout_proc3 [2025-08-21 19:28:01,007][02158] Starting process rollout_proc4 [2025-08-21 19:28:01,007][02158] Starting process rollout_proc5 [2025-08-21 19:28:01,007][02158] Starting process rollout_proc6 [2025-08-21 19:28:01,007][02158] Starting process rollout_proc7 [2025-08-21 19:28:18,263][02316] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-08-21 19:28:18,264][02316] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0 [2025-08-21 19:28:18,395][02316] Num visible devices: 1 [2025-08-21 19:28:18,410][02316] Starting seed is not provided [2025-08-21 19:28:18,410][02316] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-08-21 19:28:18,411][02316] Initializing actor-critic model on device cuda:0 [2025-08-21 19:28:18,412][02316] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 19:28:18,414][02316] RunningMeanStd input shape: (1,) [2025-08-21 19:28:18,568][02316] ConvEncoder: input_channels=3 [2025-08-21 19:28:18,729][02329] Worker 0 uses CPU cores [0] [2025-08-21 19:28:19,178][02336] Worker 6 uses CPU cores [0] [2025-08-21 19:28:19,206][02337] Worker 7 uses CPU cores [1] [2025-08-21 19:28:19,479][02331] Worker 2 uses CPU cores [0] [2025-08-21 19:28:19,568][02334] Worker 5 uses CPU cores [1] [2025-08-21 19:28:19,718][02330] Worker 1 uses CPU cores [1] [2025-08-21 19:28:19,730][02332] Worker 3 uses CPU cores [1] [2025-08-21 19:28:19,864][02316] Conv encoder output size: 512 [2025-08-21 19:28:19,866][02316] Policy head output size: 512 [2025-08-21 19:28:19,934][02335] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-08-21 19:28:19,935][02335] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0 [2025-08-21 19:28:19,993][02335] Num visible devices: 1 [2025-08-21 19:28:19,994][02316] Created Actor Critic model with architecture: [2025-08-21 19:28:19,994][02316] 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) ) ) [2025-08-21 19:28:20,148][02333] Worker 4 uses CPU cores [0] [2025-08-21 19:28:20,558][02316] Using optimizer [2025-08-21 19:28:20,890][02158] Heartbeat connected on Batcher_0 [2025-08-21 19:28:20,900][02158] Heartbeat connected on InferenceWorker_p0-w0 [2025-08-21 19:28:20,910][02158] Heartbeat connected on RolloutWorker_w0 [2025-08-21 19:28:20,918][02158] Heartbeat connected on RolloutWorker_w2 [2025-08-21 19:28:20,921][02158] Heartbeat connected on RolloutWorker_w1 [2025-08-21 19:28:20,921][02158] Heartbeat connected on RolloutWorker_w3 [2025-08-21 19:28:20,924][02158] Heartbeat connected on RolloutWorker_w4 [2025-08-21 19:28:20,929][02158] Heartbeat connected on RolloutWorker_w5 [2025-08-21 19:28:20,932][02158] Heartbeat connected on RolloutWorker_w6 [2025-08-21 19:28:20,936][02158] Heartbeat connected on RolloutWorker_w7 [2025-08-21 19:28:26,543][02316] No checkpoints found [2025-08-21 19:28:26,543][02316] Did not load from checkpoint, starting from scratch! [2025-08-21 19:28:26,543][02316] Initialized policy 0 weights for model version 0 [2025-08-21 19:28:26,547][02316] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-08-21 19:28:26,553][02316] LearnerWorker_p0 finished initialization! [2025-08-21 19:28:26,556][02158] Heartbeat connected on LearnerWorker_p0 [2025-08-21 19:28:26,658][02335] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 19:28:26,659][02335] RunningMeanStd input shape: (1,) [2025-08-21 19:28:26,669][02335] ConvEncoder: input_channels=3 [2025-08-21 19:28:26,759][02335] Conv encoder output size: 512 [2025-08-21 19:28:26,759][02335] Policy head output size: 512 [2025-08-21 19:28:26,792][02158] Inference worker 0-0 is ready! [2025-08-21 19:28:26,793][02158] All inference workers are ready! Signal rollout workers to start! [2025-08-21 19:28:26,981][02331] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:28:26,980][02330] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:28:26,979][02337] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:28:26,982][02333] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:28:26,978][02329] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:28:26,985][02336] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:28:26,989][02332] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:28:26,987][02334] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:28:27,951][02337] Decorrelating experience for 0 frames... [2025-08-21 19:28:27,955][02330] Decorrelating experience for 0 frames... [2025-08-21 19:28:28,256][02329] Decorrelating experience for 0 frames... [2025-08-21 19:28:28,259][02336] Decorrelating experience for 0 frames... [2025-08-21 19:28:28,266][02333] Decorrelating experience for 0 frames... [2025-08-21 19:28:29,008][02329] Decorrelating experience for 32 frames... [2025-08-21 19:28:29,019][02333] Decorrelating experience for 32 frames... [2025-08-21 19:28:29,113][02330] Decorrelating experience for 32 frames... [2025-08-21 19:28:29,118][02332] Decorrelating experience for 0 frames... [2025-08-21 19:28:29,787][02337] Decorrelating experience for 32 frames... [2025-08-21 19:28:29,784][02334] Decorrelating experience for 0 frames... [2025-08-21 19:28:30,015][02336] Decorrelating experience for 32 frames... [2025-08-21 19:28:30,080][02158] 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) [2025-08-21 19:28:30,199][02329] Decorrelating experience for 64 frames... [2025-08-21 19:28:30,652][02332] Decorrelating experience for 32 frames... [2025-08-21 19:28:31,034][02330] Decorrelating experience for 64 frames... [2025-08-21 19:28:31,374][02334] Decorrelating experience for 32 frames... [2025-08-21 19:28:31,608][02331] Decorrelating experience for 0 frames... [2025-08-21 19:28:31,954][02333] Decorrelating experience for 64 frames... [2025-08-21 19:28:32,515][02336] Decorrelating experience for 64 frames... [2025-08-21 19:28:32,554][02337] Decorrelating experience for 64 frames... [2025-08-21 19:28:33,214][02332] Decorrelating experience for 64 frames... [2025-08-21 19:28:33,573][02330] Decorrelating experience for 96 frames... [2025-08-21 19:28:33,589][02331] Decorrelating experience for 32 frames... [2025-08-21 19:28:33,592][02329] Decorrelating experience for 96 frames... [2025-08-21 19:28:34,280][02333] Decorrelating experience for 96 frames... [2025-08-21 19:28:34,320][02337] Decorrelating experience for 96 frames... [2025-08-21 19:28:35,073][02336] Decorrelating experience for 96 frames... [2025-08-21 19:28:35,080][02158] 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) [2025-08-21 19:28:35,086][02332] Decorrelating experience for 96 frames... [2025-08-21 19:28:35,476][02334] Decorrelating experience for 64 frames... [2025-08-21 19:28:36,039][02331] Decorrelating experience for 64 frames... [2025-08-21 19:28:37,498][02334] Decorrelating experience for 96 frames... [2025-08-21 19:28:38,649][02331] Decorrelating experience for 96 frames... [2025-08-21 19:28:39,003][02316] Signal inference workers to stop experience collection... [2025-08-21 19:28:39,025][02335] InferenceWorker_p0-w0: stopping experience collection [2025-08-21 19:28:40,080][02158] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 241.8. Samples: 2418. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) [2025-08-21 19:28:40,081][02158] Avg episode reward: [(0, '2.952')] [2025-08-21 19:28:41,235][02316] Signal inference workers to resume experience collection... [2025-08-21 19:28:41,237][02335] InferenceWorker_p0-w0: resuming experience collection [2025-08-21 19:28:45,080][02158] Fps is (10 sec: 2457.6, 60 sec: 1638.4, 300 sec: 1638.4). Total num frames: 24576. Throughput: 0: 439.7. Samples: 6596. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:28:45,086][02158] Avg episode reward: [(0, '3.619')] [2025-08-21 19:28:50,080][02158] Fps is (10 sec: 3686.4, 60 sec: 1843.2, 300 sec: 1843.2). Total num frames: 36864. Throughput: 0: 437.6. Samples: 8752. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:28:50,085][02158] Avg episode reward: [(0, '3.902')] [2025-08-21 19:28:50,225][02335] Updated weights for policy 0, policy_version 10 (0.0089) [2025-08-21 19:28:55,080][02158] Fps is (10 sec: 3686.5, 60 sec: 2457.6, 300 sec: 2457.6). Total num frames: 61440. Throughput: 0: 603.4. Samples: 15086. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:28:55,081][02158] Avg episode reward: [(0, '4.337')] [2025-08-21 19:28:58,868][02335] Updated weights for policy 0, policy_version 20 (0.0014) [2025-08-21 19:29:00,083][02158] Fps is (10 sec: 4913.5, 60 sec: 2866.9, 300 sec: 2866.9). Total num frames: 86016. Throughput: 0: 735.0. Samples: 22052. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:29:00,085][02158] Avg episode reward: [(0, '4.350')] [2025-08-21 19:29:05,080][02158] Fps is (10 sec: 4096.0, 60 sec: 2925.7, 300 sec: 2925.7). Total num frames: 102400. Throughput: 0: 693.0. Samples: 24254. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:29:05,083][02158] Avg episode reward: [(0, '4.342')] [2025-08-21 19:29:05,090][02316] Saving new best policy, reward=4.342! [2025-08-21 19:29:09,462][02335] Updated weights for policy 0, policy_version 30 (0.0021) [2025-08-21 19:29:10,080][02158] Fps is (10 sec: 3687.7, 60 sec: 3072.0, 300 sec: 3072.0). Total num frames: 122880. Throughput: 0: 765.1. Samples: 30602. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:29:10,084][02158] Avg episode reward: [(0, '4.404')] [2025-08-21 19:29:10,087][02316] Saving new best policy, reward=4.404! [2025-08-21 19:29:15,080][02158] Fps is (10 sec: 4096.0, 60 sec: 3185.8, 300 sec: 3185.8). Total num frames: 143360. Throughput: 0: 814.5. Samples: 36654. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:29:15,083][02158] Avg episode reward: [(0, '4.343')] [2025-08-21 19:29:20,080][02158] Fps is (10 sec: 3686.3, 60 sec: 3194.9, 300 sec: 3194.9). Total num frames: 159744. Throughput: 0: 860.6. Samples: 38728. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:29:20,084][02158] Avg episode reward: [(0, '4.422')] [2025-08-21 19:29:20,087][02316] Saving new best policy, reward=4.422! [2025-08-21 19:29:20,845][02335] Updated weights for policy 0, policy_version 40 (0.0012) [2025-08-21 19:29:25,080][02158] Fps is (10 sec: 3686.4, 60 sec: 3276.8, 300 sec: 3276.8). Total num frames: 180224. Throughput: 0: 945.8. Samples: 44978. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:29:25,081][02158] Avg episode reward: [(0, '4.362')] [2025-08-21 19:29:30,082][02158] Fps is (10 sec: 4095.2, 60 sec: 3344.9, 300 sec: 3344.9). Total num frames: 200704. Throughput: 0: 999.7. Samples: 51584. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:29:30,083][02158] Avg episode reward: [(0, '4.410')] [2025-08-21 19:29:30,213][02335] Updated weights for policy 0, policy_version 50 (0.0023) [2025-08-21 19:29:35,080][02158] Fps is (10 sec: 3686.4, 60 sec: 3618.1, 300 sec: 3339.8). Total num frames: 217088. Throughput: 0: 997.2. Samples: 53624. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:29:35,081][02158] Avg episode reward: [(0, '4.460')] [2025-08-21 19:29:35,087][02316] Saving new best policy, reward=4.460! [2025-08-21 19:29:40,080][02158] Fps is (10 sec: 4096.8, 60 sec: 4027.7, 300 sec: 3452.3). Total num frames: 241664. Throughput: 0: 992.6. Samples: 59754. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:29:40,081][02158] Avg episode reward: [(0, '4.504')] [2025-08-21 19:29:40,089][02316] Saving new best policy, reward=4.504! [2025-08-21 19:29:40,810][02335] Updated weights for policy 0, policy_version 60 (0.0015) [2025-08-21 19:29:45,081][02158] Fps is (10 sec: 4095.5, 60 sec: 3891.1, 300 sec: 3440.6). Total num frames: 258048. Throughput: 0: 966.4. Samples: 65540. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:29:45,082][02158] Avg episode reward: [(0, '4.595')] [2025-08-21 19:29:45,087][02316] Saving new best policy, reward=4.595! [2025-08-21 19:29:50,080][02158] Fps is (10 sec: 3276.9, 60 sec: 3959.5, 300 sec: 3430.4). Total num frames: 274432. Throughput: 0: 958.8. Samples: 67400. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:29:50,082][02158] Avg episode reward: [(0, '4.531')] [2025-08-21 19:29:52,178][02335] Updated weights for policy 0, policy_version 70 (0.0011) [2025-08-21 19:29:55,080][02158] Fps is (10 sec: 4096.5, 60 sec: 3959.5, 300 sec: 3517.7). Total num frames: 299008. Throughput: 0: 969.3. Samples: 74220. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:29:55,084][02158] Avg episode reward: [(0, '4.446')] [2025-08-21 19:29:55,091][02316] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000073_299008.pth... [2025-08-21 19:30:00,080][02158] Fps is (10 sec: 4505.6, 60 sec: 3891.4, 300 sec: 3549.9). Total num frames: 319488. Throughput: 0: 969.0. Samples: 80258. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:30:00,090][02158] Avg episode reward: [(0, '4.331')] [2025-08-21 19:30:02,747][02335] Updated weights for policy 0, policy_version 80 (0.0017) [2025-08-21 19:30:05,080][02158] Fps is (10 sec: 3686.4, 60 sec: 3891.2, 300 sec: 3535.5). Total num frames: 335872. Throughput: 0: 968.6. Samples: 82316. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) [2025-08-21 19:30:05,084][02158] Avg episode reward: [(0, '4.274')] [2025-08-21 19:30:10,080][02158] Fps is (10 sec: 3686.4, 60 sec: 3891.2, 300 sec: 3563.5). Total num frames: 356352. Throughput: 0: 980.9. Samples: 89118. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:30:10,083][02158] Avg episode reward: [(0, '4.268')] [2025-08-21 19:30:12,006][02335] Updated weights for policy 0, policy_version 90 (0.0019) [2025-08-21 19:30:15,080][02158] Fps is (10 sec: 4096.0, 60 sec: 3891.2, 300 sec: 3588.9). Total num frames: 376832. Throughput: 0: 967.5. Samples: 95120. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:30:15,084][02158] Avg episode reward: [(0, '4.318')] [2025-08-21 19:30:20,080][02158] Fps is (10 sec: 3686.4, 60 sec: 3891.2, 300 sec: 3574.7). Total num frames: 393216. Throughput: 0: 966.6. Samples: 97120. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:30:20,081][02158] Avg episode reward: [(0, '4.311')] [2025-08-21 19:30:22,908][02335] Updated weights for policy 0, policy_version 100 (0.0032) [2025-08-21 19:30:25,080][02158] Fps is (10 sec: 4096.0, 60 sec: 3959.5, 300 sec: 3633.0). Total num frames: 417792. Throughput: 0: 982.6. Samples: 103972. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:30:25,083][02158] Avg episode reward: [(0, '4.630')] [2025-08-21 19:30:25,089][02316] Saving new best policy, reward=4.630! [2025-08-21 19:30:30,084][02158] Fps is (10 sec: 4094.1, 60 sec: 3891.0, 300 sec: 3618.0). Total num frames: 434176. Throughput: 0: 984.8. Samples: 109858. Policy #0 lag: (min: 0.0, avg: 0.3, max: 2.0) [2025-08-21 19:30:30,086][02158] Avg episode reward: [(0, '4.643')] [2025-08-21 19:30:30,088][02316] Saving new best policy, reward=4.643! [2025-08-21 19:30:33,866][02335] Updated weights for policy 0, policy_version 110 (0.0012) [2025-08-21 19:30:35,080][02158] Fps is (10 sec: 3686.4, 60 sec: 3959.5, 300 sec: 3637.2). Total num frames: 454656. Throughput: 0: 991.9. Samples: 112036. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:30:35,082][02158] Avg episode reward: [(0, '4.356')] [2025-08-21 19:30:40,080][02158] Fps is (10 sec: 4507.7, 60 sec: 3959.5, 300 sec: 3686.4). Total num frames: 479232. Throughput: 0: 999.8. Samples: 119210. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:30:40,083][02158] Avg episode reward: [(0, '4.361')] [2025-08-21 19:30:42,489][02335] Updated weights for policy 0, policy_version 120 (0.0021) [2025-08-21 19:30:45,080][02158] Fps is (10 sec: 4096.0, 60 sec: 3959.5, 300 sec: 3671.2). Total num frames: 495616. Throughput: 0: 995.1. Samples: 125038. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:30:45,081][02158] Avg episode reward: [(0, '4.522')] [2025-08-21 19:30:50,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.7, 300 sec: 3686.4). Total num frames: 516096. Throughput: 0: 1000.0. Samples: 127318. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:30:50,081][02158] Avg episode reward: [(0, '4.632')] [2025-08-21 19:30:53,377][02335] Updated weights for policy 0, policy_version 130 (0.0022) [2025-08-21 19:30:55,080][02158] Fps is (10 sec: 4096.0, 60 sec: 3959.5, 300 sec: 3700.5). Total num frames: 536576. Throughput: 0: 1002.2. Samples: 134218. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:30:55,084][02158] Avg episode reward: [(0, '4.758')] [2025-08-21 19:30:55,090][02316] Saving new best policy, reward=4.758! [2025-08-21 19:31:00,081][02158] Fps is (10 sec: 4095.6, 60 sec: 3959.4, 300 sec: 3713.7). Total num frames: 557056. Throughput: 0: 992.9. Samples: 139800. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:31:00,082][02158] Avg episode reward: [(0, '4.691')] [2025-08-21 19:31:05,087][02158] Fps is (10 sec: 3274.5, 60 sec: 3890.7, 300 sec: 3673.0). Total num frames: 569344. Throughput: 0: 987.6. Samples: 141570. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:31:05,090][02158] Avg episode reward: [(0, '4.630')] [2025-08-21 19:31:06,214][02335] Updated weights for policy 0, policy_version 140 (0.0017) [2025-08-21 19:31:10,080][02158] Fps is (10 sec: 3277.0, 60 sec: 3891.2, 300 sec: 3686.4). Total num frames: 589824. Throughput: 0: 954.6. Samples: 146930. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:31:10,081][02158] Avg episode reward: [(0, '4.455')] [2025-08-21 19:31:15,080][02158] Fps is (10 sec: 3689.0, 60 sec: 3822.9, 300 sec: 3674.0). Total num frames: 606208. Throughput: 0: 952.9. Samples: 152732. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:31:15,084][02158] Avg episode reward: [(0, '4.363')] [2025-08-21 19:31:16,870][02335] Updated weights for policy 0, policy_version 150 (0.0020) [2025-08-21 19:31:20,080][02158] Fps is (10 sec: 3686.5, 60 sec: 3891.2, 300 sec: 3686.4). Total num frames: 626688. Throughput: 0: 953.8. Samples: 154956. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:31:20,087][02158] Avg episode reward: [(0, '4.421')] [2025-08-21 19:31:25,080][02158] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3698.1). Total num frames: 647168. Throughput: 0: 946.1. Samples: 161784. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:31:25,081][02158] Avg episode reward: [(0, '4.609')] [2025-08-21 19:31:26,168][02335] Updated weights for policy 0, policy_version 160 (0.0020) [2025-08-21 19:31:30,081][02158] Fps is (10 sec: 4095.5, 60 sec: 3891.4, 300 sec: 3709.1). Total num frames: 667648. Throughput: 0: 945.5. Samples: 167588. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:31:30,082][02158] Avg episode reward: [(0, '4.615')] [2025-08-21 19:31:35,081][02158] Fps is (10 sec: 3686.0, 60 sec: 3822.9, 300 sec: 3697.4). Total num frames: 684032. Throughput: 0: 946.0. Samples: 169890. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:31:35,084][02158] Avg episode reward: [(0, '4.605')] [2025-08-21 19:31:36,954][02335] Updated weights for policy 0, policy_version 170 (0.0024) [2025-08-21 19:31:40,080][02158] Fps is (10 sec: 4096.5, 60 sec: 3822.9, 300 sec: 3729.5). Total num frames: 708608. Throughput: 0: 945.1. Samples: 176748. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:31:40,081][02158] Avg episode reward: [(0, '4.501')] [2025-08-21 19:31:45,080][02158] Fps is (10 sec: 4096.5, 60 sec: 3822.9, 300 sec: 3717.9). Total num frames: 724992. Throughput: 0: 943.6. Samples: 182260. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:31:45,081][02158] Avg episode reward: [(0, '4.322')] [2025-08-21 19:31:47,861][02335] Updated weights for policy 0, policy_version 180 (0.0018) [2025-08-21 19:31:50,080][02158] Fps is (10 sec: 3686.4, 60 sec: 3822.9, 300 sec: 3727.4). Total num frames: 745472. Throughput: 0: 960.3. Samples: 184778. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:31:50,081][02158] Avg episode reward: [(0, '4.526')] [2025-08-21 19:31:55,080][02158] Fps is (10 sec: 4096.0, 60 sec: 3822.9, 300 sec: 3736.4). Total num frames: 765952. Throughput: 0: 994.2. Samples: 191668. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:31:55,084][02158] Avg episode reward: [(0, '4.784')] [2025-08-21 19:31:55,110][02316] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000188_770048.pth... [2025-08-21 19:31:55,224][02316] Saving new best policy, reward=4.784! [2025-08-21 19:31:57,170][02335] Updated weights for policy 0, policy_version 190 (0.0022) [2025-08-21 19:32:00,080][02158] Fps is (10 sec: 4096.0, 60 sec: 3823.0, 300 sec: 3744.9). Total num frames: 786432. Throughput: 0: 983.6. Samples: 196994. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:32:00,081][02158] Avg episode reward: [(0, '4.607')] [2025-08-21 19:32:05,080][02158] Fps is (10 sec: 3686.4, 60 sec: 3891.7, 300 sec: 3734.0). Total num frames: 802816. Throughput: 0: 992.8. Samples: 199632. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:32:05,083][02158] Avg episode reward: [(0, '4.632')] [2025-08-21 19:32:07,792][02335] Updated weights for policy 0, policy_version 200 (0.0016) [2025-08-21 19:32:10,080][02158] Fps is (10 sec: 4096.0, 60 sec: 3959.5, 300 sec: 3760.9). Total num frames: 827392. Throughput: 0: 993.5. Samples: 206490. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:32:10,083][02158] Avg episode reward: [(0, '4.740')] [2025-08-21 19:32:15,080][02158] Fps is (10 sec: 4096.0, 60 sec: 3959.5, 300 sec: 3750.1). Total num frames: 843776. Throughput: 0: 983.8. Samples: 211856. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:32:15,084][02158] Avg episode reward: [(0, '4.690')] [2025-08-21 19:32:18,804][02335] Updated weights for policy 0, policy_version 210 (0.0013) [2025-08-21 19:32:20,080][02158] Fps is (10 sec: 3686.4, 60 sec: 3959.5, 300 sec: 3757.6). Total num frames: 864256. Throughput: 0: 992.1. Samples: 214534. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:32:20,083][02158] Avg episode reward: [(0, '4.600')] [2025-08-21 19:32:25,081][02158] Fps is (10 sec: 4505.1, 60 sec: 4027.7, 300 sec: 3782.2). Total num frames: 888832. Throughput: 0: 992.7. Samples: 221420. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:32:25,084][02158] Avg episode reward: [(0, '4.535')] [2025-08-21 19:32:28,425][02335] Updated weights for policy 0, policy_version 220 (0.0011) [2025-08-21 19:32:30,083][02158] Fps is (10 sec: 4094.6, 60 sec: 3959.3, 300 sec: 3771.7). Total num frames: 905216. Throughput: 0: 988.9. Samples: 226764. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:32:30,085][02158] Avg episode reward: [(0, '4.607')] [2025-08-21 19:32:35,080][02158] Fps is (10 sec: 3686.8, 60 sec: 4027.8, 300 sec: 3778.4). Total num frames: 925696. Throughput: 0: 994.8. Samples: 229544. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:32:35,084][02158] Avg episode reward: [(0, '4.567')] [2025-08-21 19:32:38,656][02335] Updated weights for policy 0, policy_version 230 (0.0019) [2025-08-21 19:32:40,080][02158] Fps is (10 sec: 4097.4, 60 sec: 3959.5, 300 sec: 3784.7). Total num frames: 946176. Throughput: 0: 993.7. Samples: 236386. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:32:40,089][02158] Avg episode reward: [(0, '4.807')] [2025-08-21 19:32:40,091][02316] Saving new best policy, reward=4.807! [2025-08-21 19:32:45,084][02158] Fps is (10 sec: 3684.9, 60 sec: 3959.2, 300 sec: 3774.7). Total num frames: 962560. Throughput: 0: 989.7. Samples: 241536. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:32:45,085][02158] Avg episode reward: [(0, '4.755')] [2025-08-21 19:32:49,291][02335] Updated weights for policy 0, policy_version 240 (0.0020) [2025-08-21 19:32:50,080][02158] Fps is (10 sec: 3686.4, 60 sec: 3959.5, 300 sec: 3780.9). Total num frames: 983040. Throughput: 0: 997.1. Samples: 244500. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:32:50,081][02158] Avg episode reward: [(0, '4.559')] [2025-08-21 19:32:55,080][02158] Fps is (10 sec: 4507.5, 60 sec: 4027.7, 300 sec: 3802.3). Total num frames: 1007616. Throughput: 0: 1006.4. Samples: 251778. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:32:55,081][02158] Avg episode reward: [(0, '4.365')] [2025-08-21 19:32:59,098][02335] Updated weights for policy 0, policy_version 250 (0.0014) [2025-08-21 19:33:00,081][02158] Fps is (10 sec: 4095.5, 60 sec: 3959.4, 300 sec: 3792.6). Total num frames: 1024000. Throughput: 0: 1005.3. Samples: 257094. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:33:00,084][02158] Avg episode reward: [(0, '4.445')] [2025-08-21 19:33:05,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 3813.0). Total num frames: 1048576. Throughput: 0: 1012.8. Samples: 260110. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:33:05,084][02158] Avg episode reward: [(0, '4.688')] [2025-08-21 19:33:08,647][02335] Updated weights for policy 0, policy_version 260 (0.0028) [2025-08-21 19:33:10,080][02158] Fps is (10 sec: 4506.2, 60 sec: 4027.7, 300 sec: 3818.1). Total num frames: 1069056. Throughput: 0: 1012.7. Samples: 266990. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:33:10,084][02158] Avg episode reward: [(0, '4.870')] [2025-08-21 19:33:10,088][02316] Saving new best policy, reward=4.870! [2025-08-21 19:33:15,083][02158] Fps is (10 sec: 3685.2, 60 sec: 4027.5, 300 sec: 3808.5). Total num frames: 1085440. Throughput: 0: 1004.1. Samples: 271946. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:33:15,084][02158] Avg episode reward: [(0, '4.764')] [2025-08-21 19:33:19,531][02335] Updated weights for policy 0, policy_version 270 (0.0017) [2025-08-21 19:33:20,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.7, 300 sec: 3813.5). Total num frames: 1105920. Throughput: 0: 1010.8. Samples: 275030. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:33:20,081][02158] Avg episode reward: [(0, '4.716')] [2025-08-21 19:33:25,080][02158] Fps is (10 sec: 4507.0, 60 sec: 4027.8, 300 sec: 3832.2). Total num frames: 1130496. Throughput: 0: 1014.5. Samples: 282038. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:33:25,088][02158] Avg episode reward: [(0, '4.593')] [2025-08-21 19:33:30,083][02158] Fps is (10 sec: 3685.3, 60 sec: 3959.5, 300 sec: 3873.8). Total num frames: 1142784. Throughput: 0: 1014.8. Samples: 287200. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:33:30,086][02158] Avg episode reward: [(0, '4.832')] [2025-08-21 19:33:30,148][02335] Updated weights for policy 0, policy_version 280 (0.0020) [2025-08-21 19:33:35,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.7, 300 sec: 3957.2). Total num frames: 1167360. Throughput: 0: 1021.5. Samples: 290468. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:33:35,084][02158] Avg episode reward: [(0, '4.984')] [2025-08-21 19:33:35,090][02316] Saving new best policy, reward=4.984! [2025-08-21 19:33:38,802][02335] Updated weights for policy 0, policy_version 290 (0.0028) [2025-08-21 19:33:40,080][02158] Fps is (10 sec: 4916.7, 60 sec: 4096.0, 300 sec: 3957.2). Total num frames: 1191936. Throughput: 0: 1016.3. Samples: 297512. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:33:40,081][02158] Avg episode reward: [(0, '4.819')] [2025-08-21 19:33:45,080][02158] Fps is (10 sec: 4095.9, 60 sec: 4096.3, 300 sec: 3971.0). Total num frames: 1208320. Throughput: 0: 1013.1. Samples: 302682. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:33:45,084][02158] Avg episode reward: [(0, '5.092')] [2025-08-21 19:33:45,090][02316] Saving new best policy, reward=5.092! [2025-08-21 19:33:49,141][02335] Updated weights for policy 0, policy_version 300 (0.0036) [2025-08-21 19:33:50,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 3971.0). Total num frames: 1232896. Throughput: 0: 1023.5. Samples: 306168. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:33:50,084][02158] Avg episode reward: [(0, '5.382')] [2025-08-21 19:33:50,087][02316] Saving new best policy, reward=5.382! [2025-08-21 19:33:55,080][02158] Fps is (10 sec: 4505.7, 60 sec: 4096.0, 300 sec: 3957.2). Total num frames: 1253376. Throughput: 0: 1031.2. Samples: 313392. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:33:55,083][02158] Avg episode reward: [(0, '5.702')] [2025-08-21 19:33:55,152][02316] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000307_1257472.pth... [2025-08-21 19:33:55,276][02316] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000073_299008.pth [2025-08-21 19:33:55,287][02316] Saving new best policy, reward=5.702! [2025-08-21 19:33:59,611][02335] Updated weights for policy 0, policy_version 310 (0.0020) [2025-08-21 19:34:00,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.1, 300 sec: 3957.2). Total num frames: 1269760. Throughput: 0: 1030.3. Samples: 318306. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:34:00,084][02158] Avg episode reward: [(0, '5.593')] [2025-08-21 19:34:05,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 3971.0). Total num frames: 1294336. Throughput: 0: 1041.7. Samples: 321906. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:34:05,081][02158] Avg episode reward: [(0, '5.400')] [2025-08-21 19:34:08,070][02335] Updated weights for policy 0, policy_version 320 (0.0022) [2025-08-21 19:34:10,080][02158] Fps is (10 sec: 4915.2, 60 sec: 4164.3, 300 sec: 3984.9). Total num frames: 1318912. Throughput: 0: 1047.9. Samples: 329194. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:34:10,083][02158] Avg episode reward: [(0, '5.315')] [2025-08-21 19:34:15,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.2, 300 sec: 3971.0). Total num frames: 1331200. Throughput: 0: 1042.2. Samples: 334098. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:34:15,081][02158] Avg episode reward: [(0, '5.260')] [2025-08-21 19:34:18,385][02335] Updated weights for policy 0, policy_version 330 (0.0020) [2025-08-21 19:34:20,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 3984.9). Total num frames: 1355776. Throughput: 0: 1049.7. Samples: 337706. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:34:20,081][02158] Avg episode reward: [(0, '5.562')] [2025-08-21 19:34:25,083][02158] Fps is (10 sec: 4913.7, 60 sec: 4164.1, 300 sec: 3998.8). Total num frames: 1380352. Throughput: 0: 1053.8. Samples: 344934. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:34:25,085][02158] Avg episode reward: [(0, '5.435')] [2025-08-21 19:34:28,378][02335] Updated weights for policy 0, policy_version 340 (0.0015) [2025-08-21 19:34:30,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4232.8, 300 sec: 3998.8). Total num frames: 1396736. Throughput: 0: 1050.5. Samples: 349956. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:34:30,084][02158] Avg episode reward: [(0, '5.254')] [2025-08-21 19:34:35,080][02158] Fps is (10 sec: 4097.2, 60 sec: 4232.5, 300 sec: 3998.8). Total num frames: 1421312. Throughput: 0: 1054.2. Samples: 353608. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:34:35,084][02158] Avg episode reward: [(0, '5.550')] [2025-08-21 19:34:37,315][02335] Updated weights for policy 0, policy_version 350 (0.0015) [2025-08-21 19:34:40,080][02158] Fps is (10 sec: 4915.2, 60 sec: 4232.5, 300 sec: 4026.6). Total num frames: 1445888. Throughput: 0: 1053.9. Samples: 360818. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:34:40,081][02158] Avg episode reward: [(0, '5.635')] [2025-08-21 19:34:45,080][02158] Fps is (10 sec: 3686.3, 60 sec: 4164.3, 300 sec: 4012.7). Total num frames: 1458176. Throughput: 0: 1055.1. Samples: 365784. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:34:45,081][02158] Avg episode reward: [(0, '5.662')] [2025-08-21 19:34:47,656][02335] Updated weights for policy 0, policy_version 360 (0.0015) [2025-08-21 19:34:50,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4012.7). Total num frames: 1482752. Throughput: 0: 1055.1. Samples: 369384. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:34:50,081][02158] Avg episode reward: [(0, '5.602')] [2025-08-21 19:34:55,080][02158] Fps is (10 sec: 4915.3, 60 sec: 4232.5, 300 sec: 4026.6). Total num frames: 1507328. Throughput: 0: 1054.3. Samples: 376638. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:34:55,083][02158] Avg episode reward: [(0, '5.167')] [2025-08-21 19:34:57,568][02335] Updated weights for policy 0, policy_version 370 (0.0022) [2025-08-21 19:35:00,080][02158] Fps is (10 sec: 4095.9, 60 sec: 4232.5, 300 sec: 4026.6). Total num frames: 1523712. Throughput: 0: 1055.6. Samples: 381602. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:35:00,083][02158] Avg episode reward: [(0, '5.232')] [2025-08-21 19:35:05,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4232.5, 300 sec: 4040.5). Total num frames: 1548288. Throughput: 0: 1055.7. Samples: 385212. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:35:05,081][02158] Avg episode reward: [(0, '6.109')] [2025-08-21 19:35:05,087][02316] Saving new best policy, reward=6.109! [2025-08-21 19:35:06,527][02335] Updated weights for policy 0, policy_version 380 (0.0011) [2025-08-21 19:35:10,083][02158] Fps is (10 sec: 4504.1, 60 sec: 4164.0, 300 sec: 4040.4). Total num frames: 1568768. Throughput: 0: 1054.3. Samples: 392376. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:35:10,084][02158] Avg episode reward: [(0, '6.492')] [2025-08-21 19:35:10,098][02316] Saving new best policy, reward=6.492! [2025-08-21 19:35:15,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4232.5, 300 sec: 4040.5). Total num frames: 1585152. Throughput: 0: 1053.5. Samples: 397362. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:35:15,081][02158] Avg episode reward: [(0, '6.100')] [2025-08-21 19:35:16,920][02335] Updated weights for policy 0, policy_version 390 (0.0021) [2025-08-21 19:35:20,080][02158] Fps is (10 sec: 4097.4, 60 sec: 4232.5, 300 sec: 4040.5). Total num frames: 1609728. Throughput: 0: 1051.8. Samples: 400940. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:35:20,081][02158] Avg episode reward: [(0, '6.141')] [2025-08-21 19:35:25,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.5, 300 sec: 4054.4). Total num frames: 1630208. Throughput: 0: 1048.3. Samples: 407992. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:35:25,086][02158] Avg episode reward: [(0, '6.681')] [2025-08-21 19:35:25,093][02316] Saving new best policy, reward=6.681! [2025-08-21 19:35:26,769][02335] Updated weights for policy 0, policy_version 400 (0.0011) [2025-08-21 19:35:30,082][02158] Fps is (10 sec: 3685.7, 60 sec: 4164.1, 300 sec: 4040.4). Total num frames: 1646592. Throughput: 0: 1032.3. Samples: 412240. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:35:30,083][02158] Avg episode reward: [(0, '6.904')] [2025-08-21 19:35:30,087][02316] Saving new best policy, reward=6.904! [2025-08-21 19:35:35,080][02158] Fps is (10 sec: 3276.7, 60 sec: 4027.7, 300 sec: 4012.7). Total num frames: 1662976. Throughput: 0: 1005.0. Samples: 414608. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:35:35,081][02158] Avg episode reward: [(0, '6.771')] [2025-08-21 19:35:37,628][02335] Updated weights for policy 0, policy_version 410 (0.0023) [2025-08-21 19:35:40,087][02158] Fps is (10 sec: 4094.0, 60 sec: 4027.3, 300 sec: 4040.4). Total num frames: 1687552. Throughput: 0: 1000.9. Samples: 421684. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:35:40,089][02158] Avg episode reward: [(0, '7.111')] [2025-08-21 19:35:40,093][02316] Saving new best policy, reward=7.111! [2025-08-21 19:35:45,080][02158] Fps is (10 sec: 4096.2, 60 sec: 4096.0, 300 sec: 4026.6). Total num frames: 1703936. Throughput: 0: 1004.1. Samples: 426788. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:35:45,081][02158] Avg episode reward: [(0, '7.367')] [2025-08-21 19:35:45,092][02316] Saving new best policy, reward=7.367! [2025-08-21 19:35:48,066][02335] Updated weights for policy 0, policy_version 420 (0.0011) [2025-08-21 19:35:50,080][02158] Fps is (10 sec: 4098.8, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 1728512. Throughput: 0: 1003.8. Samples: 430382. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:35:50,081][02158] Avg episode reward: [(0, '7.651')] [2025-08-21 19:35:50,085][02316] Saving new best policy, reward=7.651! [2025-08-21 19:35:55,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4027.7, 300 sec: 4040.5). Total num frames: 1748992. Throughput: 0: 996.5. Samples: 437216. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:35:55,084][02158] Avg episode reward: [(0, '7.611')] [2025-08-21 19:35:55,093][02316] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000427_1748992.pth... [2025-08-21 19:35:55,245][02316] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000188_770048.pth [2025-08-21 19:35:58,464][02335] Updated weights for policy 0, policy_version 430 (0.0016) [2025-08-21 19:36:00,083][02158] Fps is (10 sec: 3685.4, 60 sec: 4027.5, 300 sec: 4054.4). Total num frames: 1765376. Throughput: 0: 1005.0. Samples: 442592. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:36:00,085][02158] Avg episode reward: [(0, '8.203')] [2025-08-21 19:36:00,112][02316] Saving new best policy, reward=8.203! [2025-08-21 19:36:05,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.7, 300 sec: 4068.2). Total num frames: 1789952. Throughput: 0: 1003.2. Samples: 446086. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:36:05,081][02158] Avg episode reward: [(0, '8.632')] [2025-08-21 19:36:05,090][02316] Saving new best policy, reward=8.632! [2025-08-21 19:36:07,082][02335] Updated weights for policy 0, policy_version 440 (0.0033) [2025-08-21 19:36:10,080][02158] Fps is (10 sec: 4507.0, 60 sec: 4028.0, 300 sec: 4082.1). Total num frames: 1810432. Throughput: 0: 992.8. Samples: 452668. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:36:10,085][02158] Avg episode reward: [(0, '8.371')] [2025-08-21 19:36:15,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4082.1). Total num frames: 1830912. Throughput: 0: 1020.8. Samples: 458174. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:36:15,081][02158] Avg episode reward: [(0, '7.797')] [2025-08-21 19:36:17,505][02335] Updated weights for policy 0, policy_version 450 (0.0029) [2025-08-21 19:36:20,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.0, 300 sec: 4096.0). Total num frames: 1855488. Throughput: 0: 1048.0. Samples: 461768. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:36:20,081][02158] Avg episode reward: [(0, '7.527')] [2025-08-21 19:36:25,080][02158] Fps is (10 sec: 4095.9, 60 sec: 4027.7, 300 sec: 4082.1). Total num frames: 1871872. Throughput: 0: 1038.5. Samples: 468410. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:36:25,087][02158] Avg episode reward: [(0, '7.681')] [2025-08-21 19:36:27,837][02335] Updated weights for policy 0, policy_version 460 (0.0011) [2025-08-21 19:36:30,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.1, 300 sec: 4096.0). Total num frames: 1892352. Throughput: 0: 1052.0. Samples: 474128. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:36:30,083][02158] Avg episode reward: [(0, '8.667')] [2025-08-21 19:36:30,085][02316] Saving new best policy, reward=8.667! [2025-08-21 19:36:35,080][02158] Fps is (10 sec: 4505.7, 60 sec: 4232.6, 300 sec: 4096.0). Total num frames: 1916928. Throughput: 0: 1051.0. Samples: 477678. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:36:35,082][02158] Avg episode reward: [(0, '9.759')] [2025-08-21 19:36:35,088][02316] Saving new best policy, reward=9.759! [2025-08-21 19:36:36,385][02335] Updated weights for policy 0, policy_version 470 (0.0012) [2025-08-21 19:36:40,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.7, 300 sec: 4109.9). Total num frames: 1937408. Throughput: 0: 1040.2. Samples: 484026. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:36:40,083][02158] Avg episode reward: [(0, '10.829')] [2025-08-21 19:36:40,084][02316] Saving new best policy, reward=10.829! [2025-08-21 19:36:45,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4096.0). Total num frames: 1953792. Throughput: 0: 1045.9. Samples: 489654. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:36:45,081][02158] Avg episode reward: [(0, '10.728')] [2025-08-21 19:36:46,925][02335] Updated weights for policy 0, policy_version 480 (0.0023) [2025-08-21 19:36:50,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 1978368. Throughput: 0: 1046.6. Samples: 493182. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:36:50,081][02158] Avg episode reward: [(0, '10.105')] [2025-08-21 19:36:55,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 1998848. Throughput: 0: 1041.9. Samples: 499554. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:36:55,084][02158] Avg episode reward: [(0, '9.095')] [2025-08-21 19:36:57,330][02335] Updated weights for policy 0, policy_version 490 (0.0015) [2025-08-21 19:37:00,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4232.7, 300 sec: 4123.8). Total num frames: 2019328. Throughput: 0: 1049.2. Samples: 505390. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:37:00,085][02158] Avg episode reward: [(0, '9.966')] [2025-08-21 19:37:05,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4232.5, 300 sec: 4123.8). Total num frames: 2043904. Throughput: 0: 1050.0. Samples: 509016. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:37:05,085][02158] Avg episode reward: [(0, '10.056')] [2025-08-21 19:37:05,810][02335] Updated weights for policy 0, policy_version 500 (0.0017) [2025-08-21 19:37:10,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 2060288. Throughput: 0: 1040.5. Samples: 515230. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:37:10,087][02158] Avg episode reward: [(0, '9.808')] [2025-08-21 19:37:15,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 2080768. Throughput: 0: 1047.8. Samples: 521280. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:37:15,083][02158] Avg episode reward: [(0, '9.347')] [2025-08-21 19:37:16,176][02335] Updated weights for policy 0, policy_version 510 (0.0018) [2025-08-21 19:37:20,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 2105344. Throughput: 0: 1047.5. Samples: 524816. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:37:20,081][02158] Avg episode reward: [(0, '9.463')] [2025-08-21 19:37:25,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 2121728. Throughput: 0: 1041.3. Samples: 530886. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:37:25,084][02158] Avg episode reward: [(0, '10.514')] [2025-08-21 19:37:26,532][02335] Updated weights for policy 0, policy_version 520 (0.0012) [2025-08-21 19:37:30,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4232.5, 300 sec: 4137.7). Total num frames: 2146304. Throughput: 0: 1054.9. Samples: 537124. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:37:30,084][02158] Avg episode reward: [(0, '12.477')] [2025-08-21 19:37:30,087][02316] Saving new best policy, reward=12.477! [2025-08-21 19:37:35,055][02335] Updated weights for policy 0, policy_version 530 (0.0022) [2025-08-21 19:37:35,080][02158] Fps is (10 sec: 4915.2, 60 sec: 4232.5, 300 sec: 4151.5). Total num frames: 2170880. Throughput: 0: 1056.0. Samples: 540704. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:37:35,081][02158] Avg episode reward: [(0, '14.177')] [2025-08-21 19:37:35,085][02316] Saving new best policy, reward=14.177! [2025-08-21 19:37:40,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4151.6). Total num frames: 2187264. Throughput: 0: 1043.6. Samples: 546516. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:37:40,086][02158] Avg episode reward: [(0, '14.876')] [2025-08-21 19:37:40,091][02316] Saving new best policy, reward=14.876! [2025-08-21 19:37:45,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4232.5, 300 sec: 4151.5). Total num frames: 2207744. Throughput: 0: 1052.6. Samples: 552756. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:37:45,085][02158] Avg episode reward: [(0, '15.394')] [2025-08-21 19:37:45,097][02316] Saving new best policy, reward=15.394! [2025-08-21 19:37:45,517][02335] Updated weights for policy 0, policy_version 540 (0.0016) [2025-08-21 19:37:50,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4232.5, 300 sec: 4151.5). Total num frames: 2232320. Throughput: 0: 1049.0. Samples: 556220. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:37:50,081][02158] Avg episode reward: [(0, '14.481')] [2025-08-21 19:37:55,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4151.6). Total num frames: 2248704. Throughput: 0: 1038.9. Samples: 561980. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:37:55,081][02158] Avg episode reward: [(0, '15.061')] [2025-08-21 19:37:55,089][02316] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000549_2248704.pth... [2025-08-21 19:37:55,235][02316] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000307_1257472.pth [2025-08-21 19:37:56,075][02335] Updated weights for policy 0, policy_version 550 (0.0022) [2025-08-21 19:38:00,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4137.7). Total num frames: 2269184. Throughput: 0: 1046.3. Samples: 568362. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:38:00,085][02158] Avg episode reward: [(0, '13.618')] [2025-08-21 19:38:04,481][02335] Updated weights for policy 0, policy_version 560 (0.0015) [2025-08-21 19:38:05,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4151.5). Total num frames: 2293760. Throughput: 0: 1047.2. Samples: 571940. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:38:05,086][02158] Avg episode reward: [(0, '12.990')] [2025-08-21 19:38:10,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4151.6). Total num frames: 2310144. Throughput: 0: 1036.9. Samples: 577548. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:38:10,082][02158] Avg episode reward: [(0, '12.770')] [2025-08-21 19:38:14,988][02335] Updated weights for policy 0, policy_version 570 (0.0022) [2025-08-21 19:38:15,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4232.5, 300 sec: 4165.4). Total num frames: 2334720. Throughput: 0: 1046.1. Samples: 584200. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:38:15,083][02158] Avg episode reward: [(0, '14.517')] [2025-08-21 19:38:20,080][02158] Fps is (10 sec: 4505.3, 60 sec: 4164.2, 300 sec: 4151.5). Total num frames: 2355200. Throughput: 0: 1042.7. Samples: 587628. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:38:20,083][02158] Avg episode reward: [(0, '14.403')] [2025-08-21 19:38:25,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4165.5). Total num frames: 2371584. Throughput: 0: 1034.7. Samples: 593076. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:38:25,084][02158] Avg episode reward: [(0, '15.084')] [2025-08-21 19:38:25,541][02335] Updated weights for policy 0, policy_version 580 (0.0020) [2025-08-21 19:38:30,080][02158] Fps is (10 sec: 4096.2, 60 sec: 4164.3, 300 sec: 4165.4). Total num frames: 2396160. Throughput: 0: 1043.2. Samples: 599700. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:38:30,084][02158] Avg episode reward: [(0, '16.020')] [2025-08-21 19:38:30,087][02316] Saving new best policy, reward=16.020! [2025-08-21 19:38:34,123][02335] Updated weights for policy 0, policy_version 590 (0.0019) [2025-08-21 19:38:35,080][02158] Fps is (10 sec: 4915.1, 60 sec: 4164.3, 300 sec: 4165.4). Total num frames: 2420736. Throughput: 0: 1043.6. Samples: 603180. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:38:35,081][02158] Avg episode reward: [(0, '15.834')] [2025-08-21 19:38:40,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4151.5). Total num frames: 2433024. Throughput: 0: 1034.2. Samples: 608520. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:38:40,084][02158] Avg episode reward: [(0, '16.158')] [2025-08-21 19:38:40,088][02316] Saving new best policy, reward=16.158! [2025-08-21 19:38:44,754][02335] Updated weights for policy 0, policy_version 600 (0.0020) [2025-08-21 19:38:45,080][02158] Fps is (10 sec: 3686.5, 60 sec: 4164.3, 300 sec: 4151.5). Total num frames: 2457600. Throughput: 0: 1039.2. Samples: 615128. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:38:45,085][02158] Avg episode reward: [(0, '16.416')] [2025-08-21 19:38:45,097][02316] Saving new best policy, reward=16.416! [2025-08-21 19:38:50,080][02158] Fps is (10 sec: 4915.2, 60 sec: 4164.3, 300 sec: 4165.4). Total num frames: 2482176. Throughput: 0: 1035.2. Samples: 618522. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:38:50,082][02158] Avg episode reward: [(0, '15.465')] [2025-08-21 19:38:55,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4151.5). Total num frames: 2494464. Throughput: 0: 1028.4. Samples: 623824. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:38:55,081][02158] Avg episode reward: [(0, '16.250')] [2025-08-21 19:38:55,245][02335] Updated weights for policy 0, policy_version 610 (0.0014) [2025-08-21 19:39:00,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4151.5). Total num frames: 2519040. Throughput: 0: 1029.2. Samples: 630512. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:39:00,081][02158] Avg episode reward: [(0, '16.379')] [2025-08-21 19:39:04,175][02335] Updated weights for policy 0, policy_version 620 (0.0014) [2025-08-21 19:39:05,083][02158] Fps is (10 sec: 4504.2, 60 sec: 4095.8, 300 sec: 4137.6). Total num frames: 2539520. Throughput: 0: 1032.1. Samples: 634076. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:39:05,084][02158] Avg episode reward: [(0, '15.571')] [2025-08-21 19:39:10,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4151.5). Total num frames: 2555904. Throughput: 0: 1026.1. Samples: 639250. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:39:10,084][02158] Avg episode reward: [(0, '15.970')] [2025-08-21 19:39:14,436][02335] Updated weights for policy 0, policy_version 630 (0.0032) [2025-08-21 19:39:15,080][02158] Fps is (10 sec: 4097.3, 60 sec: 4096.0, 300 sec: 4151.5). Total num frames: 2580480. Throughput: 0: 1030.8. Samples: 646086. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:39:15,082][02158] Avg episode reward: [(0, '15.880')] [2025-08-21 19:39:20,082][02158] Fps is (10 sec: 4914.2, 60 sec: 4164.2, 300 sec: 4151.6). Total num frames: 2605056. Throughput: 0: 1031.0. Samples: 649578. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:39:20,084][02158] Avg episode reward: [(0, '17.246')] [2025-08-21 19:39:20,087][02316] Saving new best policy, reward=17.246! [2025-08-21 19:39:25,079][02335] Updated weights for policy 0, policy_version 640 (0.0031) [2025-08-21 19:39:25,081][02158] Fps is (10 sec: 4095.7, 60 sec: 4164.2, 300 sec: 4151.5). Total num frames: 2621440. Throughput: 0: 1024.1. Samples: 654606. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:39:25,082][02158] Avg episode reward: [(0, '17.860')] [2025-08-21 19:39:25,087][02316] Saving new best policy, reward=17.860! [2025-08-21 19:39:30,080][02158] Fps is (10 sec: 3687.0, 60 sec: 4096.0, 300 sec: 4137.6). Total num frames: 2641920. Throughput: 0: 1029.5. Samples: 661458. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:39:30,081][02158] Avg episode reward: [(0, '18.267')] [2025-08-21 19:39:30,088][02316] Saving new best policy, reward=18.267! [2025-08-21 19:39:33,879][02335] Updated weights for policy 0, policy_version 650 (0.0017) [2025-08-21 19:39:35,084][02158] Fps is (10 sec: 4504.2, 60 sec: 4095.7, 300 sec: 4137.6). Total num frames: 2666496. Throughput: 0: 1032.3. Samples: 664980. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:39:35,085][02158] Avg episode reward: [(0, '19.785')] [2025-08-21 19:39:35,097][02316] Saving new best policy, reward=19.785! [2025-08-21 19:39:40,080][02158] Fps is (10 sec: 4096.1, 60 sec: 4164.3, 300 sec: 4151.5). Total num frames: 2682880. Throughput: 0: 1021.3. Samples: 669784. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:39:40,081][02158] Avg episode reward: [(0, '19.947')] [2025-08-21 19:39:40,083][02316] Saving new best policy, reward=19.947! [2025-08-21 19:39:44,535][02335] Updated weights for policy 0, policy_version 660 (0.0023) [2025-08-21 19:39:45,080][02158] Fps is (10 sec: 3687.9, 60 sec: 4096.0, 300 sec: 4137.7). Total num frames: 2703360. Throughput: 0: 1028.0. Samples: 676770. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:39:45,081][02158] Avg episode reward: [(0, '19.140')] [2025-08-21 19:39:50,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.7, 300 sec: 4123.8). Total num frames: 2723840. Throughput: 0: 1026.9. Samples: 680284. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:39:50,085][02158] Avg episode reward: [(0, '18.758')] [2025-08-21 19:39:55,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4123.8). Total num frames: 2740224. Throughput: 0: 1021.2. Samples: 685206. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:39:55,082][02158] Avg episode reward: [(0, '19.772')] [2025-08-21 19:39:55,136][02335] Updated weights for policy 0, policy_version 670 (0.0017) [2025-08-21 19:39:55,137][02316] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000670_2744320.pth... [2025-08-21 19:39:55,248][02316] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000427_1748992.pth [2025-08-21 19:40:00,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4123.8). Total num frames: 2764800. Throughput: 0: 1021.6. Samples: 692058. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:40:00,087][02158] Avg episode reward: [(0, '19.472')] [2025-08-21 19:40:04,069][02335] Updated weights for policy 0, policy_version 680 (0.0014) [2025-08-21 19:40:05,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.2, 300 sec: 4123.8). Total num frames: 2785280. Throughput: 0: 1022.1. Samples: 695570. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:40:05,083][02158] Avg episode reward: [(0, '20.974')] [2025-08-21 19:40:05,091][02316] Saving new best policy, reward=20.974! [2025-08-21 19:40:10,080][02158] Fps is (10 sec: 3686.3, 60 sec: 4096.0, 300 sec: 4123.8). Total num frames: 2801664. Throughput: 0: 1018.5. Samples: 700436. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:40:10,085][02158] Avg episode reward: [(0, '21.681')] [2025-08-21 19:40:10,096][02316] Saving new best policy, reward=21.681! [2025-08-21 19:40:14,462][02335] Updated weights for policy 0, policy_version 690 (0.0018) [2025-08-21 19:40:15,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4123.8). Total num frames: 2826240. Throughput: 0: 1020.8. Samples: 707394. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:40:15,085][02158] Avg episode reward: [(0, '21.505')] [2025-08-21 19:40:20,080][02158] Fps is (10 sec: 4505.7, 60 sec: 4027.9, 300 sec: 4123.8). Total num frames: 2846720. Throughput: 0: 1019.2. Samples: 710840. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:40:20,084][02158] Avg episode reward: [(0, '21.167')] [2025-08-21 19:40:25,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.8, 300 sec: 4123.8). Total num frames: 2863104. Throughput: 0: 1018.4. Samples: 715614. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:40:25,081][02158] Avg episode reward: [(0, '20.521')] [2025-08-21 19:40:25,336][02335] Updated weights for policy 0, policy_version 700 (0.0013) [2025-08-21 19:40:30,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4151.5). Total num frames: 2887680. Throughput: 0: 1020.2. Samples: 722678. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:40:30,081][02158] Avg episode reward: [(0, '19.435')] [2025-08-21 19:40:34,410][02335] Updated weights for policy 0, policy_version 710 (0.0012) [2025-08-21 19:40:35,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4028.0, 300 sec: 4137.8). Total num frames: 2908160. Throughput: 0: 1022.0. Samples: 726272. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:40:35,084][02158] Avg episode reward: [(0, '18.893')] [2025-08-21 19:40:40,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.7, 300 sec: 4137.7). Total num frames: 2924544. Throughput: 0: 1018.4. Samples: 731036. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:40:40,081][02158] Avg episode reward: [(0, '20.301')] [2025-08-21 19:40:44,671][02335] Updated weights for policy 0, policy_version 720 (0.0014) [2025-08-21 19:40:45,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4137.7). Total num frames: 2949120. Throughput: 0: 1023.6. Samples: 738120. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:40:45,081][02158] Avg episode reward: [(0, '18.974')] [2025-08-21 19:40:50,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.0, 300 sec: 4137.7). Total num frames: 2969600. Throughput: 0: 1020.3. Samples: 741482. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:40:50,081][02158] Avg episode reward: [(0, '19.337')] [2025-08-21 19:40:55,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4137.7). Total num frames: 2985984. Throughput: 0: 1022.3. Samples: 746438. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:40:55,085][02158] Avg episode reward: [(0, '19.696')] [2025-08-21 19:40:55,265][02335] Updated weights for policy 0, policy_version 730 (0.0041) [2025-08-21 19:41:00,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4137.7). Total num frames: 3010560. Throughput: 0: 1023.2. Samples: 753438. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:41:00,084][02158] Avg episode reward: [(0, '17.958')] [2025-08-21 19:41:04,672][02335] Updated weights for policy 0, policy_version 740 (0.0014) [2025-08-21 19:41:05,081][02158] Fps is (10 sec: 4504.9, 60 sec: 4095.9, 300 sec: 4137.6). Total num frames: 3031040. Throughput: 0: 1025.2. Samples: 756976. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:41:05,083][02158] Avg episode reward: [(0, '17.979')] [2025-08-21 19:41:10,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4137.7). Total num frames: 3051520. Throughput: 0: 1029.3. Samples: 761934. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:41:10,081][02158] Avg episode reward: [(0, '18.327')] [2025-08-21 19:41:14,388][02335] Updated weights for policy 0, policy_version 750 (0.0021) [2025-08-21 19:41:15,080][02158] Fps is (10 sec: 4096.6, 60 sec: 4096.0, 300 sec: 4123.8). Total num frames: 3072000. Throughput: 0: 1028.9. Samples: 768980. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:41:15,084][02158] Avg episode reward: [(0, '18.867')] [2025-08-21 19:41:20,080][02158] Fps is (10 sec: 4095.9, 60 sec: 4096.0, 300 sec: 4137.7). Total num frames: 3092480. Throughput: 0: 1023.4. Samples: 772326. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:41:20,085][02158] Avg episode reward: [(0, '19.830')] [2025-08-21 19:41:25,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4123.8). Total num frames: 3108864. Throughput: 0: 1028.6. Samples: 777324. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:41:25,081][02158] Avg episode reward: [(0, '19.722')] [2025-08-21 19:41:25,102][02335] Updated weights for policy 0, policy_version 760 (0.0032) [2025-08-21 19:41:30,080][02158] Fps is (10 sec: 4096.1, 60 sec: 4096.0, 300 sec: 4123.8). Total num frames: 3133440. Throughput: 0: 1029.7. Samples: 784458. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:41:30,081][02158] Avg episode reward: [(0, '19.530')] [2025-08-21 19:41:34,276][02335] Updated weights for policy 0, policy_version 770 (0.0022) [2025-08-21 19:41:35,082][02158] Fps is (10 sec: 4504.7, 60 sec: 4095.9, 300 sec: 4123.7). Total num frames: 3153920. Throughput: 0: 1030.6. Samples: 787860. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:41:35,083][02158] Avg episode reward: [(0, '21.378')] [2025-08-21 19:41:40,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4137.7). Total num frames: 3174400. Throughput: 0: 1032.0. Samples: 792876. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:41:40,085][02158] Avg episode reward: [(0, '21.307')] [2025-08-21 19:41:44,327][02335] Updated weights for policy 0, policy_version 780 (0.0027) [2025-08-21 19:41:45,080][02158] Fps is (10 sec: 4096.8, 60 sec: 4096.0, 300 sec: 4123.8). Total num frames: 3194880. Throughput: 0: 1033.4. Samples: 799942. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:41:45,081][02158] Avg episode reward: [(0, '20.714')] [2025-08-21 19:41:50,080][02158] Fps is (10 sec: 4095.9, 60 sec: 4096.0, 300 sec: 4123.8). Total num frames: 3215360. Throughput: 0: 1029.0. Samples: 803278. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:41:50,081][02158] Avg episode reward: [(0, '20.585')] [2025-08-21 19:41:54,832][02335] Updated weights for policy 0, policy_version 790 (0.0014) [2025-08-21 19:41:55,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 3235840. Throughput: 0: 1030.9. Samples: 808326. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:41:55,081][02158] Avg episode reward: [(0, '19.940')] [2025-08-21 19:41:55,088][02316] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000790_3235840.pth... [2025-08-21 19:41:55,207][02316] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000549_2248704.pth [2025-08-21 19:42:00,080][02158] Fps is (10 sec: 4505.7, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 3260416. Throughput: 0: 1030.0. Samples: 815330. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:42:00,081][02158] Avg episode reward: [(0, '17.558')] [2025-08-21 19:42:04,032][02335] Updated weights for policy 0, policy_version 800 (0.0011) [2025-08-21 19:42:05,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.1, 300 sec: 4123.8). Total num frames: 3276800. Throughput: 0: 1031.4. Samples: 818740. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:42:05,084][02158] Avg episode reward: [(0, '17.162')] [2025-08-21 19:42:10,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4123.8). Total num frames: 3297280. Throughput: 0: 1032.8. Samples: 823800. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:42:10,082][02158] Avg episode reward: [(0, '18.265')] [2025-08-21 19:42:13,906][02335] Updated weights for policy 0, policy_version 810 (0.0014) [2025-08-21 19:42:15,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 3321856. Throughput: 0: 1033.6. Samples: 830968. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:42:15,082][02158] Avg episode reward: [(0, '19.944')] [2025-08-21 19:42:20,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4123.8). Total num frames: 3338240. Throughput: 0: 1030.3. Samples: 834222. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:42:20,084][02158] Avg episode reward: [(0, '20.334')] [2025-08-21 19:42:24,463][02335] Updated weights for policy 0, policy_version 820 (0.0018) [2025-08-21 19:42:25,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 3358720. Throughput: 0: 1031.6. Samples: 839300. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:42:25,081][02158] Avg episode reward: [(0, '22.194')] [2025-08-21 19:42:25,096][02316] Saving new best policy, reward=22.194! [2025-08-21 19:42:30,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 3383296. Throughput: 0: 1030.0. Samples: 846290. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:42:30,081][02158] Avg episode reward: [(0, '23.070')] [2025-08-21 19:42:30,084][02316] Saving new best policy, reward=23.070! [2025-08-21 19:42:34,335][02335] Updated weights for policy 0, policy_version 830 (0.0019) [2025-08-21 19:42:35,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.1, 300 sec: 4109.9). Total num frames: 3399680. Throughput: 0: 1026.6. Samples: 849476. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:42:35,081][02158] Avg episode reward: [(0, '22.532')] [2025-08-21 19:42:40,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 3420160. Throughput: 0: 1028.3. Samples: 854600. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:42:40,081][02158] Avg episode reward: [(0, '21.858')] [2025-08-21 19:42:44,011][02335] Updated weights for policy 0, policy_version 840 (0.0012) [2025-08-21 19:42:45,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 3444736. Throughput: 0: 1031.8. Samples: 861762. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:42:45,084][02158] Avg episode reward: [(0, '20.251')] [2025-08-21 19:42:50,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 3461120. Throughput: 0: 1027.5. Samples: 864978. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:42:50,081][02158] Avg episode reward: [(0, '19.968')] [2025-08-21 19:42:54,464][02335] Updated weights for policy 0, policy_version 850 (0.0024) [2025-08-21 19:42:55,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 3481600. Throughput: 0: 1032.4. Samples: 870256. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:42:55,081][02158] Avg episode reward: [(0, '20.787')] [2025-08-21 19:43:00,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 3506176. Throughput: 0: 1030.8. Samples: 877352. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:43:00,085][02158] Avg episode reward: [(0, '20.669')] [2025-08-21 19:43:03,685][02335] Updated weights for policy 0, policy_version 860 (0.0014) [2025-08-21 19:43:05,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 3526656. Throughput: 0: 1030.4. Samples: 880588. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:43:05,084][02158] Avg episode reward: [(0, '20.597')] [2025-08-21 19:43:10,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 3547136. Throughput: 0: 1039.0. Samples: 886056. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:43:10,085][02158] Avg episode reward: [(0, '21.015')] [2025-08-21 19:43:13,289][02335] Updated weights for policy 0, policy_version 870 (0.0021) [2025-08-21 19:43:15,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 3567616. Throughput: 0: 1041.2. Samples: 893142. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:43:15,085][02158] Avg episode reward: [(0, '19.756')] [2025-08-21 19:43:20,080][02158] Fps is (10 sec: 4095.8, 60 sec: 4164.2, 300 sec: 4123.8). Total num frames: 3588096. Throughput: 0: 1038.0. Samples: 896188. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:43:20,084][02158] Avg episode reward: [(0, '19.390')] [2025-08-21 19:43:23,844][02335] Updated weights for policy 0, policy_version 880 (0.0023) [2025-08-21 19:43:25,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 3608576. Throughput: 0: 1045.0. Samples: 901626. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:43:25,084][02158] Avg episode reward: [(0, '20.778')] [2025-08-21 19:43:30,080][02158] Fps is (10 sec: 4505.8, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 3633152. Throughput: 0: 1042.5. Samples: 908676. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:43:30,081][02158] Avg episode reward: [(0, '22.068')] [2025-08-21 19:43:32,844][02335] Updated weights for policy 0, policy_version 890 (0.0011) [2025-08-21 19:43:35,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 3649536. Throughput: 0: 1039.6. Samples: 911762. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:43:35,085][02158] Avg episode reward: [(0, '22.330')] [2025-08-21 19:43:40,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 3670016. Throughput: 0: 1045.2. Samples: 917290. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:43:40,081][02158] Avg episode reward: [(0, '23.611')] [2025-08-21 19:43:40,082][02316] Saving new best policy, reward=23.611! [2025-08-21 19:43:42,970][02335] Updated weights for policy 0, policy_version 900 (0.0028) [2025-08-21 19:43:45,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 3694592. Throughput: 0: 1046.0. Samples: 924420. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:43:45,085][02158] Avg episode reward: [(0, '22.402')] [2025-08-21 19:43:50,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 3710976. Throughput: 0: 1038.5. Samples: 927320. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:43:50,082][02158] Avg episode reward: [(0, '21.547')] [2025-08-21 19:43:53,451][02335] Updated weights for policy 0, policy_version 910 (0.0026) [2025-08-21 19:43:55,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 3731456. Throughput: 0: 1042.3. Samples: 932960. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:43:55,081][02158] Avg episode reward: [(0, '21.501')] [2025-08-21 19:43:55,106][02316] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000912_3735552.pth... [2025-08-21 19:43:55,218][02316] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000670_2744320.pth [2025-08-21 19:44:00,080][02158] Fps is (10 sec: 4505.5, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 3756032. Throughput: 0: 1042.3. Samples: 940044. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:44:00,082][02158] Avg episode reward: [(0, '19.866')] [2025-08-21 19:44:02,325][02335] Updated weights for policy 0, policy_version 920 (0.0017) [2025-08-21 19:44:05,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4137.7). Total num frames: 3776512. Throughput: 0: 1038.5. Samples: 942922. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:44:05,085][02158] Avg episode reward: [(0, '19.345')] [2025-08-21 19:44:10,081][02158] Fps is (10 sec: 4095.6, 60 sec: 4164.2, 300 sec: 4123.8). Total num frames: 3796992. Throughput: 0: 1045.4. Samples: 948670. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:44:10,082][02158] Avg episode reward: [(0, '20.593')] [2025-08-21 19:44:13,878][02335] Updated weights for policy 0, policy_version 930 (0.0014) [2025-08-21 19:44:15,080][02158] Fps is (10 sec: 3276.8, 60 sec: 4027.7, 300 sec: 4082.1). Total num frames: 3809280. Throughput: 0: 998.9. Samples: 953626. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:44:15,084][02158] Avg episode reward: [(0, '19.446')] [2025-08-21 19:44:20,080][02158] Fps is (10 sec: 2867.5, 60 sec: 3959.5, 300 sec: 4082.1). Total num frames: 3825664. Throughput: 0: 992.2. Samples: 956410. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:44:20,081][02158] Avg episode reward: [(0, '18.481')] [2025-08-21 19:44:24,709][02335] Updated weights for policy 0, policy_version 940 (0.0016) [2025-08-21 19:44:25,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.7, 300 sec: 4096.0). Total num frames: 3850240. Throughput: 0: 996.4. Samples: 962126. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:44:25,088][02158] Avg episode reward: [(0, '18.672')] [2025-08-21 19:44:30,080][02158] Fps is (10 sec: 4915.2, 60 sec: 4027.7, 300 sec: 4096.1). Total num frames: 3874816. Throughput: 0: 999.8. Samples: 969410. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:44:30,085][02158] Avg episode reward: [(0, '18.944')] [2025-08-21 19:44:34,811][02335] Updated weights for policy 0, policy_version 950 (0.0011) [2025-08-21 19:44:35,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.7, 300 sec: 4096.0). Total num frames: 3891200. Throughput: 0: 995.8. Samples: 972132. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) [2025-08-21 19:44:35,081][02158] Avg episode reward: [(0, '17.985')] [2025-08-21 19:44:40,080][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.7, 300 sec: 4096.0). Total num frames: 3911680. Throughput: 0: 1000.2. Samples: 977968. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:44:40,081][02158] Avg episode reward: [(0, '18.378')] [2025-08-21 19:44:43,574][02335] Updated weights for policy 0, policy_version 960 (0.0013) [2025-08-21 19:44:45,080][02158] Fps is (10 sec: 4505.6, 60 sec: 4027.7, 300 sec: 4109.9). Total num frames: 3936256. Throughput: 0: 1002.3. Samples: 985148. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:44:45,085][02158] Avg episode reward: [(0, '21.339')] [2025-08-21 19:44:50,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.7, 300 sec: 4109.9). Total num frames: 3952640. Throughput: 0: 995.8. Samples: 987732. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:44:50,081][02158] Avg episode reward: [(0, '22.020')] [2025-08-21 19:44:54,077][02335] Updated weights for policy 0, policy_version 970 (0.0023) [2025-08-21 19:44:55,080][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 3977216. Throughput: 0: 999.8. Samples: 993660. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:44:55,081][02158] Avg episode reward: [(0, '22.876')] [2025-08-21 19:45:00,084][02158] Fps is (10 sec: 4913.2, 60 sec: 4095.7, 300 sec: 4123.7). Total num frames: 4001792. Throughput: 0: 1049.4. Samples: 1000854. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:45:00,086][02158] Avg episode reward: [(0, '23.877')] [2025-08-21 19:45:00,087][02316] Saving new best policy, reward=23.877! [2025-08-21 19:45:01,315][02316] Stopping Batcher_0... [2025-08-21 19:45:01,315][02316] Loop batcher_evt_loop terminating... [2025-08-21 19:45:01,316][02158] Component Batcher_0 stopped! [2025-08-21 19:45:01,321][02316] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:45:01,406][02335] Weights refcount: 2 0 [2025-08-21 19:45:01,414][02158] Component InferenceWorker_p0-w0 stopped! [2025-08-21 19:45:01,417][02335] Stopping InferenceWorker_p0-w0... [2025-08-21 19:45:01,418][02335] Loop inference_proc0-0_evt_loop terminating... [2025-08-21 19:45:01,501][02316] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000790_3235840.pth [2025-08-21 19:45:01,509][02316] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:45:01,714][02158] Component LearnerWorker_p0 stopped! [2025-08-21 19:45:01,717][02316] Stopping LearnerWorker_p0... [2025-08-21 19:45:01,717][02316] Loop learner_proc0_evt_loop terminating... [2025-08-21 19:45:01,886][02158] Component RolloutWorker_w4 stopped! [2025-08-21 19:45:01,890][02333] Stopping RolloutWorker_w4... [2025-08-21 19:45:01,890][02333] Loop rollout_proc4_evt_loop terminating... [2025-08-21 19:45:01,934][02158] Component RolloutWorker_w0 stopped! [2025-08-21 19:45:01,935][02329] Stopping RolloutWorker_w0... [2025-08-21 19:45:01,947][02329] Loop rollout_proc0_evt_loop terminating... [2025-08-21 19:45:01,998][02158] Component RolloutWorker_w6 stopped! [2025-08-21 19:45:02,003][02336] Stopping RolloutWorker_w6... [2025-08-21 19:45:02,003][02336] Loop rollout_proc6_evt_loop terminating... [2025-08-21 19:45:02,043][02158] Component RolloutWorker_w2 stopped! [2025-08-21 19:45:02,047][02331] Stopping RolloutWorker_w2... [2025-08-21 19:45:02,048][02331] Loop rollout_proc2_evt_loop terminating... [2025-08-21 19:45:02,166][02158] Component RolloutWorker_w3 stopped! [2025-08-21 19:45:02,167][02332] Stopping RolloutWorker_w3... [2025-08-21 19:45:02,168][02332] Loop rollout_proc3_evt_loop terminating... [2025-08-21 19:45:02,179][02158] Component RolloutWorker_w7 stopped! [2025-08-21 19:45:02,180][02337] Stopping RolloutWorker_w7... [2025-08-21 19:45:02,180][02337] Loop rollout_proc7_evt_loop terminating... [2025-08-21 19:45:02,217][02158] Component RolloutWorker_w1 stopped! [2025-08-21 19:45:02,218][02330] Stopping RolloutWorker_w1... [2025-08-21 19:45:02,219][02330] Loop rollout_proc1_evt_loop terminating... [2025-08-21 19:45:02,231][02158] Component RolloutWorker_w5 stopped! [2025-08-21 19:45:02,236][02158] Waiting for process learner_proc0 to stop... [2025-08-21 19:45:02,237][02334] Stopping RolloutWorker_w5... [2025-08-21 19:45:02,238][02334] Loop rollout_proc5_evt_loop terminating... [2025-08-21 19:45:04,062][02158] Waiting for process inference_proc0-0 to join... [2025-08-21 19:45:04,113][02158] Waiting for process rollout_proc0 to join... [2025-08-21 19:45:06,275][02158] Waiting for process rollout_proc1 to join... [2025-08-21 19:45:06,276][02158] Waiting for process rollout_proc2 to join... [2025-08-21 19:45:06,281][02158] Waiting for process rollout_proc3 to join... [2025-08-21 19:45:06,282][02158] Waiting for process rollout_proc4 to join... [2025-08-21 19:45:06,283][02158] Waiting for process rollout_proc5 to join... [2025-08-21 19:45:06,284][02158] Waiting for process rollout_proc6 to join... [2025-08-21 19:45:06,285][02158] Waiting for process rollout_proc7 to join... [2025-08-21 19:45:06,286][02158] Batcher 0 profile tree view: batching: 25.2408, releasing_batches: 0.0231 [2025-08-21 19:45:06,287][02158] InferenceWorker_p0-w0 profile tree view: wait_policy: 0.0009 wait_policy_total: 420.7426 update_model: 7.3242 weight_update: 0.0013 one_step: 0.0024 handle_policy_step: 529.3381 deserialize: 13.8035, stack: 2.8135, obs_to_device_normalize: 112.7857, forward: 268.2359, send_messages: 27.3678 prepare_outputs: 80.8961 to_cpu: 51.1559 [2025-08-21 19:45:06,288][02158] Learner 0 profile tree view: misc: 0.0043, prepare_batch: 12.3363 train: 70.8892 epoch_init: 0.0047, minibatch_init: 0.0062, losses_postprocess: 0.6432, kl_divergence: 0.6006, after_optimizer: 32.4209 calculate_losses: 25.0066 losses_init: 0.0077, forward_head: 1.3103, bptt_initial: 16.4919, tail: 1.0465, advantages_returns: 0.2785, losses: 3.5946 bptt: 2.0570 bptt_forward_core: 1.9683 update: 11.6119 clip: 0.8903 [2025-08-21 19:45:06,289][02158] RolloutWorker_w0 profile tree view: wait_for_trajectories: 0.2611, enqueue_policy_requests: 105.8157, env_step: 781.4104, overhead: 11.6536, complete_rollouts: 7.2810 save_policy_outputs: 18.2975 split_output_tensors: 7.2448 [2025-08-21 19:45:06,290][02158] RolloutWorker_w7 profile tree view: wait_for_trajectories: 0.3584, enqueue_policy_requests: 113.3558, env_step: 769.0173, overhead: 12.9300, complete_rollouts: 6.1323 save_policy_outputs: 18.2378 split_output_tensors: 7.1571 [2025-08-21 19:45:06,292][02158] Loop Runner_EvtLoop terminating... [2025-08-21 19:45:06,293][02158] Runner profile tree view: main_loop: 1025.3570 [2025-08-21 19:45:06,294][02158] Collected {0: 4005888}, FPS: 3906.8 [2025-08-21 19:45:06,583][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 19:45:06,584][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 19:45:06,585][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 19:45:06,587][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 19:45:06,588][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 19:45:06,589][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 19:45:06,590][02158] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 19:45:06,591][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 19:45:06,592][02158] Adding new argument 'push_to_hub'=False that is not in the saved config file! [2025-08-21 19:45:06,593][02158] Adding new argument 'hf_repository'=None that is not in the saved config file! [2025-08-21 19:45:06,594][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 19:45:06,595][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 19:45:06,596][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 19:45:06,597][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 19:45:06,598][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 19:45:06,626][02158] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:45:06,628][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 19:45:06,630][02158] RunningMeanStd input shape: (1,) [2025-08-21 19:45:06,641][02158] ConvEncoder: input_channels=3 [2025-08-21 19:45:06,729][02158] Conv encoder output size: 512 [2025-08-21 19:45:06,730][02158] Policy head output size: 512 [2025-08-21 19:45:06,894][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:45:06,898][02158] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/serialization.py", line 1529, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. (1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. (2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message. WeightsUnpickler error: Unsupported global: GLOBAL numpy.core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([numpy.core.multiarray.scalar])` or the `torch.serialization.safe_globals([numpy.core.multiarray.scalar])` context manager to allowlist this global if you trust this class/function. Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html. [2025-08-21 19:45:06,900][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:45:06,902][02158] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/serialization.py", line 1529, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. (1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. (2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message. WeightsUnpickler error: Unsupported global: GLOBAL numpy.core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([numpy.core.multiarray.scalar])` or the `torch.serialization.safe_globals([numpy.core.multiarray.scalar])` context manager to allowlist this global if you trust this class/function. Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html. [2025-08-21 19:45:06,904][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:45:06,905][02158] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/serialization.py", line 1529, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. (1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. (2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message. WeightsUnpickler error: Unsupported global: GLOBAL numpy.core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([numpy.core.multiarray.scalar])` or the `torch.serialization.safe_globals([numpy.core.multiarray.scalar])` context manager to allowlist this global if you trust this class/function. Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html. [2025-08-21 19:45:29,585][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 19:45:29,586][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 19:45:29,587][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 19:45:29,588][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 19:45:29,589][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 19:45:29,590][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 19:45:29,591][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 19:45:29,592][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 19:45:29,592][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 19:45:29,593][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 19:45:29,594][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 19:45:29,595][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 19:45:29,596][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 19:45:29,597][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 19:45:29,599][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 19:45:29,628][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 19:45:29,629][02158] RunningMeanStd input shape: (1,) [2025-08-21 19:45:29,638][02158] ConvEncoder: input_channels=3 [2025-08-21 19:45:29,683][02158] Conv encoder output size: 512 [2025-08-21 19:45:29,684][02158] Policy head output size: 512 [2025-08-21 19:45:29,700][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:45:29,701][02158] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/serialization.py", line 1529, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. (1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. (2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message. WeightsUnpickler error: Unsupported global: GLOBAL numpy.core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([numpy.core.multiarray.scalar])` or the `torch.serialization.safe_globals([numpy.core.multiarray.scalar])` context manager to allowlist this global if you trust this class/function. Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html. [2025-08-21 19:45:29,702][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:45:29,704][02158] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/serialization.py", line 1529, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. (1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. (2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message. WeightsUnpickler error: Unsupported global: GLOBAL numpy.core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([numpy.core.multiarray.scalar])` or the `torch.serialization.safe_globals([numpy.core.multiarray.scalar])` context manager to allowlist this global if you trust this class/function. Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html. [2025-08-21 19:45:29,705][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:45:29,707][02158] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/serialization.py", line 1529, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. (1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. (2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message. WeightsUnpickler error: Unsupported global: GLOBAL numpy.core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([numpy.core.multiarray.scalar])` or the `torch.serialization.safe_globals([numpy.core.multiarray.scalar])` context manager to allowlist this global if you trust this class/function. Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html. [2025-08-21 19:47:35,631][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 19:47:35,632][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 19:47:35,633][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 19:47:35,634][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 19:47:35,636][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 19:47:35,637][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 19:47:35,638][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 19:47:35,639][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 19:47:35,640][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 19:47:35,641][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 19:47:35,642][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 19:47:35,643][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 19:47:35,644][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 19:47:35,645][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 19:47:35,646][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 19:47:35,668][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 19:47:35,669][02158] RunningMeanStd input shape: (1,) [2025-08-21 19:47:35,677][02158] ConvEncoder: input_channels=3 [2025-08-21 19:47:35,709][02158] Conv encoder output size: 512 [2025-08-21 19:47:35,710][02158] Policy head output size: 512 [2025-08-21 19:47:35,726][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:47:35,728][02158] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/serialization.py", line 1529, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. Please file an issue with the following so that we can make `weights_only=True` compatible with your use case: WeightsUnpickler error: Can only build Tensor, Parameter, OrderedDict or types allowlisted via `add_safe_globals`, but got Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html. [2025-08-21 19:47:35,729][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:47:35,730][02158] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/serialization.py", line 1529, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. Please file an issue with the following so that we can make `weights_only=True` compatible with your use case: WeightsUnpickler error: Can only build Tensor, Parameter, OrderedDict or types allowlisted via `add_safe_globals`, but got Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html. [2025-08-21 19:47:35,731][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:47:35,733][02158] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/serialization.py", line 1529, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. Please file an issue with the following so that we can make `weights_only=True` compatible with your use case: WeightsUnpickler error: Can only build Tensor, Parameter, OrderedDict or types allowlisted via `add_safe_globals`, but got Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html. [2025-08-21 19:48:09,236][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 19:48:09,237][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 19:48:09,238][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 19:48:09,239][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 19:48:09,240][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 19:48:09,240][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 19:48:09,241][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 19:48:09,242][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 19:48:09,243][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 19:48:09,243][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 19:48:09,244][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 19:48:09,246][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 19:48:09,247][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 19:48:09,248][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 19:48:09,249][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 19:48:09,276][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 19:48:09,279][02158] RunningMeanStd input shape: (1,) [2025-08-21 19:48:09,287][02158] ConvEncoder: input_channels=3 [2025-08-21 19:48:09,318][02158] Conv encoder output size: 512 [2025-08-21 19:48:09,319][02158] Policy head output size: 512 [2025-08-21 19:48:09,338][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:48:10,061][02158] Num frames 100... [2025-08-21 19:48:10,185][02158] Num frames 200... [2025-08-21 19:48:10,309][02158] Num frames 300... [2025-08-21 19:48:10,432][02158] Num frames 400... [2025-08-21 19:48:10,564][02158] Num frames 500... [2025-08-21 19:48:10,692][02158] Num frames 600... [2025-08-21 19:48:10,816][02158] Num frames 700... [2025-08-21 19:48:10,940][02158] Num frames 800... [2025-08-21 19:48:11,062][02158] Num frames 900... [2025-08-21 19:48:11,196][02158] Avg episode rewards: #0: 22.600, true rewards: #0: 9.600 [2025-08-21 19:48:11,197][02158] Avg episode reward: 22.600, avg true_objective: 9.600 [2025-08-21 19:48:11,248][02158] Num frames 1000... [2025-08-21 19:48:11,371][02158] Num frames 1100... [2025-08-21 19:48:11,492][02158] Num frames 1200... [2025-08-21 19:48:11,622][02158] Num frames 1300... [2025-08-21 19:48:11,747][02158] Num frames 1400... [2025-08-21 19:48:11,866][02158] Num frames 1500... [2025-08-21 19:48:11,989][02158] Num frames 1600... [2025-08-21 19:48:12,110][02158] Num frames 1700... [2025-08-21 19:48:12,239][02158] Num frames 1800... [2025-08-21 19:48:12,403][02158] Avg episode rewards: #0: 20.940, true rewards: #0: 9.440 [2025-08-21 19:48:12,404][02158] Avg episode reward: 20.940, avg true_objective: 9.440 [2025-08-21 19:48:12,423][02158] Num frames 1900... [2025-08-21 19:48:12,540][02158] Num frames 2000... [2025-08-21 19:48:12,670][02158] Num frames 2100... [2025-08-21 19:48:12,788][02158] Num frames 2200... [2025-08-21 19:48:12,905][02158] Num frames 2300... [2025-08-21 19:48:13,028][02158] Num frames 2400... [2025-08-21 19:48:13,150][02158] Num frames 2500... [2025-08-21 19:48:13,273][02158] Num frames 2600... [2025-08-21 19:48:13,396][02158] Num frames 2700... [2025-08-21 19:48:13,518][02158] Avg episode rewards: #0: 19.507, true rewards: #0: 9.173 [2025-08-21 19:48:13,519][02158] Avg episode reward: 19.507, avg true_objective: 9.173 [2025-08-21 19:48:13,581][02158] Num frames 2800... [2025-08-21 19:48:13,713][02158] Num frames 2900... [2025-08-21 19:48:13,832][02158] Num frames 3000... [2025-08-21 19:48:13,958][02158] Num frames 3100... [2025-08-21 19:48:14,078][02158] Num frames 3200... [2025-08-21 19:48:14,202][02158] Num frames 3300... [2025-08-21 19:48:14,323][02158] Num frames 3400... [2025-08-21 19:48:14,444][02158] Num frames 3500... [2025-08-21 19:48:14,564][02158] Num frames 3600... [2025-08-21 19:48:14,694][02158] Num frames 3700... [2025-08-21 19:48:14,812][02158] Num frames 3800... [2025-08-21 19:48:14,931][02158] Num frames 3900... [2025-08-21 19:48:15,048][02158] Num frames 4000... [2025-08-21 19:48:15,166][02158] Num frames 4100... [2025-08-21 19:48:15,263][02158] Avg episode rewards: #0: 22.335, true rewards: #0: 10.335 [2025-08-21 19:48:15,264][02158] Avg episode reward: 22.335, avg true_objective: 10.335 [2025-08-21 19:48:15,346][02158] Num frames 4200... [2025-08-21 19:48:15,466][02158] Num frames 4300... [2025-08-21 19:48:15,590][02158] Num frames 4400... [2025-08-21 19:48:15,724][02158] Num frames 4500... [2025-08-21 19:48:15,846][02158] Num frames 4600... [2025-08-21 19:48:15,967][02158] Num frames 4700... [2025-08-21 19:48:16,087][02158] Num frames 4800... [2025-08-21 19:48:16,211][02158] Num frames 4900... [2025-08-21 19:48:16,341][02158] Num frames 5000... [2025-08-21 19:48:16,468][02158] Num frames 5100... [2025-08-21 19:48:16,589][02158] Num frames 5200... [2025-08-21 19:48:16,712][02158] Num frames 5300... [2025-08-21 19:48:16,844][02158] Num frames 5400... [2025-08-21 19:48:16,975][02158] Num frames 5500... [2025-08-21 19:48:17,110][02158] Num frames 5600... [2025-08-21 19:48:17,234][02158] Num frames 5700... [2025-08-21 19:48:17,390][02158] Avg episode rewards: #0: 25.962, true rewards: #0: 11.562 [2025-08-21 19:48:17,391][02158] Avg episode reward: 25.962, avg true_objective: 11.562 [2025-08-21 19:48:17,417][02158] Num frames 5800... [2025-08-21 19:48:17,537][02158] Num frames 5900... [2025-08-21 19:48:17,667][02158] Num frames 6000... [2025-08-21 19:48:17,808][02158] Num frames 6100... [2025-08-21 19:48:17,927][02158] Num frames 6200... [2025-08-21 19:48:18,046][02158] Num frames 6300... [2025-08-21 19:48:18,166][02158] Num frames 6400... [2025-08-21 19:48:18,288][02158] Num frames 6500... [2025-08-21 19:48:18,414][02158] Num frames 6600... [2025-08-21 19:48:18,572][02158] Num frames 6700... [2025-08-21 19:48:18,750][02158] Num frames 6800... [2025-08-21 19:48:18,929][02158] Num frames 6900... [2025-08-21 19:48:18,995][02158] Avg episode rewards: #0: 25.502, true rewards: #0: 11.502 [2025-08-21 19:48:18,996][02158] Avg episode reward: 25.502, avg true_objective: 11.502 [2025-08-21 19:48:19,188][02158] Num frames 7000... [2025-08-21 19:48:19,405][02158] Num frames 7100... [2025-08-21 19:48:19,576][02158] Num frames 7200... [2025-08-21 19:48:19,751][02158] Num frames 7300... [2025-08-21 19:48:19,947][02158] Num frames 7400... [2025-08-21 19:48:20,126][02158] Num frames 7500... [2025-08-21 19:48:20,203][02158] Avg episode rewards: #0: 23.299, true rewards: #0: 10.727 [2025-08-21 19:48:20,204][02158] Avg episode reward: 23.299, avg true_objective: 10.727 [2025-08-21 19:48:20,366][02158] Num frames 7600... [2025-08-21 19:48:20,547][02158] Num frames 7700... [2025-08-21 19:48:20,612][02158] Avg episode rewards: #0: 20.876, true rewards: #0: 9.626 [2025-08-21 19:48:20,613][02158] Avg episode reward: 20.876, avg true_objective: 9.626 [2025-08-21 19:48:20,775][02158] Num frames 7800... [2025-08-21 19:48:20,898][02158] Num frames 7900... [2025-08-21 19:48:21,029][02158] Num frames 8000... [2025-08-21 19:48:21,153][02158] Num frames 8100... [2025-08-21 19:48:21,294][02158] Num frames 8200... [2025-08-21 19:48:21,457][02158] Avg episode rewards: #0: 19.641, true rewards: #0: 9.197 [2025-08-21 19:48:21,458][02158] Avg episode reward: 19.641, avg true_objective: 9.197 [2025-08-21 19:48:21,489][02158] Num frames 8300... [2025-08-21 19:48:21,616][02158] Num frames 8400... [2025-08-21 19:48:21,745][02158] Num frames 8500... [2025-08-21 19:48:21,872][02158] Num frames 8600... [2025-08-21 19:48:22,022][02158] Avg episode rewards: #0: 18.061, true rewards: #0: 8.661 [2025-08-21 19:48:22,023][02158] Avg episode reward: 18.061, avg true_objective: 8.661 [2025-08-21 19:49:14,491][02158] Replay video saved to /content/train_dir/default_experiment/replay.mp4! [2025-08-21 19:49:20,462][02158] The model has been pushed to https://huggingface.co/jmartin233/rl_course_vizdoom_health_gathering_supreme [2025-08-21 19:50:28,916][02158] Environment doom_basic already registered, overwriting... [2025-08-21 19:50:28,917][02158] Environment doom_two_colors_easy already registered, overwriting... [2025-08-21 19:50:28,918][02158] Environment doom_two_colors_hard already registered, overwriting... [2025-08-21 19:50:28,919][02158] Environment doom_dm already registered, overwriting... [2025-08-21 19:50:28,920][02158] Environment doom_dwango5 already registered, overwriting... [2025-08-21 19:50:28,921][02158] Environment doom_my_way_home_flat_actions already registered, overwriting... [2025-08-21 19:50:28,922][02158] Environment doom_defend_the_center_flat_actions already registered, overwriting... [2025-08-21 19:50:28,924][02158] Environment doom_my_way_home already registered, overwriting... [2025-08-21 19:50:28,924][02158] Environment doom_deadly_corridor already registered, overwriting... [2025-08-21 19:50:28,926][02158] Environment doom_defend_the_center already registered, overwriting... [2025-08-21 19:50:28,926][02158] Environment doom_defend_the_line already registered, overwriting... [2025-08-21 19:50:28,927][02158] Environment doom_health_gathering already registered, overwriting... [2025-08-21 19:50:28,927][02158] Environment doom_health_gathering_supreme already registered, overwriting... [2025-08-21 19:50:28,928][02158] Environment doom_battle already registered, overwriting... [2025-08-21 19:50:28,929][02158] Environment doom_battle2 already registered, overwriting... [2025-08-21 19:50:28,930][02158] Environment doom_duel_bots already registered, overwriting... [2025-08-21 19:50:28,930][02158] Environment doom_deathmatch_bots already registered, overwriting... [2025-08-21 19:50:28,931][02158] Environment doom_duel already registered, overwriting... [2025-08-21 19:50:28,932][02158] Environment doom_deathmatch_full already registered, overwriting... [2025-08-21 19:50:28,932][02158] Environment doom_benchmark already registered, overwriting... [2025-08-21 19:50:28,933][02158] register_encoder_factory: [2025-08-21 19:50:28,942][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 19:50:28,943][02158] Overriding arg 'train_for_env_steps' with value 5000000 passed from command line [2025-08-21 19:50:28,948][02158] Experiment dir /content/train_dir/default_experiment already exists! [2025-08-21 19:50:28,948][02158] Resuming existing experiment from /content/train_dir/default_experiment... [2025-08-21 19:50:28,949][02158] Weights and Biases integration disabled [2025-08-21 19:50:28,952][02158] Environment var CUDA_VISIBLE_DEVICES is 0 [2025-08-21 19:50:31,047][02158] 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=5000000 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 [2025-08-21 19:50:31,048][02158] Saving configuration to /content/train_dir/default_experiment/config.json... [2025-08-21 19:50:31,050][02158] Rollout worker 0 uses device cpu [2025-08-21 19:50:31,051][02158] Rollout worker 1 uses device cpu [2025-08-21 19:50:31,052][02158] Rollout worker 2 uses device cpu [2025-08-21 19:50:31,053][02158] Rollout worker 3 uses device cpu [2025-08-21 19:50:31,055][02158] Rollout worker 4 uses device cpu [2025-08-21 19:50:31,056][02158] Rollout worker 5 uses device cpu [2025-08-21 19:50:31,057][02158] Rollout worker 6 uses device cpu [2025-08-21 19:50:31,058][02158] Rollout worker 7 uses device cpu [2025-08-21 19:50:31,125][02158] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-08-21 19:50:31,126][02158] InferenceWorker_p0-w0: min num requests: 2 [2025-08-21 19:50:31,155][02158] Starting all processes... [2025-08-21 19:50:31,155][02158] Starting process learner_proc0 [2025-08-21 19:50:31,217][02158] Starting all processes... [2025-08-21 19:50:31,222][02158] Starting process inference_proc0-0 [2025-08-21 19:50:31,222][02158] Starting process rollout_proc0 [2025-08-21 19:50:31,224][02158] Starting process rollout_proc1 [2025-08-21 19:50:31,224][02158] Starting process rollout_proc2 [2025-08-21 19:50:31,224][02158] Starting process rollout_proc3 [2025-08-21 19:50:31,224][02158] Starting process rollout_proc4 [2025-08-21 19:50:31,224][02158] Starting process rollout_proc5 [2025-08-21 19:50:31,224][02158] Starting process rollout_proc6 [2025-08-21 19:50:31,224][02158] Starting process rollout_proc7 [2025-08-21 19:50:46,821][11373] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-08-21 19:50:46,821][11373] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0 [2025-08-21 19:50:46,954][11373] Num visible devices: 1 [2025-08-21 19:50:46,982][11373] Starting seed is not provided [2025-08-21 19:50:46,982][11373] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-08-21 19:50:46,982][11373] Initializing actor-critic model on device cuda:0 [2025-08-21 19:50:46,982][11373] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 19:50:46,987][11373] RunningMeanStd input shape: (1,) [2025-08-21 19:50:47,106][11373] ConvEncoder: input_channels=3 [2025-08-21 19:50:47,468][11386] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-08-21 19:50:47,470][11386] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0 [2025-08-21 19:50:47,624][11386] Num visible devices: 1 [2025-08-21 19:50:47,658][11394] Worker 4 uses CPU cores [0] [2025-08-21 19:50:47,686][11393] Worker 6 uses CPU cores [0] [2025-08-21 19:50:47,721][11391] Worker 5 uses CPU cores [1] [2025-08-21 19:50:47,812][11387] Worker 0 uses CPU cores [0] [2025-08-21 19:50:47,831][11392] Worker 7 uses CPU cores [1] [2025-08-21 19:50:47,837][11389] Worker 2 uses CPU cores [0] [2025-08-21 19:50:47,906][11388] Worker 1 uses CPU cores [1] [2025-08-21 19:50:47,941][11390] Worker 3 uses CPU cores [1] [2025-08-21 19:50:47,950][11373] Conv encoder output size: 512 [2025-08-21 19:50:47,950][11373] Policy head output size: 512 [2025-08-21 19:50:47,972][11373] Created Actor Critic model with architecture: [2025-08-21 19:50:47,972][11373] 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) ) ) [2025-08-21 19:50:48,129][11373] Using optimizer [2025-08-21 19:50:49,134][11373] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:50:49,137][11373] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/serialization.py", line 1529, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. (1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. (2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message. WeightsUnpickler error: Unsupported global: GLOBAL numpy.core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([numpy.core.multiarray.scalar])` or the `torch.serialization.safe_globals([numpy.core.multiarray.scalar])` context manager to allowlist this global if you trust this class/function. Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html. [2025-08-21 19:50:49,138][11373] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:50:49,139][11373] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/serialization.py", line 1529, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. (1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. (2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message. WeightsUnpickler error: Unsupported global: GLOBAL numpy.core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([numpy.core.multiarray.scalar])` or the `torch.serialization.safe_globals([numpy.core.multiarray.scalar])` context manager to allowlist this global if you trust this class/function. Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html. [2025-08-21 19:50:49,139][11373] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-08-21 19:50:49,140][11373] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.12/dist-packages/torch/serialization.py", line 1529, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint. (1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. (2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message. WeightsUnpickler error: Unsupported global: GLOBAL numpy.core.multiarray.scalar was not an allowed global by default. Please use `torch.serialization.add_safe_globals([numpy.core.multiarray.scalar])` or the `torch.serialization.safe_globals([numpy.core.multiarray.scalar])` context manager to allowlist this global if you trust this class/function. Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html. [2025-08-21 19:50:49,141][11373] Did not load from checkpoint, starting from scratch! [2025-08-21 19:50:49,141][11373] Initialized policy 0 weights for model version 0 [2025-08-21 19:50:49,144][11373] LearnerWorker_p0 finished initialization! [2025-08-21 19:50:49,145][11373] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-08-21 19:50:49,272][11386] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 19:50:49,273][11386] RunningMeanStd input shape: (1,) [2025-08-21 19:50:49,283][11386] ConvEncoder: input_channels=3 [2025-08-21 19:50:49,378][11386] Conv encoder output size: 512 [2025-08-21 19:50:49,379][11386] Policy head output size: 512 [2025-08-21 19:50:49,422][02158] Inference worker 0-0 is ready! [2025-08-21 19:50:49,423][02158] All inference workers are ready! Signal rollout workers to start! [2025-08-21 19:50:49,647][11388] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:50:49,656][11389] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:50:49,669][11392] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:50:49,682][11393] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:50:49,723][11391] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:50:49,725][11390] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:50:49,730][11394] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:50:49,742][11387] Doom resolution: 160x120, resize resolution: (128, 72) [2025-08-21 19:50:50,979][11388] Decorrelating experience for 0 frames... [2025-08-21 19:50:50,981][11392] Decorrelating experience for 0 frames... [2025-08-21 19:50:50,993][11389] Decorrelating experience for 0 frames... [2025-08-21 19:50:51,005][11391] Decorrelating experience for 0 frames... [2025-08-21 19:50:51,017][11394] Decorrelating experience for 0 frames... [2025-08-21 19:50:51,020][11387] Decorrelating experience for 0 frames... [2025-08-21 19:50:51,118][02158] Heartbeat connected on Batcher_0 [2025-08-21 19:50:51,122][02158] Heartbeat connected on LearnerWorker_p0 [2025-08-21 19:50:51,161][02158] Heartbeat connected on InferenceWorker_p0-w0 [2025-08-21 19:50:52,227][11388] Decorrelating experience for 32 frames... [2025-08-21 19:50:52,306][11387] Decorrelating experience for 32 frames... [2025-08-21 19:50:52,343][11391] Decorrelating experience for 32 frames... [2025-08-21 19:50:52,345][11394] Decorrelating experience for 32 frames... [2025-08-21 19:50:52,372][11390] Decorrelating experience for 0 frames... [2025-08-21 19:50:53,394][11389] Decorrelating experience for 32 frames... [2025-08-21 19:50:53,607][11387] Decorrelating experience for 64 frames... [2025-08-21 19:50:53,616][11392] Decorrelating experience for 32 frames... [2025-08-21 19:50:53,686][11390] Decorrelating experience for 32 frames... [2025-08-21 19:50:53,952][02158] 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) [2025-08-21 19:50:54,069][11391] Decorrelating experience for 64 frames... [2025-08-21 19:50:54,587][11394] Decorrelating experience for 64 frames... [2025-08-21 19:50:54,606][11388] Decorrelating experience for 64 frames... [2025-08-21 19:50:54,741][11393] Decorrelating experience for 0 frames... [2025-08-21 19:50:54,879][11392] Decorrelating experience for 64 frames... [2025-08-21 19:50:54,901][11387] Decorrelating experience for 96 frames... [2025-08-21 19:50:55,054][02158] Heartbeat connected on RolloutWorker_w0 [2025-08-21 19:50:55,750][11393] Decorrelating experience for 32 frames... [2025-08-21 19:50:56,169][11388] Decorrelating experience for 96 frames... [2025-08-21 19:50:56,404][02158] Heartbeat connected on RolloutWorker_w1 [2025-08-21 19:50:56,505][11390] Decorrelating experience for 64 frames... [2025-08-21 19:50:56,606][11392] Decorrelating experience for 96 frames... [2025-08-21 19:50:56,894][02158] Heartbeat connected on RolloutWorker_w7 [2025-08-21 19:50:57,660][11391] Decorrelating experience for 96 frames... [2025-08-21 19:50:57,728][11393] Decorrelating experience for 64 frames... [2025-08-21 19:50:58,062][02158] Heartbeat connected on RolloutWorker_w5 [2025-08-21 19:50:58,952][02158] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 262.4. Samples: 1312. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) [2025-08-21 19:50:58,953][02158] Avg episode reward: [(0, '3.213')] [2025-08-21 19:50:59,945][11390] Decorrelating experience for 96 frames... [2025-08-21 19:50:59,971][11389] Decorrelating experience for 64 frames... [2025-08-21 19:51:00,891][02158] Heartbeat connected on RolloutWorker_w3 [2025-08-21 19:51:02,749][11394] Decorrelating experience for 96 frames... [2025-08-21 19:51:02,991][02158] Heartbeat connected on RolloutWorker_w4 [2025-08-21 19:51:03,952][02158] Fps is (10 sec: 1228.8, 60 sec: 1228.8, 300 sec: 1228.8). Total num frames: 12288. Throughput: 0: 203.8. Samples: 2038. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0) [2025-08-21 19:51:03,956][02158] Avg episode reward: [(0, '3.349')] [2025-08-21 19:51:04,106][11393] Decorrelating experience for 96 frames... [2025-08-21 19:51:04,434][11389] Decorrelating experience for 96 frames... [2025-08-21 19:51:04,597][02158] Heartbeat connected on RolloutWorker_w6 [2025-08-21 19:51:04,770][02158] Heartbeat connected on RolloutWorker_w2 [2025-08-21 19:51:08,952][02158] Fps is (10 sec: 3276.8, 60 sec: 2184.6, 300 sec: 2184.6). Total num frames: 32768. Throughput: 0: 507.1. Samples: 7606. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:51:08,953][02158] Avg episode reward: [(0, '3.774')] [2025-08-21 19:51:10,335][11386] Updated weights for policy 0, policy_version 10 (0.0012) [2025-08-21 19:51:13,952][02158] Fps is (10 sec: 4096.0, 60 sec: 2662.4, 300 sec: 2662.4). Total num frames: 53248. Throughput: 0: 716.7. Samples: 14334. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:51:13,956][02158] Avg episode reward: [(0, '4.148')] [2025-08-21 19:51:18,952][02158] Fps is (10 sec: 4096.0, 60 sec: 2949.1, 300 sec: 2949.1). Total num frames: 73728. Throughput: 0: 658.3. Samples: 16458. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:51:18,953][02158] Avg episode reward: [(0, '4.275')] [2025-08-21 19:51:20,719][11386] Updated weights for policy 0, policy_version 20 (0.0044) [2025-08-21 19:51:23,952][02158] Fps is (10 sec: 4096.0, 60 sec: 3140.3, 300 sec: 3140.3). Total num frames: 94208. Throughput: 0: 765.6. Samples: 22968. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:51:23,953][02158] Avg episode reward: [(0, '4.449')] [2025-08-21 19:51:28,956][02158] Fps is (10 sec: 4094.3, 60 sec: 3276.4, 300 sec: 3276.4). Total num frames: 114688. Throughput: 0: 831.4. Samples: 29104. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:51:28,962][02158] Avg episode reward: [(0, '4.313')] [2025-08-21 19:51:28,967][11373] Saving new best policy, reward=4.313! [2025-08-21 19:51:31,646][11386] Updated weights for policy 0, policy_version 30 (0.0020) [2025-08-21 19:51:33,952][02158] Fps is (10 sec: 3686.4, 60 sec: 3276.8, 300 sec: 3276.8). Total num frames: 131072. Throughput: 0: 780.4. Samples: 31216. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:51:33,957][02158] Avg episode reward: [(0, '4.347')] [2025-08-21 19:51:33,960][11373] Saving new best policy, reward=4.347! [2025-08-21 19:51:38,952][02158] Fps is (10 sec: 4097.7, 60 sec: 3458.9, 300 sec: 3458.9). Total num frames: 155648. Throughput: 0: 843.1. Samples: 37938. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:51:38,956][02158] Avg episode reward: [(0, '4.473')] [2025-08-21 19:51:38,962][11373] Saving new best policy, reward=4.473! [2025-08-21 19:51:40,505][11386] Updated weights for policy 0, policy_version 40 (0.0018) [2025-08-21 19:51:43,954][02158] Fps is (10 sec: 4504.8, 60 sec: 3522.4, 300 sec: 3522.4). Total num frames: 176128. Throughput: 0: 958.7. Samples: 44454. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:51:43,955][02158] Avg episode reward: [(0, '4.363')] [2025-08-21 19:51:48,952][02158] Fps is (10 sec: 3686.4, 60 sec: 3500.2, 300 sec: 3500.2). Total num frames: 192512. Throughput: 0: 990.9. Samples: 46628. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:51:48,954][02158] Avg episode reward: [(0, '4.401')] [2025-08-21 19:51:51,016][11386] Updated weights for policy 0, policy_version 50 (0.0021) [2025-08-21 19:51:53,952][02158] Fps is (10 sec: 4096.8, 60 sec: 3618.1, 300 sec: 3618.1). Total num frames: 217088. Throughput: 0: 1024.1. Samples: 53690. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:51:53,956][02158] Avg episode reward: [(0, '4.474')] [2025-08-21 19:51:53,959][11373] Saving new best policy, reward=4.474! [2025-08-21 19:51:58,953][02158] Fps is (10 sec: 4505.1, 60 sec: 3959.4, 300 sec: 3654.8). Total num frames: 237568. Throughput: 0: 1010.5. Samples: 59806. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:51:58,954][02158] Avg episode reward: [(0, '4.491')] [2025-08-21 19:51:58,958][11373] Saving new best policy, reward=4.491! [2025-08-21 19:52:01,362][11386] Updated weights for policy 0, policy_version 60 (0.0016) [2025-08-21 19:52:03,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.7, 300 sec: 3627.9). Total num frames: 253952. Throughput: 0: 1009.4. Samples: 61882. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) [2025-08-21 19:52:03,954][02158] Avg episode reward: [(0, '4.529')] [2025-08-21 19:52:03,955][11373] Saving new best policy, reward=4.529! [2025-08-21 19:52:08,952][02158] Fps is (10 sec: 4096.4, 60 sec: 4096.0, 300 sec: 3713.7). Total num frames: 278528. Throughput: 0: 1022.8. Samples: 68992. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:52:08,953][02158] Avg episode reward: [(0, '4.425')] [2025-08-21 19:52:10,083][11386] Updated weights for policy 0, policy_version 70 (0.0023) [2025-08-21 19:52:13,952][02158] Fps is (10 sec: 4505.7, 60 sec: 4096.0, 300 sec: 3737.6). Total num frames: 299008. Throughput: 0: 1027.8. Samples: 75350. Policy #0 lag: (min: 0.0, avg: 0.3, max: 2.0) [2025-08-21 19:52:13,953][02158] Avg episode reward: [(0, '4.430')] [2025-08-21 19:52:18,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 3758.7). Total num frames: 319488. Throughput: 0: 1031.0. Samples: 77612. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:52:18,956][02158] Avg episode reward: [(0, '4.325')] [2025-08-21 19:52:20,621][11386] Updated weights for policy 0, policy_version 80 (0.0018) [2025-08-21 19:52:23,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 3777.4). Total num frames: 339968. Throughput: 0: 1039.8. Samples: 84730. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:52:23,956][02158] Avg episode reward: [(0, '4.423')] [2025-08-21 19:52:28,953][02158] Fps is (10 sec: 4095.6, 60 sec: 4096.2, 300 sec: 3794.2). Total num frames: 360448. Throughput: 0: 1030.5. Samples: 90826. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:52:28,955][02158] Avg episode reward: [(0, '4.404')] [2025-08-21 19:52:28,965][11373] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000088_360448.pth... [2025-08-21 19:52:29,121][11373] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000088_360448.pth [2025-08-21 19:52:31,090][11386] Updated weights for policy 0, policy_version 90 (0.0032) [2025-08-21 19:52:33,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 3809.3). Total num frames: 380928. Throughput: 0: 1032.2. Samples: 93076. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:52:33,955][02158] Avg episode reward: [(0, '4.345')] [2025-08-21 19:52:38,952][02158] Fps is (10 sec: 4506.0, 60 sec: 4164.3, 300 sec: 3862.0). Total num frames: 405504. Throughput: 0: 1033.7. Samples: 100208. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:52:38,956][02158] Avg episode reward: [(0, '4.407')] [2025-08-21 19:52:39,783][11386] Updated weights for policy 0, policy_version 100 (0.0013) [2025-08-21 19:52:43,953][02158] Fps is (10 sec: 4095.7, 60 sec: 4096.1, 300 sec: 3835.3). Total num frames: 421888. Throughput: 0: 1031.8. Samples: 106236. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:52:43,954][02158] Avg episode reward: [(0, '4.560')] [2025-08-21 19:52:43,960][11373] Saving new best policy, reward=4.560! [2025-08-21 19:52:48,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 3846.7). Total num frames: 442368. Throughput: 0: 1036.0. Samples: 108504. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:52:48,953][02158] Avg episode reward: [(0, '4.643')] [2025-08-21 19:52:48,958][11373] Saving new best policy, reward=4.643! [2025-08-21 19:52:51,011][11386] Updated weights for policy 0, policy_version 110 (0.0022) [2025-08-21 19:52:53,952][02158] Fps is (10 sec: 3686.7, 60 sec: 4027.7, 300 sec: 3822.9). Total num frames: 458752. Throughput: 0: 1003.1. Samples: 114130. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:52:53,957][02158] Avg episode reward: [(0, '4.656')] [2025-08-21 19:52:53,960][11373] Saving new best policy, reward=4.656! [2025-08-21 19:52:58,952][02158] Fps is (10 sec: 2867.2, 60 sec: 3891.3, 300 sec: 3768.3). Total num frames: 471040. Throughput: 0: 965.4. Samples: 118794. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:52:58,953][02158] Avg episode reward: [(0, '4.468')] [2025-08-21 19:53:03,544][11386] Updated weights for policy 0, policy_version 120 (0.0029) [2025-08-21 19:53:03,952][02158] Fps is (10 sec: 3276.8, 60 sec: 3959.5, 300 sec: 3780.9). Total num frames: 491520. Throughput: 0: 970.0. Samples: 121260. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:53:03,953][02158] Avg episode reward: [(0, '4.417')] [2025-08-21 19:53:08,952][02158] Fps is (10 sec: 4096.0, 60 sec: 3891.2, 300 sec: 3792.6). Total num frames: 512000. Throughput: 0: 959.7. Samples: 127916. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:53:08,956][02158] Avg episode reward: [(0, '4.505')] [2025-08-21 19:53:13,673][11386] Updated weights for policy 0, policy_version 130 (0.0020) [2025-08-21 19:53:13,952][02158] Fps is (10 sec: 4096.0, 60 sec: 3891.2, 300 sec: 3803.4). Total num frames: 532480. Throughput: 0: 951.4. Samples: 133640. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:53:13,954][02158] Avg episode reward: [(0, '4.564')] [2025-08-21 19:53:18,952][02158] Fps is (10 sec: 4096.0, 60 sec: 3891.2, 300 sec: 3813.5). Total num frames: 552960. Throughput: 0: 957.3. Samples: 136154. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:53:18,957][02158] Avg episode reward: [(0, '4.709')] [2025-08-21 19:53:18,964][11373] Saving new best policy, reward=4.709! [2025-08-21 19:53:23,114][11386] Updated weights for policy 0, policy_version 140 (0.0013) [2025-08-21 19:53:23,952][02158] Fps is (10 sec: 4096.0, 60 sec: 3891.2, 300 sec: 3822.9). Total num frames: 573440. Throughput: 0: 953.2. Samples: 143100. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:53:23,953][02158] Avg episode reward: [(0, '4.582')] [2025-08-21 19:53:28,952][02158] Fps is (10 sec: 3686.4, 60 sec: 3823.0, 300 sec: 3805.3). Total num frames: 589824. Throughput: 0: 942.7. Samples: 148658. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:53:28,953][02158] Avg episode reward: [(0, '4.401')] [2025-08-21 19:53:33,764][11386] Updated weights for policy 0, policy_version 150 (0.0026) [2025-08-21 19:53:33,952][02158] Fps is (10 sec: 4096.0, 60 sec: 3891.2, 300 sec: 3840.0). Total num frames: 614400. Throughput: 0: 950.8. Samples: 151292. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:53:33,953][02158] Avg episode reward: [(0, '4.551')] [2025-08-21 19:53:38,952][02158] Fps is (10 sec: 4505.6, 60 sec: 3822.9, 300 sec: 3847.8). Total num frames: 634880. Throughput: 0: 982.9. Samples: 158360. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:53:38,957][02158] Avg episode reward: [(0, '4.554')] [2025-08-21 19:53:43,938][11386] Updated weights for policy 0, policy_version 160 (0.0012) [2025-08-21 19:53:43,954][02158] Fps is (10 sec: 4095.1, 60 sec: 3891.1, 300 sec: 3855.0). Total num frames: 655360. Throughput: 0: 1004.9. Samples: 164016. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:53:43,958][02158] Avg episode reward: [(0, '4.429')] [2025-08-21 19:53:48,952][02158] Fps is (10 sec: 4096.0, 60 sec: 3891.2, 300 sec: 3861.9). Total num frames: 675840. Throughput: 0: 1014.9. Samples: 166930. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:53:48,954][02158] Avg episode reward: [(0, '4.520')] [2025-08-21 19:53:52,942][11386] Updated weights for policy 0, policy_version 170 (0.0012) [2025-08-21 19:53:53,952][02158] Fps is (10 sec: 4506.6, 60 sec: 4027.7, 300 sec: 3891.2). Total num frames: 700416. Throughput: 0: 1025.7. Samples: 174072. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:53:53,953][02158] Avg episode reward: [(0, '4.569')] [2025-08-21 19:53:58,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 3874.6). Total num frames: 716800. Throughput: 0: 1017.8. Samples: 179440. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:53:58,953][02158] Avg episode reward: [(0, '4.627')] [2025-08-21 19:54:03,581][11386] Updated weights for policy 0, policy_version 180 (0.0016) [2025-08-21 19:54:03,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 3880.4). Total num frames: 737280. Throughput: 0: 1026.6. Samples: 182352. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:54:03,953][02158] Avg episode reward: [(0, '4.713')] [2025-08-21 19:54:03,955][11373] Saving new best policy, reward=4.713! [2025-08-21 19:54:08,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 3885.9). Total num frames: 757760. Throughput: 0: 1025.2. Samples: 189234. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:54:08,956][02158] Avg episode reward: [(0, '4.681')] [2025-08-21 19:54:13,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.7, 300 sec: 3870.7). Total num frames: 774144. Throughput: 0: 1015.9. Samples: 194372. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:54:13,953][02158] Avg episode reward: [(0, '4.603')] [2025-08-21 19:54:14,429][11386] Updated weights for policy 0, policy_version 190 (0.0024) [2025-08-21 19:54:18,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 3896.2). Total num frames: 798720. Throughput: 0: 1025.7. Samples: 197448. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:54:18,954][02158] Avg episode reward: [(0, '4.365')] [2025-08-21 19:54:23,241][11386] Updated weights for policy 0, policy_version 200 (0.0020) [2025-08-21 19:54:23,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.0, 300 sec: 3901.0). Total num frames: 819200. Throughput: 0: 1023.3. Samples: 204408. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:54:23,956][02158] Avg episode reward: [(0, '4.449')] [2025-08-21 19:54:28,955][02158] Fps is (10 sec: 3685.4, 60 sec: 4095.8, 300 sec: 3886.4). Total num frames: 835584. Throughput: 0: 1010.2. Samples: 209476. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:54:28,956][02158] Avg episode reward: [(0, '4.520')] [2025-08-21 19:54:28,967][11373] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000204_835584.pth... [2025-08-21 19:54:29,131][11373] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000204_835584.pth [2025-08-21 19:54:33,952][02158] Fps is (10 sec: 3686.3, 60 sec: 4027.7, 300 sec: 3891.2). Total num frames: 856064. Throughput: 0: 1014.8. Samples: 212598. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) [2025-08-21 19:54:33,953][02158] Avg episode reward: [(0, '4.653')] [2025-08-21 19:54:34,016][11386] Updated weights for policy 0, policy_version 210 (0.0028) [2025-08-21 19:54:38,952][02158] Fps is (10 sec: 4506.9, 60 sec: 4096.0, 300 sec: 3914.0). Total num frames: 880640. Throughput: 0: 1014.0. Samples: 219702. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:54:38,956][02158] Avg episode reward: [(0, '4.551')] [2025-08-21 19:54:43,952][02158] Fps is (10 sec: 4096.1, 60 sec: 4027.9, 300 sec: 3900.1). Total num frames: 897024. Throughput: 0: 1013.6. Samples: 225050. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:54:43,954][02158] Avg episode reward: [(0, '4.337')] [2025-08-21 19:54:44,534][11386] Updated weights for policy 0, policy_version 220 (0.0018) [2025-08-21 19:54:48,952][02158] Fps is (10 sec: 4095.9, 60 sec: 4096.0, 300 sec: 3921.7). Total num frames: 921600. Throughput: 0: 1018.4. Samples: 228182. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:54:48,957][02158] Avg episode reward: [(0, '4.568')] [2025-08-21 19:54:53,195][11386] Updated weights for policy 0, policy_version 230 (0.0023) [2025-08-21 19:54:53,952][02158] Fps is (10 sec: 4915.3, 60 sec: 4096.0, 300 sec: 3942.4). Total num frames: 946176. Throughput: 0: 1024.6. Samples: 235340. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:54:53,956][02158] Avg episode reward: [(0, '4.938')] [2025-08-21 19:54:53,959][11373] Saving new best policy, reward=4.938! [2025-08-21 19:54:58,952][02158] Fps is (10 sec: 3686.5, 60 sec: 4027.7, 300 sec: 3912.1). Total num frames: 958464. Throughput: 0: 1023.3. Samples: 240420. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) [2025-08-21 19:54:58,955][02158] Avg episode reward: [(0, '4.913')] [2025-08-21 19:55:03,866][11386] Updated weights for policy 0, policy_version 240 (0.0011) [2025-08-21 19:55:03,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 3932.2). Total num frames: 983040. Throughput: 0: 1026.8. Samples: 243652. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:55:03,955][02158] Avg episode reward: [(0, '4.589')] [2025-08-21 19:55:08,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.0, 300 sec: 3935.4). Total num frames: 1003520. Throughput: 0: 1029.1. Samples: 250716. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) [2025-08-21 19:55:08,957][02158] Avg episode reward: [(0, '4.424')] [2025-08-21 19:55:13,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 3922.7). Total num frames: 1019904. Throughput: 0: 1031.3. Samples: 255880. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:55:13,953][02158] Avg episode reward: [(0, '4.468')] [2025-08-21 19:55:14,400][11386] Updated weights for policy 0, policy_version 250 (0.0019) [2025-08-21 19:55:18,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 3941.4). Total num frames: 1044480. Throughput: 0: 1035.1. Samples: 259176. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:55:18,957][02158] Avg episode reward: [(0, '4.453')] [2025-08-21 19:55:22,940][11386] Updated weights for policy 0, policy_version 260 (0.0014) [2025-08-21 19:55:23,955][02158] Fps is (10 sec: 4913.8, 60 sec: 4164.1, 300 sec: 3959.4). Total num frames: 1069056. Throughput: 0: 1035.2. Samples: 266290. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:55:23,960][02158] Avg episode reward: [(0, '4.380')] [2025-08-21 19:55:28,955][02158] Fps is (10 sec: 3685.3, 60 sec: 4096.0, 300 sec: 3932.1). Total num frames: 1081344. Throughput: 0: 1028.3. Samples: 271326. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:55:28,956][02158] Avg episode reward: [(0, '4.429')] [2025-08-21 19:55:33,637][11386] Updated weights for policy 0, policy_version 270 (0.0023) [2025-08-21 19:55:33,952][02158] Fps is (10 sec: 3687.5, 60 sec: 4164.3, 300 sec: 3949.7). Total num frames: 1105920. Throughput: 0: 1030.9. Samples: 274574. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:55:33,955][02158] Avg episode reward: [(0, '4.567')] [2025-08-21 19:55:38,952][02158] Fps is (10 sec: 4916.8, 60 sec: 4164.3, 300 sec: 3966.7). Total num frames: 1130496. Throughput: 0: 1029.0. Samples: 281644. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:55:38,956][02158] Avg episode reward: [(0, '4.547')] [2025-08-21 19:55:43,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 3940.6). Total num frames: 1142784. Throughput: 0: 1026.8. Samples: 286624. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:55:43,954][02158] Avg episode reward: [(0, '4.782')] [2025-08-21 19:55:44,137][11386] Updated weights for policy 0, policy_version 280 (0.0016) [2025-08-21 19:55:48,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 3957.2). Total num frames: 1167360. Throughput: 0: 1030.0. Samples: 290004. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:55:48,956][02158] Avg episode reward: [(0, '4.779')] [2025-08-21 19:55:52,941][11386] Updated weights for policy 0, policy_version 290 (0.0018) [2025-08-21 19:55:53,955][02158] Fps is (10 sec: 4913.8, 60 sec: 4095.8, 300 sec: 4040.4). Total num frames: 1191936. Throughput: 0: 1029.6. Samples: 297050. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:55:53,956][02158] Avg episode reward: [(0, '4.813')] [2025-08-21 19:55:58,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 1204224. Throughput: 0: 1018.0. Samples: 301692. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:55:58,957][02158] Avg episode reward: [(0, '4.948')] [2025-08-21 19:55:58,963][11373] Saving new best policy, reward=4.948! [2025-08-21 19:56:03,781][11386] Updated weights for policy 0, policy_version 300 (0.0014) [2025-08-21 19:56:03,952][02158] Fps is (10 sec: 3687.5, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 1228800. Throughput: 0: 1017.8. Samples: 304978. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:56:03,956][02158] Avg episode reward: [(0, '4.890')] [2025-08-21 19:56:08,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 1249280. Throughput: 0: 1017.3. Samples: 312064. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:56:08,955][02158] Avg episode reward: [(0, '4.846')] [2025-08-21 19:56:13,953][02158] Fps is (10 sec: 3686.1, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 1265664. Throughput: 0: 1014.0. Samples: 316954. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:56:13,960][02158] Avg episode reward: [(0, '4.947')] [2025-08-21 19:56:14,339][11386] Updated weights for policy 0, policy_version 310 (0.0014) [2025-08-21 19:56:18,952][02158] Fps is (10 sec: 4095.9, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 1290240. Throughput: 0: 1021.3. Samples: 320532. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:56:18,954][02158] Avg episode reward: [(0, '4.840')] [2025-08-21 19:56:23,035][11386] Updated weights for policy 0, policy_version 320 (0.0017) [2025-08-21 19:56:23,953][02158] Fps is (10 sec: 4505.4, 60 sec: 4027.9, 300 sec: 4054.4). Total num frames: 1310720. Throughput: 0: 1020.2. Samples: 327552. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:56:23,954][02158] Avg episode reward: [(0, '4.695')] [2025-08-21 19:56:28,952][02158] Fps is (10 sec: 3686.5, 60 sec: 4096.2, 300 sec: 4054.3). Total num frames: 1327104. Throughput: 0: 1013.4. Samples: 332228. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:56:28,953][02158] Avg episode reward: [(0, '4.691')] [2025-08-21 19:56:28,959][11373] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000324_1327104.pth... [2025-08-21 19:56:29,067][11373] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000324_1327104.pth [2025-08-21 19:56:33,948][11386] Updated weights for policy 0, policy_version 330 (0.0018) [2025-08-21 19:56:33,952][02158] Fps is (10 sec: 4096.4, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 1351680. Throughput: 0: 1011.8. Samples: 335534. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:56:33,953][02158] Avg episode reward: [(0, '4.550')] [2025-08-21 19:56:38,957][02158] Fps is (10 sec: 4503.4, 60 sec: 4027.4, 300 sec: 4054.3). Total num frames: 1372160. Throughput: 0: 1012.8. Samples: 342628. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:56:38,968][02158] Avg episode reward: [(0, '4.702')] [2025-08-21 19:56:43,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 1388544. Throughput: 0: 1017.9. Samples: 347496. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:56:43,955][02158] Avg episode reward: [(0, '4.746')] [2025-08-21 19:56:44,504][11386] Updated weights for policy 0, policy_version 340 (0.0022) [2025-08-21 19:56:48,952][02158] Fps is (10 sec: 3688.2, 60 sec: 4027.7, 300 sec: 4040.5). Total num frames: 1409024. Throughput: 0: 1023.0. Samples: 351012. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:56:48,957][02158] Avg episode reward: [(0, '4.895')] [2025-08-21 19:56:53,382][11386] Updated weights for policy 0, policy_version 350 (0.0018) [2025-08-21 19:56:53,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4027.9, 300 sec: 4054.4). Total num frames: 1433600. Throughput: 0: 1022.9. Samples: 358094. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:56:53,955][02158] Avg episode reward: [(0, '4.852')] [2025-08-21 19:56:58,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 1449984. Throughput: 0: 1019.3. Samples: 362824. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:56:58,953][02158] Avg episode reward: [(0, '4.796')] [2025-08-21 19:57:03,838][11386] Updated weights for policy 0, policy_version 360 (0.0015) [2025-08-21 19:57:03,952][02158] Fps is (10 sec: 4095.9, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 1474560. Throughput: 0: 1019.2. Samples: 366398. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:57:03,955][02158] Avg episode reward: [(0, '4.977')] [2025-08-21 19:57:03,960][11373] Saving new best policy, reward=4.977! [2025-08-21 19:57:08,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.7, 300 sec: 4040.5). Total num frames: 1490944. Throughput: 0: 1017.4. Samples: 373336. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:57:08,953][02158] Avg episode reward: [(0, '4.857')] [2025-08-21 19:57:13,952][02158] Fps is (10 sec: 2867.3, 60 sec: 3959.5, 300 sec: 4012.7). Total num frames: 1503232. Throughput: 0: 997.7. Samples: 377126. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:57:13,953][02158] Avg episode reward: [(0, '4.617')] [2025-08-21 19:57:16,202][11386] Updated weights for policy 0, policy_version 370 (0.0022) [2025-08-21 19:57:18,952][02158] Fps is (10 sec: 3686.4, 60 sec: 3959.5, 300 sec: 4026.6). Total num frames: 1527808. Throughput: 0: 983.4. Samples: 379786. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:57:18,956][02158] Avg episode reward: [(0, '4.567')] [2025-08-21 19:57:23,952][02158] Fps is (10 sec: 4505.6, 60 sec: 3959.5, 300 sec: 4026.6). Total num frames: 1548288. Throughput: 0: 978.4. Samples: 386652. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:57:23,953][02158] Avg episode reward: [(0, '4.531')] [2025-08-21 19:57:25,099][11386] Updated weights for policy 0, policy_version 380 (0.0015) [2025-08-21 19:57:28,954][02158] Fps is (10 sec: 3685.7, 60 sec: 3959.3, 300 sec: 4012.7). Total num frames: 1564672. Throughput: 0: 996.4. Samples: 392336. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:57:28,955][02158] Avg episode reward: [(0, '4.548')] [2025-08-21 19:57:33,952][02158] Fps is (10 sec: 3686.4, 60 sec: 3891.2, 300 sec: 3998.8). Total num frames: 1585152. Throughput: 0: 975.0. Samples: 394888. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:57:33,956][02158] Avg episode reward: [(0, '4.784')] [2025-08-21 19:57:35,683][11386] Updated weights for policy 0, policy_version 390 (0.0020) [2025-08-21 19:57:38,952][02158] Fps is (10 sec: 4506.4, 60 sec: 3959.8, 300 sec: 4026.6). Total num frames: 1609728. Throughput: 0: 975.2. Samples: 401980. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:57:38,953][02158] Avg episode reward: [(0, '4.840')] [2025-08-21 19:57:43,953][02158] Fps is (10 sec: 4095.5, 60 sec: 3959.4, 300 sec: 4012.7). Total num frames: 1626112. Throughput: 0: 995.1. Samples: 407604. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:57:43,958][02158] Avg episode reward: [(0, '5.066')] [2025-08-21 19:57:44,021][11373] Saving new best policy, reward=5.066! [2025-08-21 19:57:46,230][11386] Updated weights for policy 0, policy_version 400 (0.0012) [2025-08-21 19:57:48,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.7, 300 sec: 4040.5). Total num frames: 1650688. Throughput: 0: 973.9. Samples: 410222. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:57:48,954][02158] Avg episode reward: [(0, '5.127')] [2025-08-21 19:57:48,964][11373] Saving new best policy, reward=5.127! [2025-08-21 19:57:53,952][02158] Fps is (10 sec: 4506.1, 60 sec: 3959.5, 300 sec: 4068.2). Total num frames: 1671168. Throughput: 0: 976.6. Samples: 417282. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:57:53,953][02158] Avg episode reward: [(0, '4.903')] [2025-08-21 19:57:54,921][11386] Updated weights for policy 0, policy_version 410 (0.0020) [2025-08-21 19:57:58,952][02158] Fps is (10 sec: 3686.3, 60 sec: 3959.4, 300 sec: 4054.3). Total num frames: 1687552. Throughput: 0: 1017.3. Samples: 422904. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:57:58,954][02158] Avg episode reward: [(0, '4.923')] [2025-08-21 19:58:03,952][02158] Fps is (10 sec: 4096.0, 60 sec: 3959.5, 300 sec: 4068.2). Total num frames: 1712128. Throughput: 0: 1021.1. Samples: 425734. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:58:03,955][02158] Avg episode reward: [(0, '5.177')] [2025-08-21 19:58:03,958][11373] Saving new best policy, reward=5.177! [2025-08-21 19:58:05,814][11386] Updated weights for policy 0, policy_version 420 (0.0014) [2025-08-21 19:58:08,952][02158] Fps is (10 sec: 4505.8, 60 sec: 4027.7, 300 sec: 4068.2). Total num frames: 1732608. Throughput: 0: 1018.3. Samples: 432476. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:58:08,956][02158] Avg episode reward: [(0, '5.248')] [2025-08-21 19:58:08,963][11373] Saving new best policy, reward=5.248! [2025-08-21 19:58:13,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 1748992. Throughput: 0: 1015.2. Samples: 438016. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:58:13,957][02158] Avg episode reward: [(0, '5.500')] [2025-08-21 19:58:13,959][11373] Saving new best policy, reward=5.500! [2025-08-21 19:58:16,412][11386] Updated weights for policy 0, policy_version 430 (0.0027) [2025-08-21 19:58:18,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.7, 300 sec: 4054.3). Total num frames: 1769472. Throughput: 0: 1021.2. Samples: 440842. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:58:18,957][02158] Avg episode reward: [(0, '5.619')] [2025-08-21 19:58:19,033][11373] Saving new best policy, reward=5.619! [2025-08-21 19:58:23,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.0, 300 sec: 4082.1). Total num frames: 1794048. Throughput: 0: 1017.6. Samples: 447770. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:58:23,953][02158] Avg episode reward: [(0, '6.154')] [2025-08-21 19:58:23,954][11373] Saving new best policy, reward=6.154! [2025-08-21 19:58:25,354][11386] Updated weights for policy 0, policy_version 440 (0.0016) [2025-08-21 19:58:28,954][02158] Fps is (10 sec: 4095.0, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 1810432. Throughput: 0: 1009.1. Samples: 453016. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:58:28,957][02158] Avg episode reward: [(0, '6.782')] [2025-08-21 19:58:28,967][11373] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000442_1810432.pth... [2025-08-21 19:58:29,113][11373] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000442_1810432.pth [2025-08-21 19:58:29,127][11373] Saving new best policy, reward=6.782! [2025-08-21 19:58:33,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 1830912. Throughput: 0: 1014.8. Samples: 455890. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 19:58:33,957][02158] Avg episode reward: [(0, '6.613')] [2025-08-21 19:58:36,032][11386] Updated weights for policy 0, policy_version 450 (0.0024) [2025-08-21 19:58:38,952][02158] Fps is (10 sec: 4506.7, 60 sec: 4096.0, 300 sec: 4068.3). Total num frames: 1855488. Throughput: 0: 1016.2. Samples: 463012. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:58:38,953][02158] Avg episode reward: [(0, '6.566')] [2025-08-21 19:58:43,954][02158] Fps is (10 sec: 4095.3, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 1871872. Throughput: 0: 1009.5. Samples: 468332. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:58:43,955][02158] Avg episode reward: [(0, '6.783')] [2025-08-21 19:58:43,958][11373] Saving new best policy, reward=6.783! [2025-08-21 19:58:46,739][11386] Updated weights for policy 0, policy_version 460 (0.0034) [2025-08-21 19:58:48,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.7, 300 sec: 4040.5). Total num frames: 1892352. Throughput: 0: 1014.1. Samples: 471368. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:58:48,955][02158] Avg episode reward: [(0, '7.128')] [2025-08-21 19:58:48,965][11373] Saving new best policy, reward=7.128! [2025-08-21 19:58:53,952][02158] Fps is (10 sec: 4506.2, 60 sec: 4096.0, 300 sec: 4068.2). Total num frames: 1916928. Throughput: 0: 1019.1. Samples: 478334. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:58:53,957][02158] Avg episode reward: [(0, '6.561')] [2025-08-21 19:58:56,114][11386] Updated weights for policy 0, policy_version 470 (0.0016) [2025-08-21 19:58:58,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 1933312. Throughput: 0: 1012.9. Samples: 483598. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:58:58,955][02158] Avg episode reward: [(0, '6.267')] [2025-08-21 19:59:03,952][02158] Fps is (10 sec: 3686.5, 60 sec: 4027.7, 300 sec: 4054.3). Total num frames: 1953792. Throughput: 0: 1014.6. Samples: 486500. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 19:59:03,953][02158] Avg episode reward: [(0, '6.402')] [2025-08-21 19:59:06,168][11386] Updated weights for policy 0, policy_version 480 (0.0023) [2025-08-21 19:59:08,952][02158] Fps is (10 sec: 4505.5, 60 sec: 4096.0, 300 sec: 4082.1). Total num frames: 1978368. Throughput: 0: 1019.7. Samples: 493658. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:59:08,957][02158] Avg episode reward: [(0, '7.022')] [2025-08-21 19:59:13,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.7, 300 sec: 4040.5). Total num frames: 1990656. Throughput: 0: 1019.2. Samples: 498878. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:59:13,955][02158] Avg episode reward: [(0, '7.348')] [2025-08-21 19:59:13,958][11373] Saving new best policy, reward=7.348! [2025-08-21 19:59:16,721][11386] Updated weights for policy 0, policy_version 490 (0.0012) [2025-08-21 19:59:18,952][02158] Fps is (10 sec: 3686.5, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 2015232. Throughput: 0: 1024.4. Samples: 501988. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:59:18,956][02158] Avg episode reward: [(0, '8.149')] [2025-08-21 19:59:18,962][11373] Saving new best policy, reward=8.149! [2025-08-21 19:59:23,952][02158] Fps is (10 sec: 4915.2, 60 sec: 4096.0, 300 sec: 4082.2). Total num frames: 2039808. Throughput: 0: 1024.3. Samples: 509106. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 19:59:23,956][02158] Avg episode reward: [(0, '8.768')] [2025-08-21 19:59:23,960][11373] Saving new best policy, reward=8.768! [2025-08-21 19:59:26,285][11386] Updated weights for policy 0, policy_version 500 (0.0026) [2025-08-21 19:59:28,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.2, 300 sec: 4068.2). Total num frames: 2056192. Throughput: 0: 1020.3. Samples: 514244. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 19:59:28,953][02158] Avg episode reward: [(0, '9.150')] [2025-08-21 19:59:28,960][11373] Saving new best policy, reward=9.150! [2025-08-21 19:59:33,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 2076672. Throughput: 0: 1020.9. Samples: 517310. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 19:59:33,956][02158] Avg episode reward: [(0, '9.589')] [2025-08-21 19:59:33,960][11373] Saving new best policy, reward=9.589! [2025-08-21 19:59:36,503][11386] Updated weights for policy 0, policy_version 510 (0.0031) [2025-08-21 19:59:38,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.7, 300 sec: 4068.2). Total num frames: 2097152. Throughput: 0: 1010.5. Samples: 523808. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 19:59:38,953][02158] Avg episode reward: [(0, '9.928')] [2025-08-21 19:59:39,015][11373] Saving new best policy, reward=9.928! [2025-08-21 19:59:43,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.8, 300 sec: 4040.5). Total num frames: 2113536. Throughput: 0: 1001.9. Samples: 528684. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) [2025-08-21 19:59:43,955][02158] Avg episode reward: [(0, '10.221')] [2025-08-21 19:59:43,960][11373] Saving new best policy, reward=10.221! [2025-08-21 19:59:47,290][11386] Updated weights for policy 0, policy_version 520 (0.0011) [2025-08-21 19:59:48,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.7, 300 sec: 4026.6). Total num frames: 2134016. Throughput: 0: 1013.2. Samples: 532094. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 19:59:48,953][02158] Avg episode reward: [(0, '9.997')] [2025-08-21 19:59:53,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4027.8, 300 sec: 4068.2). Total num frames: 2158592. Throughput: 0: 1010.2. Samples: 539118. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) [2025-08-21 19:59:53,955][02158] Avg episode reward: [(0, '9.318')] [2025-08-21 19:59:56,968][11386] Updated weights for policy 0, policy_version 530 (0.0014) [2025-08-21 19:59:58,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.7, 300 sec: 4040.5). Total num frames: 2174976. Throughput: 0: 1001.2. Samples: 543932. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) [2025-08-21 19:59:58,955][02158] Avg episode reward: [(0, '8.927')] [2025-08-21 20:00:03,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.7, 300 sec: 4040.5). Total num frames: 2195456. Throughput: 0: 1005.2. Samples: 547220. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 20:00:03,953][02158] Avg episode reward: [(0, '9.344')] [2025-08-21 20:00:06,576][11386] Updated weights for policy 0, policy_version 540 (0.0015) [2025-08-21 20:00:08,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4027.8, 300 sec: 4068.2). Total num frames: 2220032. Throughput: 0: 1009.6. Samples: 554536. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:00:08,956][02158] Avg episode reward: [(0, '9.447')] [2025-08-21 20:00:13,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 2236416. Throughput: 0: 1007.6. Samples: 559584. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 20:00:13,956][02158] Avg episode reward: [(0, '9.531')] [2025-08-21 20:00:17,229][11386] Updated weights for policy 0, policy_version 550 (0.0015) [2025-08-21 20:00:18,952][02158] Fps is (10 sec: 4095.9, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 2260992. Throughput: 0: 1012.8. Samples: 562888. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:00:18,956][02158] Avg episode reward: [(0, '9.328')] [2025-08-21 20:00:23,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4027.7, 300 sec: 4068.3). Total num frames: 2281472. Throughput: 0: 1026.6. Samples: 570006. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:00:23,957][02158] Avg episode reward: [(0, '10.144')] [2025-08-21 20:00:27,439][11386] Updated weights for policy 0, policy_version 560 (0.0021) [2025-08-21 20:00:28,953][02158] Fps is (10 sec: 3686.1, 60 sec: 4027.7, 300 sec: 4040.4). Total num frames: 2297856. Throughput: 0: 1027.5. Samples: 574924. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:00:28,954][02158] Avg episode reward: [(0, '10.493')] [2025-08-21 20:00:28,963][11373] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000561_2297856.pth... [2025-08-21 20:00:29,086][11373] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000561_2297856.pth [2025-08-21 20:00:29,097][11373] Saving new best policy, reward=10.493! [2025-08-21 20:00:33,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4027.7, 300 sec: 4026.6). Total num frames: 2318336. Throughput: 0: 1020.9. Samples: 578036. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 20:00:33,956][02158] Avg episode reward: [(0, '10.958')] [2025-08-21 20:00:33,958][11373] Saving new best policy, reward=10.958! [2025-08-21 20:00:36,826][11386] Updated weights for policy 0, policy_version 570 (0.0024) [2025-08-21 20:00:38,955][02158] Fps is (10 sec: 4095.3, 60 sec: 4027.6, 300 sec: 4054.3). Total num frames: 2338816. Throughput: 0: 1020.0. Samples: 585020. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:00:38,959][02158] Avg episode reward: [(0, '11.665')] [2025-08-21 20:00:38,985][11373] Saving new best policy, reward=11.665! [2025-08-21 20:00:43,953][02158] Fps is (10 sec: 4095.6, 60 sec: 4095.9, 300 sec: 4040.4). Total num frames: 2359296. Throughput: 0: 1023.2. Samples: 589978. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:00:43,958][02158] Avg episode reward: [(0, '12.141')] [2025-08-21 20:00:43,964][11373] Saving new best policy, reward=12.141! [2025-08-21 20:00:47,301][11386] Updated weights for policy 0, policy_version 580 (0.0011) [2025-08-21 20:00:48,952][02158] Fps is (10 sec: 4097.1, 60 sec: 4096.0, 300 sec: 4026.6). Total num frames: 2379776. Throughput: 0: 1025.0. Samples: 593344. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) [2025-08-21 20:00:48,957][02158] Avg episode reward: [(0, '12.436')] [2025-08-21 20:00:48,965][11373] Saving new best policy, reward=12.436! [2025-08-21 20:00:53,952][02158] Fps is (10 sec: 4506.0, 60 sec: 4096.0, 300 sec: 4068.2). Total num frames: 2404352. Throughput: 0: 1016.9. Samples: 600296. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 20:00:53,955][02158] Avg episode reward: [(0, '12.543')] [2025-08-21 20:00:53,956][11373] Saving new best policy, reward=12.543! [2025-08-21 20:00:58,135][11386] Updated weights for policy 0, policy_version 590 (0.0019) [2025-08-21 20:00:58,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 2420736. Throughput: 0: 1013.4. Samples: 605186. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) [2025-08-21 20:00:58,957][02158] Avg episode reward: [(0, '12.270')] [2025-08-21 20:01:03,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 2441216. Throughput: 0: 1018.1. Samples: 608704. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:01:03,959][02158] Avg episode reward: [(0, '11.996')] [2025-08-21 20:01:06,685][11386] Updated weights for policy 0, policy_version 600 (0.0020) [2025-08-21 20:01:08,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.7, 300 sec: 4054.4). Total num frames: 2461696. Throughput: 0: 1011.6. Samples: 615528. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) [2025-08-21 20:01:08,955][02158] Avg episode reward: [(0, '11.801')] [2025-08-21 20:01:13,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 2482176. Throughput: 0: 1015.2. Samples: 620608. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:01:13,953][02158] Avg episode reward: [(0, '12.340')] [2025-08-21 20:01:17,269][11386] Updated weights for policy 0, policy_version 610 (0.0016) [2025-08-21 20:01:18,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.7, 300 sec: 4040.5). Total num frames: 2502656. Throughput: 0: 1024.8. Samples: 624152. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:01:18,953][02158] Avg episode reward: [(0, '12.233')] [2025-08-21 20:01:23,952][02158] Fps is (10 sec: 3686.4, 60 sec: 3959.5, 300 sec: 4040.5). Total num frames: 2519040. Throughput: 0: 1005.1. Samples: 630246. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:01:23,955][02158] Avg episode reward: [(0, '12.637')] [2025-08-21 20:01:23,958][11373] Saving new best policy, reward=12.637! [2025-08-21 20:01:28,952][02158] Fps is (10 sec: 2867.1, 60 sec: 3891.2, 300 sec: 3998.8). Total num frames: 2531328. Throughput: 0: 974.5. Samples: 633832. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:01:28,955][02158] Avg episode reward: [(0, '12.977')] [2025-08-21 20:01:28,963][11373] Saving new best policy, reward=12.977! [2025-08-21 20:01:30,064][11386] Updated weights for policy 0, policy_version 620 (0.0034) [2025-08-21 20:01:33,952][02158] Fps is (10 sec: 3686.4, 60 sec: 3959.5, 300 sec: 4012.8). Total num frames: 2555904. Throughput: 0: 965.2. Samples: 636776. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) [2025-08-21 20:01:33,953][02158] Avg episode reward: [(0, '12.797')] [2025-08-21 20:01:38,810][11386] Updated weights for policy 0, policy_version 630 (0.0021) [2025-08-21 20:01:38,953][02158] Fps is (10 sec: 4914.8, 60 sec: 4027.8, 300 sec: 4040.4). Total num frames: 2580480. Throughput: 0: 971.7. Samples: 644022. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 20:01:38,955][02158] Avg episode reward: [(0, '13.390')] [2025-08-21 20:01:38,965][11373] Saving new best policy, reward=13.390! [2025-08-21 20:01:43,952][02158] Fps is (10 sec: 3686.4, 60 sec: 3891.3, 300 sec: 4012.7). Total num frames: 2592768. Throughput: 0: 976.0. Samples: 649106. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:01:43,956][02158] Avg episode reward: [(0, '12.968')] [2025-08-21 20:01:48,952][02158] Fps is (10 sec: 3686.8, 60 sec: 3959.5, 300 sec: 4012.7). Total num frames: 2617344. Throughput: 0: 975.9. Samples: 652620. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:01:48,956][02158] Avg episode reward: [(0, '13.766')] [2025-08-21 20:01:48,964][11373] Saving new best policy, reward=13.766! [2025-08-21 20:01:49,371][11386] Updated weights for policy 0, policy_version 640 (0.0035) [2025-08-21 20:01:53,952][02158] Fps is (10 sec: 4915.0, 60 sec: 3959.4, 300 sec: 4040.5). Total num frames: 2641920. Throughput: 0: 978.7. Samples: 659572. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 20:01:53,956][02158] Avg episode reward: [(0, '13.949')] [2025-08-21 20:01:53,960][11373] Saving new best policy, reward=13.949! [2025-08-21 20:01:58,952][02158] Fps is (10 sec: 3686.5, 60 sec: 3891.2, 300 sec: 3998.8). Total num frames: 2654208. Throughput: 0: 973.0. Samples: 664394. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:01:58,957][02158] Avg episode reward: [(0, '13.550')] [2025-08-21 20:02:00,021][11386] Updated weights for policy 0, policy_version 650 (0.0019) [2025-08-21 20:02:03,952][02158] Fps is (10 sec: 3686.5, 60 sec: 3959.5, 300 sec: 4026.6). Total num frames: 2678784. Throughput: 0: 968.8. Samples: 667750. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:02:03,957][02158] Avg episode reward: [(0, '14.365')] [2025-08-21 20:02:03,961][11373] Saving new best policy, reward=14.365! [2025-08-21 20:02:08,843][11386] Updated weights for policy 0, policy_version 660 (0.0013) [2025-08-21 20:02:08,952][02158] Fps is (10 sec: 4915.2, 60 sec: 4027.7, 300 sec: 4068.2). Total num frames: 2703360. Throughput: 0: 991.8. Samples: 674876. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:02:08,953][02158] Avg episode reward: [(0, '14.826')] [2025-08-21 20:02:08,964][11373] Saving new best policy, reward=14.826! [2025-08-21 20:02:13,952][02158] Fps is (10 sec: 3686.3, 60 sec: 3891.2, 300 sec: 4026.6). Total num frames: 2715648. Throughput: 0: 1021.8. Samples: 679814. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:02:13,957][02158] Avg episode reward: [(0, '16.664')] [2025-08-21 20:02:13,962][11373] Saving new best policy, reward=16.664! [2025-08-21 20:02:18,952][02158] Fps is (10 sec: 3686.4, 60 sec: 3959.5, 300 sec: 4040.5). Total num frames: 2740224. Throughput: 0: 1035.5. Samples: 683372. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 20:02:18,953][02158] Avg episode reward: [(0, '17.674')] [2025-08-21 20:02:18,963][11373] Saving new best policy, reward=17.674! [2025-08-21 20:02:19,396][11386] Updated weights for policy 0, policy_version 670 (0.0031) [2025-08-21 20:02:23,952][02158] Fps is (10 sec: 4915.1, 60 sec: 4096.0, 300 sec: 4068.3). Total num frames: 2764800. Throughput: 0: 1031.2. Samples: 690426. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:02:23,954][02158] Avg episode reward: [(0, '18.301')] [2025-08-21 20:02:23,955][11373] Saving new best policy, reward=18.301! [2025-08-21 20:02:28,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 2777088. Throughput: 0: 1024.6. Samples: 695212. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 20:02:28,953][02158] Avg episode reward: [(0, '18.961')] [2025-08-21 20:02:28,993][11373] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000679_2781184.pth... [2025-08-21 20:02:29,103][11373] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000679_2781184.pth [2025-08-21 20:02:29,117][11373] Saving new best policy, reward=18.961! [2025-08-21 20:02:30,013][11386] Updated weights for policy 0, policy_version 680 (0.0023) [2025-08-21 20:02:33,952][02158] Fps is (10 sec: 3686.5, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 2801664. Throughput: 0: 1021.0. Samples: 698564. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:02:33,956][02158] Avg episode reward: [(0, '20.341')] [2025-08-21 20:02:33,960][11373] Saving new best policy, reward=20.341! [2025-08-21 20:02:38,957][02158] Fps is (10 sec: 4503.4, 60 sec: 4027.5, 300 sec: 4054.3). Total num frames: 2822144. Throughput: 0: 1025.7. Samples: 705732. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:02:38,959][02158] Avg episode reward: [(0, '19.651')] [2025-08-21 20:02:39,133][11386] Updated weights for policy 0, policy_version 690 (0.0016) [2025-08-21 20:02:43,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4040.5). Total num frames: 2842624. Throughput: 0: 1025.7. Samples: 710550. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 20:02:43,953][02158] Avg episode reward: [(0, '19.938')] [2025-08-21 20:02:48,952][02158] Fps is (10 sec: 4098.0, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 2863104. Throughput: 0: 1033.1. Samples: 714238. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:02:48,953][02158] Avg episode reward: [(0, '17.840')] [2025-08-21 20:02:49,183][11386] Updated weights for policy 0, policy_version 700 (0.0031) [2025-08-21 20:02:53,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.0, 300 sec: 4068.2). Total num frames: 2887680. Throughput: 0: 1036.3. Samples: 721510. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:02:53,956][02158] Avg episode reward: [(0, '16.792')] [2025-08-21 20:02:58,952][02158] Fps is (10 sec: 4095.9, 60 sec: 4164.2, 300 sec: 4040.5). Total num frames: 2904064. Throughput: 0: 1031.1. Samples: 726212. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 20:02:58,957][02158] Avg episode reward: [(0, '16.327')] [2025-08-21 20:02:59,677][11386] Updated weights for policy 0, policy_version 710 (0.0011) [2025-08-21 20:03:03,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 2924544. Throughput: 0: 1026.6. Samples: 729568. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:03:03,955][02158] Avg episode reward: [(0, '17.190')] [2025-08-21 20:03:08,952][02158] Fps is (10 sec: 4096.1, 60 sec: 4027.7, 300 sec: 4054.3). Total num frames: 2945024. Throughput: 0: 1026.7. Samples: 736626. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:03:08,953][02158] Avg episode reward: [(0, '17.399')] [2025-08-21 20:03:08,975][11386] Updated weights for policy 0, policy_version 720 (0.0018) [2025-08-21 20:03:13,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4054.3). Total num frames: 2965504. Throughput: 0: 1032.1. Samples: 741658. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:03:13,956][02158] Avg episode reward: [(0, '19.338')] [2025-08-21 20:03:18,952][02158] Fps is (10 sec: 4095.9, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 2985984. Throughput: 0: 1034.3. Samples: 745106. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:03:18,956][02158] Avg episode reward: [(0, '19.399')] [2025-08-21 20:03:19,106][11386] Updated weights for policy 0, policy_version 730 (0.0015) [2025-08-21 20:03:23,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.8, 300 sec: 4054.4). Total num frames: 3006464. Throughput: 0: 1028.1. Samples: 751992. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:03:23,955][02158] Avg episode reward: [(0, '17.165')] [2025-08-21 20:03:28,952][02158] Fps is (10 sec: 4096.1, 60 sec: 4164.3, 300 sec: 4054.3). Total num frames: 3026944. Throughput: 0: 1032.2. Samples: 756998. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:03:28,955][02158] Avg episode reward: [(0, '16.841')] [2025-08-21 20:03:29,654][11386] Updated weights for policy 0, policy_version 740 (0.0021) [2025-08-21 20:03:33,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 3047424. Throughput: 0: 1025.8. Samples: 760398. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:03:33,953][02158] Avg episode reward: [(0, '16.885')] [2025-08-21 20:03:38,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.3, 300 sec: 4054.4). Total num frames: 3067904. Throughput: 0: 1013.6. Samples: 767124. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:03:38,954][02158] Avg episode reward: [(0, '16.441')] [2025-08-21 20:03:39,348][11386] Updated weights for policy 0, policy_version 750 (0.0016) [2025-08-21 20:03:43,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 3088384. Throughput: 0: 1025.0. Samples: 772338. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:03:43,956][02158] Avg episode reward: [(0, '17.512')] [2025-08-21 20:03:48,900][11386] Updated weights for policy 0, policy_version 760 (0.0011) [2025-08-21 20:03:48,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4054.3). Total num frames: 3112960. Throughput: 0: 1030.3. Samples: 775932. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:03:48,953][02158] Avg episode reward: [(0, '19.792')] [2025-08-21 20:03:53,955][02158] Fps is (10 sec: 4094.9, 60 sec: 4027.6, 300 sec: 4054.3). Total num frames: 3129344. Throughput: 0: 1022.6. Samples: 782644. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:03:53,956][02158] Avg episode reward: [(0, '19.469')] [2025-08-21 20:03:58,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 3149824. Throughput: 0: 1028.2. Samples: 787928. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 20:03:58,955][02158] Avg episode reward: [(0, '20.072')] [2025-08-21 20:03:59,517][11386] Updated weights for policy 0, policy_version 770 (0.0028) [2025-08-21 20:04:03,952][02158] Fps is (10 sec: 4097.0, 60 sec: 4096.0, 300 sec: 4040.5). Total num frames: 3170304. Throughput: 0: 1026.7. Samples: 791306. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:04:03,955][02158] Avg episode reward: [(0, '20.490')] [2025-08-21 20:04:04,008][11373] Saving new best policy, reward=20.490! [2025-08-21 20:04:08,956][02158] Fps is (10 sec: 4094.3, 60 sec: 4095.7, 300 sec: 4068.2). Total num frames: 3190784. Throughput: 0: 1019.2. Samples: 797860. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 20:04:08,963][02158] Avg episode reward: [(0, '19.201')] [2025-08-21 20:04:09,521][11386] Updated weights for policy 0, policy_version 780 (0.0014) [2025-08-21 20:04:13,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 3211264. Throughput: 0: 1030.6. Samples: 803374. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:04:13,954][02158] Avg episode reward: [(0, '18.859')] [2025-08-21 20:04:18,393][11386] Updated weights for policy 0, policy_version 790 (0.0013) [2025-08-21 20:04:18,952][02158] Fps is (10 sec: 4507.4, 60 sec: 4164.3, 300 sec: 4054.3). Total num frames: 3235840. Throughput: 0: 1036.1. Samples: 807022. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 20:04:18,955][02158] Avg episode reward: [(0, '18.449')] [2025-08-21 20:04:23,954][02158] Fps is (10 sec: 4504.9, 60 sec: 4164.1, 300 sec: 4068.2). Total num frames: 3256320. Throughput: 0: 1031.4. Samples: 813540. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:04:23,956][02158] Avg episode reward: [(0, '17.691')] [2025-08-21 20:04:28,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4054.3). Total num frames: 3272704. Throughput: 0: 1037.1. Samples: 819008. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 20:04:28,953][02158] Avg episode reward: [(0, '18.019')] [2025-08-21 20:04:29,029][11386] Updated weights for policy 0, policy_version 800 (0.0026) [2025-08-21 20:04:29,035][11373] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000800_3276800.pth... [2025-08-21 20:04:29,139][11373] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000800_3276800.pth [2025-08-21 20:04:33,952][02158] Fps is (10 sec: 4096.8, 60 sec: 4164.3, 300 sec: 4068.2). Total num frames: 3297280. Throughput: 0: 1036.0. Samples: 822554. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:04:33,953][02158] Avg episode reward: [(0, '18.609')] [2025-08-21 20:04:38,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4068.2). Total num frames: 3313664. Throughput: 0: 1027.2. Samples: 828866. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:04:38,956][02158] Avg episode reward: [(0, '18.731')] [2025-08-21 20:04:39,115][11386] Updated weights for policy 0, policy_version 810 (0.0026) [2025-08-21 20:04:43,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4082.1). Total num frames: 3338240. Throughput: 0: 1034.8. Samples: 834494. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:04:43,953][02158] Avg episode reward: [(0, '19.464')] [2025-08-21 20:04:48,085][11386] Updated weights for policy 0, policy_version 820 (0.0016) [2025-08-21 20:04:48,952][02158] Fps is (10 sec: 4915.2, 60 sec: 4164.3, 300 sec: 4082.1). Total num frames: 3362816. Throughput: 0: 1039.7. Samples: 838094. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 20:04:48,953][02158] Avg episode reward: [(0, '17.473')] [2025-08-21 20:04:53,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.4, 300 sec: 4082.1). Total num frames: 3379200. Throughput: 0: 1037.5. Samples: 844542. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 20:04:53,954][02158] Avg episode reward: [(0, '16.843')] [2025-08-21 20:04:58,696][11386] Updated weights for policy 0, policy_version 830 (0.0017) [2025-08-21 20:04:58,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4082.1). Total num frames: 3399680. Throughput: 0: 1039.6. Samples: 850156. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 20:04:58,957][02158] Avg episode reward: [(0, '16.699')] [2025-08-21 20:05:03,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4232.6, 300 sec: 4082.1). Total num frames: 3424256. Throughput: 0: 1039.8. Samples: 853812. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 20:05:03,955][02158] Avg episode reward: [(0, '16.793')] [2025-08-21 20:05:08,478][11386] Updated weights for policy 0, policy_version 840 (0.0011) [2025-08-21 20:05:08,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.6, 300 sec: 4082.1). Total num frames: 3440640. Throughput: 0: 1035.0. Samples: 860114. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:05:08,955][02158] Avg episode reward: [(0, '18.635')] [2025-08-21 20:05:13,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4068.2). Total num frames: 3461120. Throughput: 0: 1044.5. Samples: 866012. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 20:05:13,957][02158] Avg episode reward: [(0, '18.603')] [2025-08-21 20:05:17,799][11386] Updated weights for policy 0, policy_version 850 (0.0013) [2025-08-21 20:05:18,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4082.1). Total num frames: 3485696. Throughput: 0: 1041.2. Samples: 869408. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:05:18,957][02158] Avg episode reward: [(0, '18.628')] [2025-08-21 20:05:23,955][02158] Fps is (10 sec: 4094.7, 60 sec: 4095.9, 300 sec: 4082.1). Total num frames: 3502080. Throughput: 0: 1040.5. Samples: 875690. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:05:23,957][02158] Avg episode reward: [(0, '17.757')] [2025-08-21 20:05:28,053][11386] Updated weights for policy 0, policy_version 860 (0.0013) [2025-08-21 20:05:28,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4082.1). Total num frames: 3522560. Throughput: 0: 1049.4. Samples: 881716. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 20:05:28,953][02158] Avg episode reward: [(0, '18.858')] [2025-08-21 20:05:33,958][02158] Fps is (10 sec: 4094.9, 60 sec: 4095.6, 300 sec: 4082.1). Total num frames: 3543040. Throughput: 0: 1045.6. Samples: 885154. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 20:05:33,960][02158] Avg episode reward: [(0, '18.018')] [2025-08-21 20:05:38,956][02158] Fps is (10 sec: 3275.5, 60 sec: 4027.5, 300 sec: 4054.3). Total num frames: 3555328. Throughput: 0: 994.4. Samples: 889294. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:05:38,957][02158] Avg episode reward: [(0, '19.531')] [2025-08-21 20:05:40,711][11386] Updated weights for policy 0, policy_version 870 (0.0033) [2025-08-21 20:05:43,952][02158] Fps is (10 sec: 3278.7, 60 sec: 3959.5, 300 sec: 4054.3). Total num frames: 3575808. Throughput: 0: 986.5. Samples: 894550. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-08-21 20:05:43,957][02158] Avg episode reward: [(0, '19.468')] [2025-08-21 20:05:48,952][02158] Fps is (10 sec: 4507.2, 60 sec: 3959.4, 300 sec: 4054.3). Total num frames: 3600384. Throughput: 0: 984.7. Samples: 898122. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:05:48,958][02158] Avg episode reward: [(0, '20.806')] [2025-08-21 20:05:48,970][11373] Saving new best policy, reward=20.806! [2025-08-21 20:05:49,506][11386] Updated weights for policy 0, policy_version 880 (0.0019) [2025-08-21 20:05:53,952][02158] Fps is (10 sec: 4096.0, 60 sec: 3959.5, 300 sec: 4054.3). Total num frames: 3616768. Throughput: 0: 987.1. Samples: 904532. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:05:53,956][02158] Avg episode reward: [(0, '21.748')] [2025-08-21 20:05:53,957][11373] Saving new best policy, reward=21.748! [2025-08-21 20:05:58,952][02158] Fps is (10 sec: 3686.5, 60 sec: 3959.5, 300 sec: 4054.3). Total num frames: 3637248. Throughput: 0: 977.7. Samples: 910010. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 20:05:58,953][02158] Avg episode reward: [(0, '21.998')] [2025-08-21 20:05:58,958][11373] Saving new best policy, reward=21.998! [2025-08-21 20:06:00,226][11386] Updated weights for policy 0, policy_version 890 (0.0015) [2025-08-21 20:06:03,952][02158] Fps is (10 sec: 4505.6, 60 sec: 3959.5, 300 sec: 4068.2). Total num frames: 3661824. Throughput: 0: 980.5. Samples: 913530. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) [2025-08-21 20:06:03,953][02158] Avg episode reward: [(0, '20.701')] [2025-08-21 20:06:08,952][02158] Fps is (10 sec: 4096.0, 60 sec: 3959.5, 300 sec: 4054.3). Total num frames: 3678208. Throughput: 0: 981.4. Samples: 919848. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 20:06:08,953][02158] Avg episode reward: [(0, '19.266')] [2025-08-21 20:06:10,628][11386] Updated weights for policy 0, policy_version 900 (0.0013) [2025-08-21 20:06:13,952][02158] Fps is (10 sec: 3686.4, 60 sec: 3959.5, 300 sec: 4054.3). Total num frames: 3698688. Throughput: 0: 978.1. Samples: 925730. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-08-21 20:06:13,956][02158] Avg episode reward: [(0, '20.564')] [2025-08-21 20:06:18,952][02158] Fps is (10 sec: 4505.6, 60 sec: 3959.5, 300 sec: 4082.1). Total num frames: 3723264. Throughput: 0: 982.3. Samples: 929352. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:06:18,953][02158] Avg episode reward: [(0, '20.301')] [2025-08-21 20:06:19,120][11386] Updated weights for policy 0, policy_version 910 (0.0018) [2025-08-21 20:06:23,952][02158] Fps is (10 sec: 4505.4, 60 sec: 4027.9, 300 sec: 4109.9). Total num frames: 3743744. Throughput: 0: 1026.9. Samples: 935502. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:06:23,961][02158] Avg episode reward: [(0, '20.050')] [2025-08-21 20:06:28,961][02158] Fps is (10 sec: 4092.4, 60 sec: 4027.1, 300 sec: 4095.9). Total num frames: 3764224. Throughput: 0: 1042.7. Samples: 941482. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) [2025-08-21 20:06:28,962][02158] Avg episode reward: [(0, '21.715')] [2025-08-21 20:06:28,970][11373] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000919_3764224.pth... [2025-08-21 20:06:29,077][11373] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000912_3735552.pth [2025-08-21 20:06:29,602][11386] Updated weights for policy 0, policy_version 920 (0.0017) [2025-08-21 20:06:33,952][02158] Fps is (10 sec: 4505.8, 60 sec: 4096.4, 300 sec: 4096.0). Total num frames: 3788800. Throughput: 0: 1043.5. Samples: 945078. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:06:33,957][02158] Avg episode reward: [(0, '22.786')] [2025-08-21 20:06:33,962][11373] Saving new best policy, reward=22.786! [2025-08-21 20:06:38,952][02158] Fps is (10 sec: 4099.6, 60 sec: 4164.5, 300 sec: 4109.9). Total num frames: 3805184. Throughput: 0: 1029.9. Samples: 950876. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:06:38,953][02158] Avg episode reward: [(0, '22.562')] [2025-08-21 20:06:40,051][11386] Updated weights for policy 0, policy_version 930 (0.0030) [2025-08-21 20:06:43,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4096.0). Total num frames: 3825664. Throughput: 0: 1037.8. Samples: 956710. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:06:43,957][02158] Avg episode reward: [(0, '22.332')] [2025-08-21 20:06:48,840][11386] Updated weights for policy 0, policy_version 940 (0.0016) [2025-08-21 20:06:48,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4096.0). Total num frames: 3850240. Throughput: 0: 1038.2. Samples: 960248. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:06:48,957][02158] Avg episode reward: [(0, '21.134')] [2025-08-21 20:06:53,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4096.0). Total num frames: 3862528. Throughput: 0: 1028.2. Samples: 966116. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:06:53,956][02158] Avg episode reward: [(0, '19.595')] [2025-08-21 20:06:58,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4096.0). Total num frames: 3887104. Throughput: 0: 1039.2. Samples: 972496. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:06:58,953][02158] Avg episode reward: [(0, '19.782')] [2025-08-21 20:06:59,124][11386] Updated weights for policy 0, policy_version 950 (0.0013) [2025-08-21 20:07:03,952][02158] Fps is (10 sec: 4915.1, 60 sec: 4164.3, 300 sec: 4096.0). Total num frames: 3911680. Throughput: 0: 1039.7. Samples: 976138. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:07:03,956][02158] Avg episode reward: [(0, '19.679')] [2025-08-21 20:07:08,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 3928064. Throughput: 0: 1030.1. Samples: 981858. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:07:08,955][02158] Avg episode reward: [(0, '20.115')] [2025-08-21 20:07:09,477][11386] Updated weights for policy 0, policy_version 960 (0.0021) [2025-08-21 20:07:13,952][02158] Fps is (10 sec: 3686.5, 60 sec: 4164.3, 300 sec: 4096.0). Total num frames: 3948544. Throughput: 0: 1036.0. Samples: 988094. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:07:13,956][02158] Avg episode reward: [(0, '22.514')] [2025-08-21 20:07:18,352][11386] Updated weights for policy 0, policy_version 970 (0.0015) [2025-08-21 20:07:18,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4096.0). Total num frames: 3973120. Throughput: 0: 1034.3. Samples: 991622. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:07:18,953][02158] Avg episode reward: [(0, '23.073')] [2025-08-21 20:07:18,960][11373] Saving new best policy, reward=23.073! [2025-08-21 20:07:23,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 3989504. Throughput: 0: 1028.8. Samples: 997170. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 20:07:23,953][02158] Avg episode reward: [(0, '23.472')] [2025-08-21 20:07:23,954][11373] Saving new best policy, reward=23.472! [2025-08-21 20:07:28,839][11386] Updated weights for policy 0, policy_version 980 (0.0015) [2025-08-21 20:07:28,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.9, 300 sec: 4109.9). Total num frames: 4014080. Throughput: 0: 1045.7. Samples: 1003768. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:07:28,953][02158] Avg episode reward: [(0, '24.309')] [2025-08-21 20:07:28,959][11373] Saving new best policy, reward=24.309! [2025-08-21 20:07:33,952][02158] Fps is (10 sec: 4505.4, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 4034560. Throughput: 0: 1040.9. Samples: 1007088. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:07:33,957][02158] Avg episode reward: [(0, '24.210')] [2025-08-21 20:07:38,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4096.0). Total num frames: 4050944. Throughput: 0: 1030.7. Samples: 1012498. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:07:38,961][02158] Avg episode reward: [(0, '23.768')] [2025-08-21 20:07:39,373][11386] Updated weights for policy 0, policy_version 990 (0.0018) [2025-08-21 20:07:43,952][02158] Fps is (10 sec: 4096.2, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 4075520. Throughput: 0: 1034.5. Samples: 1019050. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:07:43,958][02158] Avg episode reward: [(0, '22.798')] [2025-08-21 20:07:48,058][11386] Updated weights for policy 0, policy_version 1000 (0.0018) [2025-08-21 20:07:48,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.0, 300 sec: 4096.0). Total num frames: 4096000. Throughput: 0: 1033.8. Samples: 1022658. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:07:48,953][02158] Avg episode reward: [(0, '23.291')] [2025-08-21 20:07:53,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4232.5, 300 sec: 4109.9). Total num frames: 4116480. Throughput: 0: 1029.6. Samples: 1028190. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:07:53,954][02158] Avg episode reward: [(0, '23.355')] [2025-08-21 20:07:58,495][11386] Updated weights for policy 0, policy_version 1010 (0.0033) [2025-08-21 20:07:58,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 4136960. Throughput: 0: 1036.2. Samples: 1034722. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:07:58,957][02158] Avg episode reward: [(0, '23.236')] [2025-08-21 20:08:03,954][02158] Fps is (10 sec: 4504.5, 60 sec: 4164.1, 300 sec: 4123.7). Total num frames: 4161536. Throughput: 0: 1037.5. Samples: 1038310. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:08:03,957][02158] Avg episode reward: [(0, '23.880')] [2025-08-21 20:08:08,907][11386] Updated weights for policy 0, policy_version 1020 (0.0020) [2025-08-21 20:08:08,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 4177920. Throughput: 0: 1032.5. Samples: 1043634. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 20:08:08,956][02158] Avg episode reward: [(0, '24.317')] [2025-08-21 20:08:08,966][11373] Saving new best policy, reward=24.317! [2025-08-21 20:08:13,956][02158] Fps is (10 sec: 3685.9, 60 sec: 4164.0, 300 sec: 4109.8). Total num frames: 4198400. Throughput: 0: 1035.1. Samples: 1050352. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:08:13,958][02158] Avg episode reward: [(0, '23.078')] [2025-08-21 20:08:17,570][11386] Updated weights for policy 0, policy_version 1030 (0.0017) [2025-08-21 20:08:18,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 4222976. Throughput: 0: 1043.3. Samples: 1054034. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:08:18,959][02158] Avg episode reward: [(0, '22.987')] [2025-08-21 20:08:23,952][02158] Fps is (10 sec: 4097.6, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 4239360. Throughput: 0: 1036.4. Samples: 1059136. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:08:23,954][02158] Avg episode reward: [(0, '23.026')] [2025-08-21 20:08:27,942][11386] Updated weights for policy 0, policy_version 1040 (0.0015) [2025-08-21 20:08:28,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 4263936. Throughput: 0: 1046.2. Samples: 1066128. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:08:28,953][02158] Avg episode reward: [(0, '22.227')] [2025-08-21 20:08:28,964][11373] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001041_4263936.pth... [2025-08-21 20:08:29,074][11373] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000919_3764224.pth [2025-08-21 20:08:33,952][02158] Fps is (10 sec: 4505.4, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 4284416. Throughput: 0: 1043.3. Samples: 1069608. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:08:33,956][02158] Avg episode reward: [(0, '22.211')] [2025-08-21 20:08:38,525][11386] Updated weights for policy 0, policy_version 1050 (0.0017) [2025-08-21 20:08:38,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 4300800. Throughput: 0: 1031.4. Samples: 1074604. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:08:38,953][02158] Avg episode reward: [(0, '22.115')] [2025-08-21 20:08:43,952][02158] Fps is (10 sec: 4096.1, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 4325376. Throughput: 0: 1042.0. Samples: 1081610. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:08:43,953][02158] Avg episode reward: [(0, '22.268')] [2025-08-21 20:08:47,058][11386] Updated weights for policy 0, policy_version 1060 (0.0019) [2025-08-21 20:08:48,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 4345856. Throughput: 0: 1042.3. Samples: 1085210. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:08:48,956][02158] Avg episode reward: [(0, '21.926')] [2025-08-21 20:08:53,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 4362240. Throughput: 0: 1037.5. Samples: 1090320. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:08:53,954][02158] Avg episode reward: [(0, '22.616')] [2025-08-21 20:08:57,595][11386] Updated weights for policy 0, policy_version 1070 (0.0017) [2025-08-21 20:08:58,954][02158] Fps is (10 sec: 4095.2, 60 sec: 4164.1, 300 sec: 4123.7). Total num frames: 4386816. Throughput: 0: 1043.4. Samples: 1097302. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:08:58,955][02158] Avg episode reward: [(0, '24.177')] [2025-08-21 20:09:03,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.2, 300 sec: 4123.8). Total num frames: 4407296. Throughput: 0: 1038.7. Samples: 1100776. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:09:03,959][02158] Avg episode reward: [(0, '25.300')] [2025-08-21 20:09:03,960][11373] Saving new best policy, reward=25.300! [2025-08-21 20:09:08,178][11386] Updated weights for policy 0, policy_version 1080 (0.0017) [2025-08-21 20:09:08,952][02158] Fps is (10 sec: 3687.2, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 4423680. Throughput: 0: 1031.8. Samples: 1105566. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:09:08,953][02158] Avg episode reward: [(0, '25.976')] [2025-08-21 20:09:08,975][11373] Saving new best policy, reward=25.976! [2025-08-21 20:09:13,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.5, 300 sec: 4109.9). Total num frames: 4448256. Throughput: 0: 1036.7. Samples: 1112778. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:09:13,957][02158] Avg episode reward: [(0, '25.579')] [2025-08-21 20:09:16,877][11386] Updated weights for policy 0, policy_version 1090 (0.0012) [2025-08-21 20:09:18,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 4468736. Throughput: 0: 1037.8. Samples: 1116308. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:09:18,953][02158] Avg episode reward: [(0, '25.280')] [2025-08-21 20:09:23,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 4485120. Throughput: 0: 1031.7. Samples: 1121032. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:09:23,953][02158] Avg episode reward: [(0, '26.251')] [2025-08-21 20:09:24,003][11373] Saving new best policy, reward=26.251! [2025-08-21 20:09:27,646][11386] Updated weights for policy 0, policy_version 1100 (0.0023) [2025-08-21 20:09:28,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 4509696. Throughput: 0: 1028.3. Samples: 1127884. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:09:28,957][02158] Avg episode reward: [(0, '25.822')] [2025-08-21 20:09:33,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.0, 300 sec: 4123.8). Total num frames: 4530176. Throughput: 0: 1028.5. Samples: 1131492. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:09:33,955][02158] Avg episode reward: [(0, '26.183')] [2025-08-21 20:09:38,952][02158] Fps is (10 sec: 3276.8, 60 sec: 4027.7, 300 sec: 4082.1). Total num frames: 4542464. Throughput: 0: 1016.9. Samples: 1136082. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:09:38,953][02158] Avg episode reward: [(0, '25.250')] [2025-08-21 20:09:39,313][11386] Updated weights for policy 0, policy_version 1110 (0.0017) [2025-08-21 20:09:43,952][02158] Fps is (10 sec: 3276.7, 60 sec: 3959.4, 300 sec: 4068.2). Total num frames: 4562944. Throughput: 0: 978.3. Samples: 1141326. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:09:43,954][02158] Avg episode reward: [(0, '25.842')] [2025-08-21 20:09:48,863][11386] Updated weights for policy 0, policy_version 1120 (0.0015) [2025-08-21 20:09:48,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4027.7, 300 sec: 4096.0). Total num frames: 4587520. Throughput: 0: 982.5. Samples: 1144990. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:09:48,956][02158] Avg episode reward: [(0, '24.291')] [2025-08-21 20:09:53,952][02158] Fps is (10 sec: 4096.1, 60 sec: 4027.7, 300 sec: 4082.1). Total num frames: 4603904. Throughput: 0: 986.1. Samples: 1149942. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 20:09:53,953][02158] Avg episode reward: [(0, '22.924')] [2025-08-21 20:09:58,866][11386] Updated weights for policy 0, policy_version 1130 (0.0013) [2025-08-21 20:09:58,952][02158] Fps is (10 sec: 4095.9, 60 sec: 4027.9, 300 sec: 4082.1). Total num frames: 4628480. Throughput: 0: 984.2. Samples: 1157066. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:09:58,956][02158] Avg episode reward: [(0, '24.246')] [2025-08-21 20:10:03,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4027.7, 300 sec: 4096.0). Total num frames: 4648960. Throughput: 0: 985.6. Samples: 1160658. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-08-21 20:10:03,957][02158] Avg episode reward: [(0, '24.179')] [2025-08-21 20:10:08,952][02158] Fps is (10 sec: 3686.5, 60 sec: 4027.7, 300 sec: 4082.1). Total num frames: 4665344. Throughput: 0: 986.6. Samples: 1165430. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:10:08,953][02158] Avg episode reward: [(0, '23.789')] [2025-08-21 20:10:09,549][11386] Updated weights for policy 0, policy_version 1140 (0.0016) [2025-08-21 20:10:13,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.7, 300 sec: 4082.1). Total num frames: 4689920. Throughput: 0: 993.0. Samples: 1172570. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:10:13,957][02158] Avg episode reward: [(0, '25.649')] [2025-08-21 20:10:18,417][11386] Updated weights for policy 0, policy_version 1150 (0.0020) [2025-08-21 20:10:18,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4027.7, 300 sec: 4096.0). Total num frames: 4710400. Throughput: 0: 994.2. Samples: 1176230. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:10:18,958][02158] Avg episode reward: [(0, '25.748')] [2025-08-21 20:10:23,952][02158] Fps is (10 sec: 3686.3, 60 sec: 4027.7, 300 sec: 4082.1). Total num frames: 4726784. Throughput: 0: 1004.5. Samples: 1181284. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:10:23,954][02158] Avg episode reward: [(0, '24.591')] [2025-08-21 20:10:28,625][11386] Updated weights for policy 0, policy_version 1160 (0.0018) [2025-08-21 20:10:28,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4027.7, 300 sec: 4096.1). Total num frames: 4751360. Throughput: 0: 1044.8. Samples: 1188340. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 20:10:28,953][02158] Avg episode reward: [(0, '23.024')] [2025-08-21 20:10:28,960][11373] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001160_4751360.pth... [2025-08-21 20:10:29,066][11373] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth [2025-08-21 20:10:33,952][02158] Fps is (10 sec: 4505.7, 60 sec: 4027.7, 300 sec: 4123.8). Total num frames: 4771840. Throughput: 0: 1042.4. Samples: 1191896. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:10:33,953][02158] Avg episode reward: [(0, '23.460')] [2025-08-21 20:10:38,952][02158] Fps is (10 sec: 3686.4, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 4788224. Throughput: 0: 1034.4. Samples: 1196488. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:10:38,957][02158] Avg episode reward: [(0, '22.489')] [2025-08-21 20:10:39,280][11386] Updated weights for policy 0, policy_version 1170 (0.0012) [2025-08-21 20:10:43,952][02158] Fps is (10 sec: 4095.9, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 4812800. Throughput: 0: 1036.0. Samples: 1203684. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:10:43,956][02158] Avg episode reward: [(0, '22.847')] [2025-08-21 20:10:48,301][11386] Updated weights for policy 0, policy_version 1180 (0.0015) [2025-08-21 20:10:48,952][02158] Fps is (10 sec: 4505.6, 60 sec: 4096.0, 300 sec: 4123.8). Total num frames: 4833280. Throughput: 0: 1035.2. Samples: 1207242. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-08-21 20:10:48,953][02158] Avg episode reward: [(0, '22.707')] [2025-08-21 20:10:53,952][02158] Fps is (10 sec: 4096.1, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 4853760. Throughput: 0: 1041.0. Samples: 1212274. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:10:53,953][02158] Avg episode reward: [(0, '24.260')] [2025-08-21 20:10:58,357][11386] Updated weights for policy 0, policy_version 1190 (0.0028) [2025-08-21 20:10:58,952][02158] Fps is (10 sec: 4095.8, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 4874240. Throughput: 0: 1035.2. Samples: 1219154. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:10:58,957][02158] Avg episode reward: [(0, '24.211')] [2025-08-21 20:11:03,955][02158] Fps is (10 sec: 4094.8, 60 sec: 4095.8, 300 sec: 4123.7). Total num frames: 4894720. Throughput: 0: 1033.0. Samples: 1222720. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:11:03,959][02158] Avg episode reward: [(0, '25.685')] [2025-08-21 20:11:08,873][11386] Updated weights for policy 0, policy_version 1200 (0.0029) [2025-08-21 20:11:08,952][02158] Fps is (10 sec: 4096.2, 60 sec: 4164.3, 300 sec: 4123.8). Total num frames: 4915200. Throughput: 0: 1030.9. Samples: 1227674. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:11:08,953][02158] Avg episode reward: [(0, '26.252')] [2025-08-21 20:11:08,958][11373] Saving new best policy, reward=26.252! [2025-08-21 20:11:13,952][02158] Fps is (10 sec: 4097.2, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 4935680. Throughput: 0: 1032.0. Samples: 1234782. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:11:13,958][02158] Avg episode reward: [(0, '26.092')] [2025-08-21 20:11:17,909][11386] Updated weights for policy 0, policy_version 1210 (0.0024) [2025-08-21 20:11:18,953][02158] Fps is (10 sec: 4095.7, 60 sec: 4096.0, 300 sec: 4109.9). Total num frames: 4956160. Throughput: 0: 1034.3. Samples: 1238440. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-08-21 20:11:18,954][02158] Avg episode reward: [(0, '26.419')] [2025-08-21 20:11:18,962][11373] Saving new best policy, reward=26.419! [2025-08-21 20:11:23,952][02158] Fps is (10 sec: 4096.0, 60 sec: 4164.3, 300 sec: 4110.0). Total num frames: 4976640. Throughput: 0: 1040.6. Samples: 1243316. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-08-21 20:11:23,959][02158] Avg episode reward: [(0, '25.139')] [2025-08-21 20:11:27,993][11386] Updated weights for policy 0, policy_version 1220 (0.0013) [2025-08-21 20:11:28,952][02158] Fps is (10 sec: 4505.9, 60 sec: 4164.3, 300 sec: 4109.9). Total num frames: 5001216. Throughput: 0: 1037.2. Samples: 1250360. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-08-21 20:11:28,953][02158] Avg episode reward: [(0, '24.128')] [2025-08-21 20:11:29,733][02158] Component Batcher_0 stopped! [2025-08-21 20:11:29,733][11373] Stopping Batcher_0... [2025-08-21 20:11:29,738][11373] Loop batcher_evt_loop terminating... [2025-08-21 20:11:29,740][11373] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:11:29,792][11386] Weights refcount: 2 0 [2025-08-21 20:11:29,797][11386] Stopping InferenceWorker_p0-w0... [2025-08-21 20:11:29,797][02158] Component InferenceWorker_p0-w0 stopped! [2025-08-21 20:11:29,803][11386] Loop inference_proc0-0_evt_loop terminating... [2025-08-21 20:11:29,841][11373] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001041_4263936.pth [2025-08-21 20:11:29,857][11373] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:11:30,014][11373] Stopping LearnerWorker_p0... [2025-08-21 20:11:30,014][11373] Loop learner_proc0_evt_loop terminating... [2025-08-21 20:11:30,017][02158] Component LearnerWorker_p0 stopped! [2025-08-21 20:11:30,164][02158] Component RolloutWorker_w4 stopped! [2025-08-21 20:11:30,166][11394] Stopping RolloutWorker_w4... [2025-08-21 20:11:30,166][11394] Loop rollout_proc4_evt_loop terminating... [2025-08-21 20:11:30,178][02158] Component RolloutWorker_w5 stopped! [2025-08-21 20:11:30,178][11391] Stopping RolloutWorker_w5... [2025-08-21 20:11:30,185][02158] Component RolloutWorker_w3 stopped! [2025-08-21 20:11:30,185][11390] Stopping RolloutWorker_w3... [2025-08-21 20:11:30,183][11391] Loop rollout_proc5_evt_loop terminating... [2025-08-21 20:11:30,192][11390] Loop rollout_proc3_evt_loop terminating... [2025-08-21 20:11:30,201][02158] Component RolloutWorker_w1 stopped! [2025-08-21 20:11:30,206][11388] Stopping RolloutWorker_w1... [2025-08-21 20:11:30,209][11388] Loop rollout_proc1_evt_loop terminating... [2025-08-21 20:11:30,226][11389] Stopping RolloutWorker_w2... [2025-08-21 20:11:30,227][11389] Loop rollout_proc2_evt_loop terminating... [2025-08-21 20:11:30,226][02158] Component RolloutWorker_w2 stopped! [2025-08-21 20:11:30,250][11393] Stopping RolloutWorker_w6... [2025-08-21 20:11:30,251][11393] Loop rollout_proc6_evt_loop terminating... [2025-08-21 20:11:30,250][02158] Component RolloutWorker_w6 stopped! [2025-08-21 20:11:30,264][02158] Component RolloutWorker_w7 stopped! [2025-08-21 20:11:30,264][11392] Stopping RolloutWorker_w7... [2025-08-21 20:11:30,270][11392] Loop rollout_proc7_evt_loop terminating... [2025-08-21 20:11:30,273][02158] Component RolloutWorker_w0 stopped! [2025-08-21 20:11:30,276][02158] Waiting for process learner_proc0 to stop... [2025-08-21 20:11:30,279][11387] Stopping RolloutWorker_w0... [2025-08-21 20:11:30,283][11387] Loop rollout_proc0_evt_loop terminating... [2025-08-21 20:11:31,496][02158] Waiting for process inference_proc0-0 to join... [2025-08-21 20:11:31,502][02158] Waiting for process rollout_proc0 to join... [2025-08-21 20:11:34,602][02158] Waiting for process rollout_proc1 to join... [2025-08-21 20:11:34,603][02158] Waiting for process rollout_proc2 to join... [2025-08-21 20:11:34,604][02158] Waiting for process rollout_proc3 to join... [2025-08-21 20:11:34,605][02158] Waiting for process rollout_proc4 to join... [2025-08-21 20:11:34,606][02158] Waiting for process rollout_proc5 to join... [2025-08-21 20:11:34,607][02158] Waiting for process rollout_proc6 to join... [2025-08-21 20:11:34,608][02158] Waiting for process rollout_proc7 to join... [2025-08-21 20:11:34,609][02158] Batcher 0 profile tree view: batching: 33.5809, releasing_batches: 0.0323 [2025-08-21 20:11:34,610][02158] InferenceWorker_p0-w0 profile tree view: wait_policy: 0.0035 wait_policy_total: 535.4931 update_model: 9.3376 weight_update: 0.0020 one_step: 0.0025 handle_policy_step: 651.5964 deserialize: 17.3898, stack: 3.3949, obs_to_device_normalize: 139.9688, forward: 329.0427, send_messages: 33.6492 prepare_outputs: 99.5629 to_cpu: 62.1138 [2025-08-21 20:11:34,611][02158] Learner 0 profile tree view: misc: 0.0062, prepare_batch: 14.9270 train: 89.6135 epoch_init: 0.0065, minibatch_init: 0.0145, losses_postprocess: 0.7535, kl_divergence: 0.8283, after_optimizer: 41.6698 calculate_losses: 31.1672 losses_init: 0.0040, forward_head: 1.5524, bptt_initial: 20.6017, tail: 1.1391, advantages_returns: 0.3993, losses: 4.4069 bptt: 2.7556 bptt_forward_core: 2.6089 update: 14.5035 clip: 1.1377 [2025-08-21 20:11:34,613][02158] RolloutWorker_w0 profile tree view: wait_for_trajectories: 0.2983, enqueue_policy_requests: 133.6954, env_step: 979.3261, overhead: 14.3485, complete_rollouts: 9.0930 save_policy_outputs: 22.6923 split_output_tensors: 9.0335 [2025-08-21 20:11:34,614][02158] RolloutWorker_w7 profile tree view: wait_for_trajectories: 0.3712, enqueue_policy_requests: 143.4527, env_step: 967.5485, overhead: 15.0307, complete_rollouts: 7.7438 save_policy_outputs: 21.7696 split_output_tensors: 8.1752 [2025-08-21 20:11:34,614][02158] Loop Runner_EvtLoop terminating... [2025-08-21 20:11:34,619][02158] Runner profile tree view: main_loop: 1263.4643 [2025-08-21 20:11:34,619][02158] Collected {0: 5005312}, FPS: 3961.6 [2025-08-21 20:11:34,645][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 20:11:34,646][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 20:11:34,647][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 20:11:34,648][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 20:11:34,649][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 20:11:34,650][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 20:11:34,651][02158] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 20:11:34,653][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 20:11:34,654][02158] Adding new argument 'push_to_hub'=False that is not in the saved config file! [2025-08-21 20:11:34,656][02158] Adding new argument 'hf_repository'=None that is not in the saved config file! [2025-08-21 20:11:34,657][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 20:11:34,661][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 20:11:34,661][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 20:11:34,662][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 20:11:34,663][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 20:11:34,709][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 20:11:34,711][02158] RunningMeanStd input shape: (1,) [2025-08-21 20:11:34,726][02158] ConvEncoder: input_channels=3 [2025-08-21 20:11:34,786][02158] Conv encoder output size: 512 [2025-08-21 20:11:34,787][02158] Policy head output size: 512 [2025-08-21 20:11:34,814][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:11:35,234][02158] Num frames 100... [2025-08-21 20:11:35,351][02158] Num frames 200... [2025-08-21 20:11:35,468][02158] Num frames 300... [2025-08-21 20:11:35,591][02158] Num frames 400... [2025-08-21 20:11:35,707][02158] Num frames 500... [2025-08-21 20:11:35,833][02158] Num frames 600... [2025-08-21 20:11:35,950][02158] Num frames 700... [2025-08-21 20:11:36,069][02158] Num frames 800... [2025-08-21 20:11:36,187][02158] Num frames 900... [2025-08-21 20:11:36,306][02158] Num frames 1000... [2025-08-21 20:11:36,463][02158] Num frames 1100... [2025-08-21 20:11:36,583][02158] Num frames 1200... [2025-08-21 20:11:36,703][02158] Num frames 1300... [2025-08-21 20:11:36,821][02158] Num frames 1400... [2025-08-21 20:11:36,951][02158] Num frames 1500... [2025-08-21 20:11:37,068][02158] Num frames 1600... [2025-08-21 20:11:37,185][02158] Num frames 1700... [2025-08-21 20:11:37,309][02158] Num frames 1800... [2025-08-21 20:11:37,451][02158] Num frames 1900... [2025-08-21 20:11:37,569][02158] Num frames 2000... [2025-08-21 20:11:37,695][02158] Num frames 2100... [2025-08-21 20:11:37,747][02158] Avg episode rewards: #0: 57.999, true rewards: #0: 21.000 [2025-08-21 20:11:37,748][02158] Avg episode reward: 57.999, avg true_objective: 21.000 [2025-08-21 20:11:37,864][02158] Num frames 2200... [2025-08-21 20:11:37,993][02158] Num frames 2300... [2025-08-21 20:11:38,109][02158] Num frames 2400... [2025-08-21 20:11:38,232][02158] Num frames 2500... [2025-08-21 20:11:38,382][02158] Avg episode rewards: #0: 32.899, true rewards: #0: 12.900 [2025-08-21 20:11:38,383][02158] Avg episode reward: 32.899, avg true_objective: 12.900 [2025-08-21 20:11:38,409][02158] Num frames 2600... [2025-08-21 20:11:38,529][02158] Num frames 2700... [2025-08-21 20:11:38,648][02158] Num frames 2800... [2025-08-21 20:11:38,770][02158] Num frames 2900... [2025-08-21 20:11:38,889][02158] Num frames 3000... [2025-08-21 20:11:39,020][02158] Num frames 3100... [2025-08-21 20:11:39,140][02158] Num frames 3200... [2025-08-21 20:11:39,264][02158] Num frames 3300... [2025-08-21 20:11:39,391][02158] Num frames 3400... [2025-08-21 20:11:39,518][02158] Num frames 3500... [2025-08-21 20:11:39,642][02158] Num frames 3600... [2025-08-21 20:11:39,761][02158] Num frames 3700... [2025-08-21 20:11:39,880][02158] Num frames 3800... [2025-08-21 20:11:40,012][02158] Num frames 3900... [2025-08-21 20:11:40,169][02158] Avg episode rewards: #0: 31.960, true rewards: #0: 13.293 [2025-08-21 20:11:40,170][02158] Avg episode reward: 31.960, avg true_objective: 13.293 [2025-08-21 20:11:40,190][02158] Num frames 4000... [2025-08-21 20:11:40,310][02158] Num frames 4100... [2025-08-21 20:11:40,431][02158] Num frames 4200... [2025-08-21 20:11:40,554][02158] Num frames 4300... [2025-08-21 20:11:40,677][02158] Num frames 4400... [2025-08-21 20:11:40,795][02158] Num frames 4500... [2025-08-21 20:11:40,913][02158] Num frames 4600... [2025-08-21 20:11:41,045][02158] Num frames 4700... [2025-08-21 20:11:41,166][02158] Num frames 4800... [2025-08-21 20:11:41,286][02158] Num frames 4900... [2025-08-21 20:11:41,406][02158] Num frames 5000... [2025-08-21 20:11:41,528][02158] Num frames 5100... [2025-08-21 20:11:41,652][02158] Num frames 5200... [2025-08-21 20:11:41,772][02158] Num frames 5300... [2025-08-21 20:11:41,891][02158] Num frames 5400... [2025-08-21 20:11:42,015][02158] Num frames 5500... [2025-08-21 20:11:42,102][02158] Avg episode rewards: #0: 33.060, true rewards: #0: 13.810 [2025-08-21 20:11:42,103][02158] Avg episode reward: 33.060, avg true_objective: 13.810 [2025-08-21 20:11:42,196][02158] Num frames 5600... [2025-08-21 20:11:42,314][02158] Num frames 5700... [2025-08-21 20:11:42,435][02158] Num frames 5800... [2025-08-21 20:11:42,554][02158] Num frames 5900... [2025-08-21 20:11:42,673][02158] Num frames 6000... [2025-08-21 20:11:42,800][02158] Num frames 6100... [2025-08-21 20:11:42,930][02158] Avg episode rewards: #0: 28.728, true rewards: #0: 12.328 [2025-08-21 20:11:42,931][02158] Avg episode reward: 28.728, avg true_objective: 12.328 [2025-08-21 20:11:42,975][02158] Num frames 6200... [2025-08-21 20:11:43,102][02158] Num frames 6300... [2025-08-21 20:11:43,223][02158] Num frames 6400... [2025-08-21 20:11:43,343][02158] Num frames 6500... [2025-08-21 20:11:43,463][02158] Num frames 6600... [2025-08-21 20:11:43,581][02158] Num frames 6700... [2025-08-21 20:11:43,698][02158] Num frames 6800... [2025-08-21 20:11:43,794][02158] Avg episode rewards: #0: 26.060, true rewards: #0: 11.393 [2025-08-21 20:11:43,795][02158] Avg episode reward: 26.060, avg true_objective: 11.393 [2025-08-21 20:11:43,870][02158] Num frames 6900... [2025-08-21 20:11:43,987][02158] Num frames 7000... [2025-08-21 20:11:44,116][02158] Num frames 7100... [2025-08-21 20:11:44,245][02158] Num frames 7200... [2025-08-21 20:11:44,362][02158] Num frames 7300... [2025-08-21 20:11:44,490][02158] Num frames 7400... [2025-08-21 20:11:44,611][02158] Num frames 7500... [2025-08-21 20:11:44,729][02158] Num frames 7600... [2025-08-21 20:11:44,845][02158] Num frames 7700... [2025-08-21 20:11:44,987][02158] Num frames 7800... [2025-08-21 20:11:45,164][02158] Num frames 7900... [2025-08-21 20:11:45,241][02158] Avg episode rewards: #0: 25.587, true rewards: #0: 11.301 [2025-08-21 20:11:45,242][02158] Avg episode reward: 25.587, avg true_objective: 11.301 [2025-08-21 20:11:45,389][02158] Num frames 8000... [2025-08-21 20:11:45,555][02158] Num frames 8100... [2025-08-21 20:11:45,721][02158] Num frames 8200... [2025-08-21 20:11:45,885][02158] Num frames 8300... [2025-08-21 20:11:46,047][02158] Num frames 8400... [2025-08-21 20:11:46,221][02158] Num frames 8500... [2025-08-21 20:11:46,392][02158] Num frames 8600... [2025-08-21 20:11:46,562][02158] Num frames 8700... [2025-08-21 20:11:46,730][02158] Num frames 8800... [2025-08-21 20:11:46,903][02158] Num frames 8900... [2025-08-21 20:11:47,083][02158] Num frames 9000... [2025-08-21 20:11:47,206][02158] Num frames 9100... [2025-08-21 20:11:47,337][02158] Num frames 9200... [2025-08-21 20:11:47,458][02158] Num frames 9300... [2025-08-21 20:11:47,614][02158] Avg episode rewards: #0: 26.855, true rewards: #0: 11.730 [2025-08-21 20:11:47,615][02158] Avg episode reward: 26.855, avg true_objective: 11.730 [2025-08-21 20:11:47,638][02158] Num frames 9400... [2025-08-21 20:11:47,759][02158] Num frames 9500... [2025-08-21 20:11:47,880][02158] Num frames 9600... [2025-08-21 20:11:48,000][02158] Num frames 9700... [2025-08-21 20:11:48,118][02158] Num frames 9800... [2025-08-21 20:11:48,239][02158] Num frames 9900... [2025-08-21 20:11:48,370][02158] Num frames 10000... [2025-08-21 20:11:48,491][02158] Num frames 10100... [2025-08-21 20:11:48,612][02158] Num frames 10200... [2025-08-21 20:11:48,733][02158] Num frames 10300... [2025-08-21 20:11:48,852][02158] Num frames 10400... [2025-08-21 20:11:49,025][02158] Avg episode rewards: #0: 26.554, true rewards: #0: 11.666 [2025-08-21 20:11:49,026][02158] Avg episode reward: 26.554, avg true_objective: 11.666 [2025-08-21 20:11:49,029][02158] Num frames 10500... [2025-08-21 20:11:49,151][02158] Num frames 10600... [2025-08-21 20:11:49,272][02158] Num frames 10700... [2025-08-21 20:11:49,400][02158] Num frames 10800... [2025-08-21 20:11:49,519][02158] Num frames 10900... [2025-08-21 20:11:49,636][02158] Num frames 11000... [2025-08-21 20:11:49,755][02158] Num frames 11100... [2025-08-21 20:11:49,873][02158] Num frames 11200... [2025-08-21 20:11:49,990][02158] Num frames 11300... [2025-08-21 20:11:50,109][02158] Num frames 11400... [2025-08-21 20:11:50,232][02158] Num frames 11500... [2025-08-21 20:11:50,363][02158] Num frames 11600... [2025-08-21 20:11:50,486][02158] Num frames 11700... [2025-08-21 20:11:50,606][02158] Num frames 11800... [2025-08-21 20:11:50,728][02158] Num frames 11900... [2025-08-21 20:11:50,848][02158] Num frames 12000... [2025-08-21 20:11:50,969][02158] Num frames 12100... [2025-08-21 20:11:51,091][02158] Num frames 12200... [2025-08-21 20:11:51,215][02158] Num frames 12300... [2025-08-21 20:11:51,334][02158] Num frames 12400... [2025-08-21 20:11:51,466][02158] Num frames 12500... [2025-08-21 20:11:51,636][02158] Avg episode rewards: #0: 29.799, true rewards: #0: 12.599 [2025-08-21 20:11:51,637][02158] Avg episode reward: 29.799, avg true_objective: 12.599 [2025-08-21 20:11:51,640][02158] Num frames 12600... [2025-08-21 20:13:08,120][02158] Replay video saved to /content/train_dir/default_experiment/replay.mp4! [2025-08-21 20:13:08,158][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 20:13:08,159][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 20:13:08,160][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 20:13:08,161][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 20:13:08,162][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 20:13:08,164][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 20:13:08,165][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 20:13:08,166][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 20:13:08,167][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 20:13:08,168][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 20:13:08,169][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 20:13:08,170][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 20:13:08,170][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 20:13:08,171][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 20:13:08,172][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 20:13:08,199][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 20:13:08,201][02158] RunningMeanStd input shape: (1,) [2025-08-21 20:13:08,212][02158] ConvEncoder: input_channels=3 [2025-08-21 20:13:08,248][02158] Conv encoder output size: 512 [2025-08-21 20:13:08,249][02158] Policy head output size: 512 [2025-08-21 20:13:08,266][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:13:08,267][02158] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:13:08,291][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:13:08,432][02158] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:13:08,454][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:13:08,456][02158] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:14:02,225][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 20:14:02,226][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 20:14:02,227][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 20:14:02,228][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 20:14:02,229][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 20:14:02,229][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 20:14:02,230][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 20:14:02,231][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 20:14:02,232][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 20:14:02,232][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 20:14:02,234][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 20:14:02,235][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 20:14:02,236][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 20:14:02,237][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 20:14:02,238][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 20:14:02,266][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 20:14:02,267][02158] RunningMeanStd input shape: (1,) [2025-08-21 20:14:02,282][02158] ConvEncoder: input_channels=3 [2025-08-21 20:14:02,316][02158] Conv encoder output size: 512 [2025-08-21 20:14:02,317][02158] Policy head output size: 512 [2025-08-21 20:14:02,333][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:14:02,335][02158] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:14:02,357][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:14:02,359][02158] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:14:02,380][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:14:02,382][02158] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:14:06,535][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 20:14:06,536][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 20:14:06,537][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 20:14:06,537][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 20:14:06,538][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 20:14:06,539][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 20:14:06,540][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 20:14:06,541][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 20:14:06,542][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 20:14:06,543][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 20:14:06,543][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 20:14:06,544][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 20:14:06,545][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 20:14:06,546][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 20:14:06,547][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 20:14:06,572][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 20:14:06,574][02158] RunningMeanStd input shape: (1,) [2025-08-21 20:14:06,582][02158] ConvEncoder: input_channels=3 [2025-08-21 20:14:06,613][02158] Conv encoder output size: 512 [2025-08-21 20:14:06,614][02158] Policy head output size: 512 [2025-08-21 20:14:06,630][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:14:06,633][02158] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:14:06,654][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:14:06,656][02158] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:14:06,678][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:14:06,681][02158] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:15:10,526][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 20:15:10,527][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 20:15:10,528][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 20:15:10,529][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 20:15:10,530][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 20:15:10,531][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 20:15:10,532][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 20:15:10,533][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 20:15:10,534][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 20:15:10,534][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 20:15:10,535][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 20:15:10,536][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 20:15:10,537][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 20:15:10,538][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 20:15:10,539][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 20:15:10,564][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 20:15:10,565][02158] RunningMeanStd input shape: (1,) [2025-08-21 20:15:10,574][02158] ConvEncoder: input_channels=3 [2025-08-21 20:15:10,603][02158] Conv encoder output size: 512 [2025-08-21 20:15:10,603][02158] Policy head output size: 512 [2025-08-21 20:15:10,619][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:15:10,621][02158] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:15:10,642][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:15:10,644][02158] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:15:10,666][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:15:10,667][02158] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:15:35,955][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 20:15:35,956][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 20:15:35,957][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 20:15:35,957][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 20:15:35,958][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 20:15:35,959][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 20:15:35,960][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 20:15:35,961][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 20:15:35,962][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 20:15:35,963][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 20:15:35,964][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 20:15:35,965][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 20:15:35,965][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 20:15:35,966][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 20:15:35,967][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 20:15:35,992][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 20:15:35,993][02158] RunningMeanStd input shape: (1,) [2025-08-21 20:15:36,003][02158] ConvEncoder: input_channels=3 [2025-08-21 20:15:36,032][02158] Conv encoder output size: 512 [2025-08-21 20:15:36,033][02158] Policy head output size: 512 [2025-08-21 20:15:36,049][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:15:36,051][02158] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:15:36,073][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:15:36,074][02158] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:15:36,097][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:15:36,099][02158] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:17:06,806][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 20:17:06,807][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 20:17:06,809][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 20:17:06,810][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 20:17:06,811][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 20:17:06,811][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 20:17:06,812][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 20:17:06,813][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 20:17:06,815][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 20:17:06,815][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 20:17:06,816][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 20:17:06,817][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 20:17:06,818][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 20:17:06,819][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 20:17:06,820][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 20:17:06,843][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 20:17:06,844][02158] RunningMeanStd input shape: (1,) [2025-08-21 20:17:06,853][02158] ConvEncoder: input_channels=3 [2025-08-21 20:17:06,883][02158] Conv encoder output size: 512 [2025-08-21 20:17:06,884][02158] Policy head output size: 512 [2025-08-21 20:17:06,900][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:17:06,902][02158] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:17:06,924][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:17:06,926][02158] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:17:06,947][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:17:06,950][02158] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:18:16,541][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 20:18:16,542][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 20:18:16,543][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 20:18:16,545][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 20:18:16,545][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 20:18:16,546][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 20:18:16,547][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 20:18:16,548][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 20:18:16,549][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 20:18:16,550][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 20:18:16,551][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 20:18:16,551][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 20:18:16,552][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 20:18:16,553][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 20:18:16,554][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 20:18:16,579][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 20:18:16,580][02158] RunningMeanStd input shape: (1,) [2025-08-21 20:18:16,589][02158] ConvEncoder: input_channels=3 [2025-08-21 20:18:16,619][02158] Conv encoder output size: 512 [2025-08-21 20:18:16,620][02158] Policy head output size: 512 [2025-08-21 20:18:16,635][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:18:16,637][02158] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:18:16,658][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:18:16,660][02158] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:18:16,682][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:18:16,684][02158] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:18:40,179][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 20:18:40,180][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 20:18:40,181][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 20:18:40,183][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 20:18:40,184][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 20:18:40,185][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 20:18:40,186][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 20:18:40,187][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 20:18:40,188][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 20:18:40,189][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 20:18:40,190][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 20:18:40,191][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 20:18:40,192][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 20:18:40,192][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 20:18:40,193][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 20:18:40,216][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 20:18:40,217][02158] RunningMeanStd input shape: (1,) [2025-08-21 20:18:40,226][02158] ConvEncoder: input_channels=3 [2025-08-21 20:18:40,255][02158] Conv encoder output size: 512 [2025-08-21 20:18:40,255][02158] Policy head output size: 512 [2025-08-21 20:18:40,271][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:18:40,273][02158] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:18:40,295][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:18:40,298][02158] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:18:40,320][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:18:40,323][02158] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:18:53,859][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 20:18:53,860][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 20:18:53,861][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 20:18:53,862][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 20:18:53,863][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 20:18:53,864][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 20:18:53,865][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 20:18:53,866][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 20:18:53,867][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 20:18:53,868][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 20:18:53,871][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 20:18:53,872][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 20:18:53,873][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 20:18:53,874][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 20:18:53,875][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 20:18:53,909][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 20:18:53,912][02158] RunningMeanStd input shape: (1,) [2025-08-21 20:18:53,922][02158] ConvEncoder: input_channels=3 [2025-08-21 20:18:53,952][02158] Conv encoder output size: 512 [2025-08-21 20:18:53,953][02158] Policy head output size: 512 [2025-08-21 20:18:53,969][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:18:53,971][02158] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:18:53,997][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:18:53,999][02158] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:18:54,020][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:18:54,022][02158] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 973 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:19:09,813][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 20:19:09,814][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 20:19:09,815][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 20:19:09,816][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 20:19:09,817][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 20:19:09,818][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 20:19:09,819][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 20:19:09,820][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 20:19:09,820][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 20:19:09,821][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 20:19:09,822][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 20:19:09,823][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 20:19:09,824][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 20:19:09,825][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 20:19:09,826][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 20:19:09,849][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 20:19:09,850][02158] RunningMeanStd input shape: (1,) [2025-08-21 20:19:09,858][02158] ConvEncoder: input_channels=3 [2025-08-21 20:19:09,887][02158] Conv encoder output size: 512 [2025-08-21 20:19:09,888][02158] Policy head output size: 512 [2025-08-21 20:19:09,903][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:19:09,905][02158] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-844920808.py", line 6, in _torch_load_allow_pickle return _orig_torch_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 972 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:19:09,926][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:19:09,928][02158] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-844920808.py", line 6, in _torch_load_allow_pickle return _orig_torch_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 972 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:19:09,950][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:19:09,953][02158] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/usr/local/lib/python3.12/dist-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-844920808.py", line 6, in _torch_load_allow_pickle return _orig_torch_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/tmp/ipython-input-1477929787.py", line 5, in _load_allow_pickle return _real_load(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ [Previous line repeated 972 more times] RecursionError: maximum recursion depth exceeded [2025-08-21 20:23:04,342][02158] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json [2025-08-21 20:23:04,343][02158] Overriding arg 'num_workers' with value 1 passed from command line [2025-08-21 20:23:04,344][02158] Adding new argument 'no_render'=True that is not in the saved config file! [2025-08-21 20:23:04,345][02158] Adding new argument 'save_video'=True that is not in the saved config file! [2025-08-21 20:23:04,346][02158] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-08-21 20:23:04,347][02158] Adding new argument 'video_name'=None that is not in the saved config file! [2025-08-21 20:23:04,348][02158] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-08-21 20:23:04,348][02158] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-08-21 20:23:04,349][02158] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-08-21 20:23:04,350][02158] Adding new argument 'hf_repository'='jmartin233/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-08-21 20:23:04,353][02158] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-08-21 20:23:04,354][02158] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-08-21 20:23:04,355][02158] Adding new argument 'train_script'=None that is not in the saved config file! [2025-08-21 20:23:04,356][02158] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-08-21 20:23:04,357][02158] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-08-21 20:23:04,380][02158] RunningMeanStd input shape: (3, 72, 128) [2025-08-21 20:23:04,381][02158] RunningMeanStd input shape: (1,) [2025-08-21 20:23:04,390][02158] ConvEncoder: input_channels=3 [2025-08-21 20:23:04,420][02158] Conv encoder output size: 512 [2025-08-21 20:23:04,420][02158] Policy head output size: 512 [2025-08-21 20:23:04,437][02158] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000001222_5005312.pth... [2025-08-21 20:23:04,852][02158] Num frames 100... [2025-08-21 20:23:04,972][02158] Num frames 200... [2025-08-21 20:23:05,090][02158] Num frames 300... [2025-08-21 20:23:05,209][02158] Num frames 400... [2025-08-21 20:23:05,327][02158] Num frames 500... [2025-08-21 20:23:05,444][02158] Num frames 600... [2025-08-21 20:23:05,577][02158] Num frames 700... [2025-08-21 20:23:05,699][02158] Num frames 800... [2025-08-21 20:23:05,819][02158] Num frames 900... [2025-08-21 20:23:05,939][02158] Num frames 1000... [2025-08-21 20:23:06,060][02158] Num frames 1100... [2025-08-21 20:23:06,177][02158] Num frames 1200... [2025-08-21 20:23:06,299][02158] Num frames 1300... [2025-08-21 20:23:06,410][02158] Avg episode rewards: #0: 32.440, true rewards: #0: 13.440 [2025-08-21 20:23:06,412][02158] Avg episode reward: 32.440, avg true_objective: 13.440 [2025-08-21 20:23:06,479][02158] Num frames 1400... [2025-08-21 20:23:06,608][02158] Num frames 1500... [2025-08-21 20:23:06,724][02158] Num frames 1600... [2025-08-21 20:23:06,839][02158] Num frames 1700... [2025-08-21 20:23:06,957][02158] Num frames 1800... [2025-08-21 20:23:07,113][02158] Avg episode rewards: #0: 19.940, true rewards: #0: 9.440 [2025-08-21 20:23:07,114][02158] Avg episode reward: 19.940, avg true_objective: 9.440 [2025-08-21 20:23:07,137][02158] Num frames 1900... [2025-08-21 20:23:07,314][02158] Num frames 2000... [2025-08-21 20:23:07,489][02158] Num frames 2100... [2025-08-21 20:23:07,677][02158] Num frames 2200... [2025-08-21 20:23:07,861][02158] Num frames 2300... [2025-08-21 20:23:08,025][02158] Num frames 2400... [2025-08-21 20:23:08,189][02158] Num frames 2500... [2025-08-21 20:23:08,283][02158] Avg episode rewards: #0: 17.740, true rewards: #0: 8.407 [2025-08-21 20:23:08,285][02158] Avg episode reward: 17.740, avg true_objective: 8.407 [2025-08-21 20:23:08,423][02158] Num frames 2600... [2025-08-21 20:23:08,600][02158] Num frames 2700... [2025-08-21 20:23:08,780][02158] Num frames 2800... [2025-08-21 20:23:08,978][02158] Num frames 2900... [2025-08-21 20:23:09,156][02158] Num frames 3000... [2025-08-21 20:23:09,334][02158] Num frames 3100... [2025-08-21 20:23:09,466][02158] Num frames 3200... [2025-08-21 20:23:09,583][02158] Num frames 3300... [2025-08-21 20:23:09,699][02158] Num frames 3400... [2025-08-21 20:23:09,824][02158] Avg episode rewards: #0: 18.625, true rewards: #0: 8.625 [2025-08-21 20:23:09,825][02158] Avg episode reward: 18.625, avg true_objective: 8.625 [2025-08-21 20:23:09,886][02158] Num frames 3500... [2025-08-21 20:23:10,008][02158] Num frames 3600... [2025-08-21 20:23:10,126][02158] Num frames 3700... [2025-08-21 20:23:10,247][02158] Num frames 3800... [2025-08-21 20:23:10,366][02158] Num frames 3900... [2025-08-21 20:23:10,485][02158] Num frames 4000... [2025-08-21 20:23:10,606][02158] Num frames 4100... [2025-08-21 20:23:10,725][02158] Num frames 4200... [2025-08-21 20:23:10,859][02158] Num frames 4300... [2025-08-21 20:23:10,993][02158] Num frames 4400... [2025-08-21 20:23:11,122][02158] Num frames 4500... [2025-08-21 20:23:11,259][02158] Avg episode rewards: #0: 20.130, true rewards: #0: 9.130 [2025-08-21 20:23:11,260][02158] Avg episode reward: 20.130, avg true_objective: 9.130 [2025-08-21 20:23:11,305][02158] Num frames 4600... [2025-08-21 20:23:11,426][02158] Num frames 4700... [2025-08-21 20:23:11,544][02158] Num frames 4800... [2025-08-21 20:23:11,660][02158] Num frames 4900... [2025-08-21 20:23:11,779][02158] Num frames 5000... [2025-08-21 20:23:11,908][02158] Num frames 5100... [2025-08-21 20:23:12,028][02158] Num frames 5200... [2025-08-21 20:23:12,146][02158] Num frames 5300... [2025-08-21 20:23:12,243][02158] Avg episode rewards: #0: 19.055, true rewards: #0: 8.888 [2025-08-21 20:23:12,244][02158] Avg episode reward: 19.055, avg true_objective: 8.888 [2025-08-21 20:23:12,324][02158] Num frames 5400... [2025-08-21 20:23:12,443][02158] Num frames 5500... [2025-08-21 20:23:12,563][02158] Num frames 5600... [2025-08-21 20:23:12,682][02158] Num frames 5700... [2025-08-21 20:23:12,801][02158] Num frames 5800... [2025-08-21 20:23:12,932][02158] Num frames 5900... [2025-08-21 20:23:13,053][02158] Num frames 6000... [2025-08-21 20:23:13,178][02158] Num frames 6100... [2025-08-21 20:23:13,300][02158] Num frames 6200... [2025-08-21 20:23:13,421][02158] Num frames 6300... [2025-08-21 20:23:13,539][02158] Num frames 6400... [2025-08-21 20:23:13,659][02158] Avg episode rewards: #0: 19.933, true rewards: #0: 9.219 [2025-08-21 20:23:13,660][02158] Avg episode reward: 19.933, avg true_objective: 9.219 [2025-08-21 20:23:13,715][02158] Num frames 6500... [2025-08-21 20:23:13,829][02158] Num frames 6600... [2025-08-21 20:23:13,956][02158] Num frames 6700... [2025-08-21 20:23:14,073][02158] Num frames 6800... [2025-08-21 20:23:14,192][02158] Num frames 6900... [2025-08-21 20:23:14,321][02158] Avg episode rewards: #0: 18.456, true rewards: #0: 8.706 [2025-08-21 20:23:14,322][02158] Avg episode reward: 18.456, avg true_objective: 8.706 [2025-08-21 20:23:14,365][02158] Num frames 7000... [2025-08-21 20:23:14,480][02158] Num frames 7100... [2025-08-21 20:23:14,597][02158] Num frames 7200... [2025-08-21 20:23:14,713][02158] Num frames 7300... [2025-08-21 20:23:14,830][02158] Num frames 7400... [2025-08-21 20:23:14,965][02158] Num frames 7500... [2025-08-21 20:23:15,091][02158] Num frames 7600... [2025-08-21 20:23:15,210][02158] Num frames 7700... [2025-08-21 20:23:15,330][02158] Num frames 7800... [2025-08-21 20:23:15,482][02158] Avg episode rewards: #0: 18.753, true rewards: #0: 8.753 [2025-08-21 20:23:15,483][02158] Avg episode reward: 18.753, avg true_objective: 8.753 [2025-08-21 20:23:15,511][02158] Num frames 7900... [2025-08-21 20:23:15,628][02158] Num frames 8000... [2025-08-21 20:23:15,746][02158] Num frames 8100... [2025-08-21 20:23:15,860][02158] Num frames 8200... [2025-08-21 20:23:15,995][02158] Num frames 8300... [2025-08-21 20:23:16,116][02158] Num frames 8400... [2025-08-21 20:23:16,243][02158] Num frames 8500... [2025-08-21 20:23:16,364][02158] Num frames 8600... [2025-08-21 20:23:16,491][02158] Num frames 8700... [2025-08-21 20:23:16,613][02158] Num frames 8800... [2025-08-21 20:23:16,737][02158] Num frames 8900... [2025-08-21 20:23:16,861][02158] Num frames 9000... [2025-08-21 20:23:16,999][02158] Num frames 9100... [2025-08-21 20:23:17,124][02158] Num frames 9200... [2025-08-21 20:23:17,246][02158] Num frames 9300... [2025-08-21 20:23:17,333][02158] Avg episode rewards: #0: 20.325, true rewards: #0: 9.325 [2025-08-21 20:23:17,334][02158] Avg episode reward: 20.325, avg true_objective: 9.325 [2025-08-21 20:24:14,103][02158] Replay video saved to /content/train_dir/default_experiment/replay.mp4!