[2025-03-28 17:53:30,457][2713170] Saving configuration to /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/config.json... [2025-03-28 17:53:30,531][2713170] Rollout worker 0 uses device cpu [2025-03-28 17:53:30,532][2713170] Rollout worker 1 uses device cpu [2025-03-28 17:53:30,533][2713170] Rollout worker 2 uses device cpu [2025-03-28 17:53:30,533][2713170] Rollout worker 3 uses device cpu [2025-03-28 17:53:30,534][2713170] Rollout worker 4 uses device cpu [2025-03-28 17:53:30,535][2713170] Rollout worker 5 uses device cpu [2025-03-28 17:53:30,536][2713170] Rollout worker 6 uses device cpu [2025-03-28 17:53:30,537][2713170] Rollout worker 7 uses device cpu [2025-03-28 17:53:30,612][2713170] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-03-28 17:53:30,613][2713170] InferenceWorker_p0-w0: min num requests: 2 [2025-03-28 17:53:30,653][2713170] Starting all processes... [2025-03-28 17:53:30,654][2713170] Starting process learner_proc0 [2025-03-28 17:53:31,157][2713170] Starting all processes... [2025-03-28 17:53:31,169][2713170] Starting process inference_proc0-0 [2025-03-28 17:53:31,171][2713170] Starting process rollout_proc0 [2025-03-28 17:53:31,171][2713170] Starting process rollout_proc1 [2025-03-28 17:53:31,172][2713170] Starting process rollout_proc2 [2025-03-28 17:53:31,172][2713170] Starting process rollout_proc3 [2025-03-28 17:53:31,173][2713170] Starting process rollout_proc4 [2025-03-28 17:53:31,173][2713170] Starting process rollout_proc5 [2025-03-28 17:53:31,173][2713170] Starting process rollout_proc6 [2025-03-28 17:53:31,173][2713170] Starting process rollout_proc7 [2025-03-28 17:53:34,373][2730012] Worker 2 uses CPU cores [64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95] [2025-03-28 17:53:34,373][2730019] Worker 4 uses CPU cores [128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159] [2025-03-28 17:53:34,373][2730011] Worker 0 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31] [2025-03-28 17:53:34,375][2730010] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-03-28 17:53:34,376][2730010] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0 [2025-03-28 17:53:34,383][2730024] Worker 7 uses CPU cores [224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255] [2025-03-28 17:53:34,385][2730021] Worker 3 uses CPU cores [96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127] [2025-03-28 17:53:34,387][2730020] Worker 5 uses CPU cores [160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191] [2025-03-28 17:53:34,400][2730023] Worker 6 uses CPU cores [192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223] [2025-03-28 17:53:34,402][2729989] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-03-28 17:53:34,402][2729989] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0 [2025-03-28 17:53:34,402][2730013] Worker 1 uses CPU cores [32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63] [2025-03-28 17:53:34,904][2730010] Num visible devices: 1 [2025-03-28 17:53:34,906][2729989] Num visible devices: 1 [2025-03-28 17:53:34,907][2729989] Starting seed is not provided [2025-03-28 17:53:34,907][2729989] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-03-28 17:53:34,908][2729989] Initializing actor-critic model on device cuda:0 [2025-03-28 17:53:34,908][2729989] RunningMeanStd input shape: (3, 72, 128) [2025-03-28 17:53:34,938][2729989] RunningMeanStd input shape: (1,) [2025-03-28 17:53:34,951][2729989] ConvEncoder: input_channels=3 [2025-03-28 17:53:35,102][2729989] Conv encoder output size: 512 [2025-03-28 17:53:35,102][2729989] Policy head output size: 512 [2025-03-28 17:53:35,116][2729989] Created Actor Critic model with architecture: [2025-03-28 17:53:35,116][2729989] 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) ) ) [2025-03-28 17:53:35,564][2729989] Using optimizer [2025-03-28 17:53:37,203][2729989] No checkpoints found [2025-03-28 17:53:37,204][2729989] Did not load from checkpoint, starting from scratch! [2025-03-28 17:53:37,204][2729989] Initialized policy 0 weights for model version 0 [2025-03-28 17:53:37,506][2729989] LearnerWorker_p0 finished initialization! [2025-03-28 17:53:37,506][2729989] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-03-28 17:53:37,835][2713170] 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) [2025-03-28 17:53:37,878][2730010] RunningMeanStd input shape: (3, 72, 128) [2025-03-28 17:53:37,906][2730010] RunningMeanStd input shape: (1,) [2025-03-28 17:53:37,916][2730010] ConvEncoder: input_channels=3 [2025-03-28 17:53:38,004][2730010] Conv encoder output size: 512 [2025-03-28 17:53:38,004][2730010] Policy head output size: 512 [2025-03-28 17:53:38,123][2713170] Inference worker 0-0 is ready! [2025-03-28 17:53:38,124][2713170] All inference workers are ready! Signal rollout workers to start! [2025-03-28 17:53:38,153][2730019] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 17:53:38,158][2730024] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 17:53:38,159][2730020] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 17:53:38,159][2730013] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 17:53:38,169][2730011] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 17:53:38,175][2730012] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 17:53:38,175][2730021] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 17:53:38,176][2730023] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 17:53:38,718][2730020] Decorrelating experience for 0 frames... [2025-03-28 17:53:38,718][2730013] Decorrelating experience for 0 frames... [2025-03-28 17:53:38,718][2730012] Decorrelating experience for 0 frames... [2025-03-28 17:53:38,718][2730019] Decorrelating experience for 0 frames... [2025-03-28 17:53:38,718][2730024] Decorrelating experience for 0 frames... [2025-03-28 17:53:38,718][2730011] Decorrelating experience for 0 frames... [2025-03-28 17:53:39,141][2730024] Decorrelating experience for 32 frames... [2025-03-28 17:53:39,153][2730021] Decorrelating experience for 0 frames... [2025-03-28 17:53:39,165][2730013] Decorrelating experience for 32 frames... [2025-03-28 17:53:39,166][2730020] Decorrelating experience for 32 frames... [2025-03-28 17:53:39,167][2730019] Decorrelating experience for 32 frames... [2025-03-28 17:53:39,169][2730012] Decorrelating experience for 32 frames... [2025-03-28 17:53:39,172][2730023] Decorrelating experience for 0 frames... [2025-03-28 17:53:39,562][2730021] Decorrelating experience for 32 frames... [2025-03-28 17:53:39,581][2730023] Decorrelating experience for 32 frames... [2025-03-28 17:53:39,582][2730011] Decorrelating experience for 32 frames... [2025-03-28 17:53:39,597][2730024] Decorrelating experience for 64 frames... [2025-03-28 17:53:39,624][2730019] Decorrelating experience for 64 frames... [2025-03-28 17:53:39,635][2730012] Decorrelating experience for 64 frames... [2025-03-28 17:53:39,641][2730020] Decorrelating experience for 64 frames... [2025-03-28 17:53:39,641][2730013] Decorrelating experience for 64 frames... [2025-03-28 17:53:40,038][2730021] Decorrelating experience for 64 frames... [2025-03-28 17:53:40,052][2730024] Decorrelating experience for 96 frames... [2025-03-28 17:53:40,052][2730011] Decorrelating experience for 64 frames... [2025-03-28 17:53:40,064][2730012] Decorrelating experience for 96 frames... [2025-03-28 17:53:40,454][2730023] Decorrelating experience for 64 frames... [2025-03-28 17:53:40,485][2730021] Decorrelating experience for 96 frames... [2025-03-28 17:53:40,486][2730013] Decorrelating experience for 96 frames... [2025-03-28 17:53:40,486][2730011] Decorrelating experience for 96 frames... [2025-03-28 17:53:40,852][2730019] Decorrelating experience for 96 frames... [2025-03-28 17:53:40,913][2730023] Decorrelating experience for 96 frames... [2025-03-28 17:53:40,921][2730020] Decorrelating experience for 96 frames... [2025-03-28 17:53:41,244][2729989] Signal inference workers to stop experience collection... [2025-03-28 17:53:41,258][2730010] InferenceWorker_p0-w0: stopping experience collection [2025-03-28 17:53:42,835][2713170] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 318.4. Samples: 1592. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) [2025-03-28 17:53:42,837][2713170] Avg episode reward: [(0, '2.205')] [2025-03-28 17:53:43,658][2729989] Signal inference workers to resume experience collection... [2025-03-28 17:53:43,659][2730010] InferenceWorker_p0-w0: resuming experience collection [2025-03-28 17:53:44,998][2730010] Updated weights for policy 0, policy_version 10 (0.0066) [2025-03-28 17:53:46,399][2730010] Updated weights for policy 0, policy_version 20 (0.0007) [2025-03-28 17:53:47,835][2713170] Fps is (10 sec: 10240.0, 60 sec: 10240.0, 300 sec: 10240.0). Total num frames: 102400. Throughput: 0: 2609.0. Samples: 26090. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 17:53:47,837][2713170] Avg episode reward: [(0, '4.093')] [2025-03-28 17:53:47,839][2729989] Saving new best policy, reward=4.093! [2025-03-28 17:53:49,331][2730010] Updated weights for policy 0, policy_version 30 (0.0007) [2025-03-28 17:53:50,602][2713170] Heartbeat connected on Batcher_0 [2025-03-28 17:53:50,606][2713170] Heartbeat connected on LearnerWorker_p0 [2025-03-28 17:53:50,614][2713170] Heartbeat connected on InferenceWorker_p0-w0 [2025-03-28 17:53:50,619][2713170] Heartbeat connected on RolloutWorker_w0 [2025-03-28 17:53:50,624][2713170] Heartbeat connected on RolloutWorker_w1 [2025-03-28 17:53:50,629][2713170] Heartbeat connected on RolloutWorker_w2 [2025-03-28 17:53:50,638][2713170] Heartbeat connected on RolloutWorker_w3 [2025-03-28 17:53:50,643][2713170] Heartbeat connected on RolloutWorker_w4 [2025-03-28 17:53:50,644][2713170] Heartbeat connected on RolloutWorker_w5 [2025-03-28 17:53:50,651][2713170] Heartbeat connected on RolloutWorker_w6 [2025-03-28 17:53:50,657][2713170] Heartbeat connected on RolloutWorker_w7 [2025-03-28 17:53:50,841][2730010] Updated weights for policy 0, policy_version 40 (0.0007) [2025-03-28 17:53:52,230][2730010] Updated weights for policy 0, policy_version 50 (0.0007) [2025-03-28 17:53:52,835][2713170] Fps is (10 sec: 22118.6, 60 sec: 14745.6, 300 sec: 14745.6). Total num frames: 221184. Throughput: 0: 2444.8. Samples: 36672. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) [2025-03-28 17:53:52,837][2713170] Avg episode reward: [(0, '4.486')] [2025-03-28 17:53:52,843][2729989] Saving new best policy, reward=4.486! [2025-03-28 17:53:53,668][2730010] Updated weights for policy 0, policy_version 60 (0.0007) [2025-03-28 17:53:55,180][2730010] Updated weights for policy 0, policy_version 70 (0.0006) [2025-03-28 17:53:56,687][2730010] Updated weights for policy 0, policy_version 80 (0.0006) [2025-03-28 17:53:57,835][2713170] Fps is (10 sec: 25395.3, 60 sec: 17817.7, 300 sec: 17817.7). Total num frames: 356352. Throughput: 0: 3934.8. Samples: 78696. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 17:53:57,836][2713170] Avg episode reward: [(0, '4.392')] [2025-03-28 17:53:58,253][2730010] Updated weights for policy 0, policy_version 90 (0.0006) [2025-03-28 17:53:59,741][2730010] Updated weights for policy 0, policy_version 100 (0.0007) [2025-03-28 17:54:01,248][2730010] Updated weights for policy 0, policy_version 110 (0.0006) [2025-03-28 17:54:02,795][2730010] Updated weights for policy 0, policy_version 120 (0.0006) [2025-03-28 17:54:02,835][2713170] Fps is (10 sec: 27033.9, 60 sec: 19660.9, 300 sec: 19660.9). Total num frames: 491520. Throughput: 0: 4778.9. Samples: 119472. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-03-28 17:54:02,836][2713170] Avg episode reward: [(0, '4.558')] [2025-03-28 17:54:02,841][2729989] Saving new best policy, reward=4.558! [2025-03-28 17:54:04,316][2730010] Updated weights for policy 0, policy_version 130 (0.0006) [2025-03-28 17:54:05,799][2730010] Updated weights for policy 0, policy_version 140 (0.0006) [2025-03-28 17:54:07,289][2730010] Updated weights for policy 0, policy_version 150 (0.0006) [2025-03-28 17:54:07,835][2713170] Fps is (10 sec: 27033.4, 60 sec: 20889.6, 300 sec: 20889.6). Total num frames: 626688. Throughput: 0: 4649.3. Samples: 139478. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 17:54:07,837][2713170] Avg episode reward: [(0, '4.413')] [2025-03-28 17:54:08,812][2730010] Updated weights for policy 0, policy_version 160 (0.0006) [2025-03-28 17:54:10,255][2730010] Updated weights for policy 0, policy_version 170 (0.0006) [2025-03-28 17:54:11,700][2730010] Updated weights for policy 0, policy_version 180 (0.0006) [2025-03-28 17:54:12,835][2713170] Fps is (10 sec: 27442.8, 60 sec: 21884.4, 300 sec: 21884.4). Total num frames: 765952. Throughput: 0: 5168.9. Samples: 180910. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 17:54:12,843][2713170] Avg episode reward: [(0, '4.549')] [2025-03-28 17:54:13,151][2730010] Updated weights for policy 0, policy_version 190 (0.0006) [2025-03-28 17:54:14,627][2730010] Updated weights for policy 0, policy_version 200 (0.0006) [2025-03-28 17:54:16,062][2730010] Updated weights for policy 0, policy_version 210 (0.0006) [2025-03-28 17:54:17,334][2730010] Updated weights for policy 0, policy_version 220 (0.0007) [2025-03-28 17:54:17,835][2713170] Fps is (10 sec: 28672.2, 60 sec: 22835.3, 300 sec: 22835.3). Total num frames: 913408. Throughput: 0: 5615.0. Samples: 224598. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 17:54:17,837][2713170] Avg episode reward: [(0, '5.481')] [2025-03-28 17:54:17,839][2729989] Saving new best policy, reward=5.481! [2025-03-28 17:54:18,695][2730010] Updated weights for policy 0, policy_version 230 (0.0006) [2025-03-28 17:54:20,104][2730010] Updated weights for policy 0, policy_version 240 (0.0006) [2025-03-28 17:54:21,638][2730010] Updated weights for policy 0, policy_version 250 (0.0006) [2025-03-28 17:54:22,835][2713170] Fps is (10 sec: 28671.9, 60 sec: 23392.7, 300 sec: 23392.7). Total num frames: 1052672. Throughput: 0: 5479.4. Samples: 246574. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) [2025-03-28 17:54:22,837][2713170] Avg episode reward: [(0, '5.019')] [2025-03-28 17:54:23,212][2730010] Updated weights for policy 0, policy_version 260 (0.0006) [2025-03-28 17:54:24,681][2730010] Updated weights for policy 0, policy_version 270 (0.0006) [2025-03-28 17:54:26,137][2730010] Updated weights for policy 0, policy_version 280 (0.0006) [2025-03-28 17:54:27,609][2730010] Updated weights for policy 0, policy_version 290 (0.0006) [2025-03-28 17:54:27,835][2713170] Fps is (10 sec: 27852.1, 60 sec: 23838.6, 300 sec: 23838.6). Total num frames: 1191936. Throughput: 0: 6351.0. Samples: 287386. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 17:54:27,837][2713170] Avg episode reward: [(0, '6.011')] [2025-03-28 17:54:27,838][2729989] Saving new best policy, reward=6.011! [2025-03-28 17:54:28,987][2730010] Updated weights for policy 0, policy_version 300 (0.0006) [2025-03-28 17:54:30,156][2730010] Updated weights for policy 0, policy_version 310 (0.0007) [2025-03-28 17:54:31,639][2730010] Updated weights for policy 0, policy_version 320 (0.0006) [2025-03-28 17:54:32,836][2713170] Fps is (10 sec: 29081.4, 60 sec: 24427.0, 300 sec: 24427.0). Total num frames: 1343488. Throughput: 0: 6796.3. Samples: 331926. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 17:54:32,839][2713170] Avg episode reward: [(0, '6.494')] [2025-03-28 17:54:32,845][2729989] Saving new best policy, reward=6.494! [2025-03-28 17:54:33,131][2730010] Updated weights for policy 0, policy_version 330 (0.0006) [2025-03-28 17:54:34,679][2730010] Updated weights for policy 0, policy_version 340 (0.0006) [2025-03-28 17:54:36,210][2730010] Updated weights for policy 0, policy_version 350 (0.0006) [2025-03-28 17:54:37,735][2730010] Updated weights for policy 0, policy_version 360 (0.0006) [2025-03-28 17:54:37,835][2713170] Fps is (10 sec: 28262.9, 60 sec: 24576.0, 300 sec: 24576.0). Total num frames: 1474560. Throughput: 0: 7006.5. Samples: 351964. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) [2025-03-28 17:54:37,837][2713170] Avg episode reward: [(0, '6.363')] [2025-03-28 17:54:39,290][2730010] Updated weights for policy 0, policy_version 370 (0.0006) [2025-03-28 17:54:40,834][2730010] Updated weights for policy 0, policy_version 380 (0.0006) [2025-03-28 17:54:42,394][2730010] Updated weights for policy 0, policy_version 390 (0.0006) [2025-03-28 17:54:42,835][2713170] Fps is (10 sec: 26214.6, 60 sec: 26760.5, 300 sec: 24702.0). Total num frames: 1605632. Throughput: 0: 6955.8. Samples: 391708. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 17:54:42,837][2713170] Avg episode reward: [(0, '6.408')] [2025-03-28 17:54:43,936][2730010] Updated weights for policy 0, policy_version 400 (0.0006) [2025-03-28 17:54:45,513][2730010] Updated weights for policy 0, policy_version 410 (0.0006) [2025-03-28 17:54:47,039][2730010] Updated weights for policy 0, policy_version 420 (0.0006) [2025-03-28 17:54:47,836][2713170] Fps is (10 sec: 26622.9, 60 sec: 27306.5, 300 sec: 24868.4). Total num frames: 1740800. Throughput: 0: 6930.5. Samples: 431348. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 17:54:47,840][2713170] Avg episode reward: [(0, '6.476')] [2025-03-28 17:54:48,561][2730010] Updated weights for policy 0, policy_version 430 (0.0006) [2025-03-28 17:54:50,114][2730010] Updated weights for policy 0, policy_version 440 (0.0006) [2025-03-28 17:54:51,659][2730010] Updated weights for policy 0, policy_version 450 (0.0007) [2025-03-28 17:54:52,835][2713170] Fps is (10 sec: 26624.0, 60 sec: 27511.4, 300 sec: 24958.3). Total num frames: 1871872. Throughput: 0: 6930.8. Samples: 451366. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 17:54:52,837][2713170] Avg episode reward: [(0, '7.730')] [2025-03-28 17:54:52,845][2729989] Saving new best policy, reward=7.730! [2025-03-28 17:54:53,179][2730010] Updated weights for policy 0, policy_version 460 (0.0006) [2025-03-28 17:54:54,724][2730010] Updated weights for policy 0, policy_version 470 (0.0006) [2025-03-28 17:54:56,274][2730010] Updated weights for policy 0, policy_version 480 (0.0006) [2025-03-28 17:54:57,752][2730010] Updated weights for policy 0, policy_version 490 (0.0006) [2025-03-28 17:54:57,835][2713170] Fps is (10 sec: 26624.9, 60 sec: 27511.4, 300 sec: 25088.0). Total num frames: 2007040. Throughput: 0: 6897.6. Samples: 491302. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 17:54:57,837][2713170] Avg episode reward: [(0, '9.450')] [2025-03-28 17:54:57,839][2729989] Saving new best policy, reward=9.450! [2025-03-28 17:54:59,188][2730010] Updated weights for policy 0, policy_version 500 (0.0006) [2025-03-28 17:55:00,652][2730010] Updated weights for policy 0, policy_version 510 (0.0006) [2025-03-28 17:55:02,126][2730010] Updated weights for policy 0, policy_version 520 (0.0006) [2025-03-28 17:55:02,835][2713170] Fps is (10 sec: 27443.3, 60 sec: 27579.7, 300 sec: 25250.6). Total num frames: 2146304. Throughput: 0: 6855.7. Samples: 533104. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 17:55:02,837][2713170] Avg episode reward: [(0, '9.217')] [2025-03-28 17:55:03,594][2730010] Updated weights for policy 0, policy_version 530 (0.0006) [2025-03-28 17:55:05,113][2730010] Updated weights for policy 0, policy_version 540 (0.0006) [2025-03-28 17:55:06,639][2730010] Updated weights for policy 0, policy_version 550 (0.0006) [2025-03-28 17:55:07,835][2713170] Fps is (10 sec: 27852.9, 60 sec: 27648.0, 300 sec: 25395.2). Total num frames: 2285568. Throughput: 0: 6827.4. Samples: 553806. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 17:55:07,836][2713170] Avg episode reward: [(0, '11.260')] [2025-03-28 17:55:07,838][2729989] Saving new best policy, reward=11.260! [2025-03-28 17:55:08,092][2730010] Updated weights for policy 0, policy_version 560 (0.0006) [2025-03-28 17:55:09,608][2730010] Updated weights for policy 0, policy_version 570 (0.0006) [2025-03-28 17:55:11,074][2730010] Updated weights for policy 0, policy_version 580 (0.0007) [2025-03-28 17:55:12,569][2730010] Updated weights for policy 0, policy_version 590 (0.0006) [2025-03-28 17:55:12,835][2713170] Fps is (10 sec: 27443.1, 60 sec: 27579.7, 300 sec: 25481.4). Total num frames: 2420736. Throughput: 0: 6836.8. Samples: 595042. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 17:55:12,837][2713170] Avg episode reward: [(0, '11.551')] [2025-03-28 17:55:12,842][2729989] Saving new best policy, reward=11.551! [2025-03-28 17:55:14,024][2730010] Updated weights for policy 0, policy_version 600 (0.0006) [2025-03-28 17:55:15,515][2730010] Updated weights for policy 0, policy_version 610 (0.0006) [2025-03-28 17:55:16,978][2730010] Updated weights for policy 0, policy_version 620 (0.0006) [2025-03-28 17:55:17,835][2713170] Fps is (10 sec: 27443.2, 60 sec: 27443.2, 300 sec: 25600.0). Total num frames: 2560000. Throughput: 0: 6770.4. Samples: 636592. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 17:55:17,837][2713170] Avg episode reward: [(0, '11.011')] [2025-03-28 17:55:18,528][2730010] Updated weights for policy 0, policy_version 630 (0.0006) [2025-03-28 17:55:20,083][2730010] Updated weights for policy 0, policy_version 640 (0.0006) [2025-03-28 17:55:21,579][2730010] Updated weights for policy 0, policy_version 650 (0.0006) [2025-03-28 17:55:22,835][2713170] Fps is (10 sec: 27443.0, 60 sec: 27374.9, 300 sec: 25668.2). Total num frames: 2695168. Throughput: 0: 6764.1. Samples: 656350. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 17:55:22,837][2713170] Avg episode reward: [(0, '11.818')] [2025-03-28 17:55:22,843][2729989] Saving /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000659_2699264.pth... [2025-03-28 17:55:22,944][2729989] Saving new best policy, reward=11.818! [2025-03-28 17:55:23,061][2730010] Updated weights for policy 0, policy_version 660 (0.0007) [2025-03-28 17:55:24,335][2730010] Updated weights for policy 0, policy_version 670 (0.0007) [2025-03-28 17:55:25,720][2730010] Updated weights for policy 0, policy_version 680 (0.0007) [2025-03-28 17:55:27,211][2730010] Updated weights for policy 0, policy_version 690 (0.0007) [2025-03-28 17:55:27,835][2713170] Fps is (10 sec: 27852.8, 60 sec: 27443.3, 300 sec: 25804.8). Total num frames: 2838528. Throughput: 0: 6846.9. Samples: 699820. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-03-28 17:55:27,842][2713170] Avg episode reward: [(0, '11.834')] [2025-03-28 17:55:27,844][2729989] Saving new best policy, reward=11.834! [2025-03-28 17:55:28,751][2730010] Updated weights for policy 0, policy_version 700 (0.0007) [2025-03-28 17:55:30,279][2730010] Updated weights for policy 0, policy_version 710 (0.0007) [2025-03-28 17:55:31,848][2730010] Updated weights for policy 0, policy_version 720 (0.0007) [2025-03-28 17:55:32,835][2713170] Fps is (10 sec: 27853.0, 60 sec: 27170.2, 300 sec: 25858.2). Total num frames: 2973696. Throughput: 0: 6854.8. Samples: 739810. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-03-28 17:55:32,837][2713170] Avg episode reward: [(0, '13.856')] [2025-03-28 17:55:32,843][2729989] Saving new best policy, reward=13.856! [2025-03-28 17:55:33,369][2730010] Updated weights for policy 0, policy_version 730 (0.0007) [2025-03-28 17:55:34,938][2730010] Updated weights for policy 0, policy_version 740 (0.0007) [2025-03-28 17:55:36,456][2730010] Updated weights for policy 0, policy_version 750 (0.0007) [2025-03-28 17:55:37,835][2713170] Fps is (10 sec: 26624.0, 60 sec: 27170.1, 300 sec: 25873.1). Total num frames: 3104768. Throughput: 0: 6853.6. Samples: 759780. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-03-28 17:55:37,837][2713170] Avg episode reward: [(0, '11.840')] [2025-03-28 17:55:38,032][2730010] Updated weights for policy 0, policy_version 760 (0.0007) [2025-03-28 17:55:39,566][2730010] Updated weights for policy 0, policy_version 770 (0.0007) [2025-03-28 17:55:41,000][2730010] Updated weights for policy 0, policy_version 780 (0.0007) [2025-03-28 17:55:42,482][2730010] Updated weights for policy 0, policy_version 790 (0.0006) [2025-03-28 17:55:42,835][2713170] Fps is (10 sec: 27033.7, 60 sec: 27306.7, 300 sec: 25952.3). Total num frames: 3244032. Throughput: 0: 6868.9. Samples: 800404. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 17:55:42,836][2713170] Avg episode reward: [(0, '15.923')] [2025-03-28 17:55:42,843][2729989] Saving new best policy, reward=15.923! [2025-03-28 17:55:43,948][2730010] Updated weights for policy 0, policy_version 800 (0.0007) [2025-03-28 17:55:45,425][2730010] Updated weights for policy 0, policy_version 810 (0.0006) [2025-03-28 17:55:46,967][2730010] Updated weights for policy 0, policy_version 820 (0.0006) [2025-03-28 17:55:47,835][2713170] Fps is (10 sec: 27443.2, 60 sec: 27306.8, 300 sec: 25993.8). Total num frames: 3379200. Throughput: 0: 6853.2. Samples: 841498. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 17:55:47,837][2713170] Avg episode reward: [(0, '17.706')] [2025-03-28 17:55:47,838][2729989] Saving new best policy, reward=17.706! [2025-03-28 17:55:48,475][2730010] Updated weights for policy 0, policy_version 830 (0.0006) [2025-03-28 17:55:50,012][2730010] Updated weights for policy 0, policy_version 840 (0.0006) [2025-03-28 17:55:51,474][2730010] Updated weights for policy 0, policy_version 850 (0.0006) [2025-03-28 17:55:52,835][2713170] Fps is (10 sec: 27033.6, 60 sec: 27374.9, 300 sec: 26032.4). Total num frames: 3514368. Throughput: 0: 6842.1. Samples: 861700. Policy #0 lag: (min: 0.0, avg: 0.9, max: 2.0) [2025-03-28 17:55:52,837][2713170] Avg episode reward: [(0, '17.510')] [2025-03-28 17:55:53,056][2730010] Updated weights for policy 0, policy_version 860 (0.0006) [2025-03-28 17:55:54,494][2730010] Updated weights for policy 0, policy_version 870 (0.0006) [2025-03-28 17:55:55,979][2730010] Updated weights for policy 0, policy_version 880 (0.0006) [2025-03-28 17:55:57,517][2730010] Updated weights for policy 0, policy_version 890 (0.0006) [2025-03-28 17:55:57,835][2713170] Fps is (10 sec: 27443.2, 60 sec: 27443.2, 300 sec: 26097.4). Total num frames: 3653632. Throughput: 0: 6839.1. Samples: 902802. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 17:55:57,837][2713170] Avg episode reward: [(0, '19.800')] [2025-03-28 17:55:57,838][2729989] Saving new best policy, reward=19.800! [2025-03-28 17:55:59,010][2730010] Updated weights for policy 0, policy_version 900 (0.0006) [2025-03-28 17:56:00,480][2730010] Updated weights for policy 0, policy_version 910 (0.0006) [2025-03-28 17:56:01,991][2730010] Updated weights for policy 0, policy_version 920 (0.0006) [2025-03-28 17:56:02,835][2713170] Fps is (10 sec: 27443.1, 60 sec: 27374.9, 300 sec: 26129.6). Total num frames: 3788800. Throughput: 0: 6822.7. Samples: 943614. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-03-28 17:56:02,837][2713170] Avg episode reward: [(0, '16.817')] [2025-03-28 17:56:03,498][2730010] Updated weights for policy 0, policy_version 930 (0.0006) [2025-03-28 17:56:05,032][2730010] Updated weights for policy 0, policy_version 940 (0.0006) [2025-03-28 17:56:06,454][2730010] Updated weights for policy 0, policy_version 950 (0.0006) [2025-03-28 17:56:07,835][2713170] Fps is (10 sec: 27443.2, 60 sec: 27374.9, 300 sec: 26187.1). Total num frames: 3928064. Throughput: 0: 6839.6. Samples: 964132. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-03-28 17:56:07,837][2713170] Avg episode reward: [(0, '20.252')] [2025-03-28 17:56:07,838][2729989] Saving new best policy, reward=20.252! [2025-03-28 17:56:07,971][2730010] Updated weights for policy 0, policy_version 960 (0.0006) [2025-03-28 17:56:09,442][2730010] Updated weights for policy 0, policy_version 970 (0.0006) [2025-03-28 17:56:10,647][2729989] Stopping Batcher_0... [2025-03-28 17:56:10,647][2729989] Saving /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 17:56:10,647][2713170] Component Batcher_0 stopped! [2025-03-28 17:56:10,648][2729989] Loop batcher_evt_loop terminating... [2025-03-28 17:56:10,680][2730010] Weights refcount: 2 0 [2025-03-28 17:56:10,738][2730013] Stopping RolloutWorker_w1... [2025-03-28 17:56:10,739][2730013] Loop rollout_proc1_evt_loop terminating... [2025-03-28 17:56:10,738][2713170] Component RolloutWorker_w1 stopped! [2025-03-28 17:56:10,741][2730010] Stopping InferenceWorker_p0-w0... [2025-03-28 17:56:10,741][2730010] Loop inference_proc0-0_evt_loop terminating... [2025-03-28 17:56:10,743][2729989] Saving /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 17:56:10,742][2713170] Component InferenceWorker_p0-w0 stopped! [2025-03-28 17:56:10,744][2713170] Component RolloutWorker_w2 stopped! [2025-03-28 17:56:10,745][2730012] Stopping RolloutWorker_w2... [2025-03-28 17:56:10,746][2730012] Loop rollout_proc2_evt_loop terminating... [2025-03-28 17:56:10,746][2730020] Stopping RolloutWorker_w5... [2025-03-28 17:56:10,747][2730020] Loop rollout_proc5_evt_loop terminating... [2025-03-28 17:56:10,746][2713170] Component RolloutWorker_w6 stopped! [2025-03-28 17:56:10,747][2730019] Stopping RolloutWorker_w4... [2025-03-28 17:56:10,748][2730019] Loop rollout_proc4_evt_loop terminating... [2025-03-28 17:56:10,746][2730023] Stopping RolloutWorker_w6... [2025-03-28 17:56:10,749][2730023] Loop rollout_proc6_evt_loop terminating... [2025-03-28 17:56:10,748][2730024] Stopping RolloutWorker_w7... [2025-03-28 17:56:10,748][2730021] Stopping RolloutWorker_w3... [2025-03-28 17:56:10,750][2730024] Loop rollout_proc7_evt_loop terminating... [2025-03-28 17:56:10,750][2730021] Loop rollout_proc3_evt_loop terminating... [2025-03-28 17:56:10,748][2713170] Component RolloutWorker_w5 stopped! [2025-03-28 17:56:10,752][2713170] Component RolloutWorker_w4 stopped! [2025-03-28 17:56:10,753][2713170] Component RolloutWorker_w3 stopped! [2025-03-28 17:56:10,753][2713170] Component RolloutWorker_w7 stopped! [2025-03-28 17:56:10,755][2730011] Stopping RolloutWorker_w0... [2025-03-28 17:56:10,757][2730011] Loop rollout_proc0_evt_loop terminating... [2025-03-28 17:56:10,755][2713170] Component RolloutWorker_w0 stopped! [2025-03-28 17:56:10,818][2729989] Stopping LearnerWorker_p0... [2025-03-28 17:56:10,818][2729989] Loop learner_proc0_evt_loop terminating... [2025-03-28 17:56:10,818][2713170] Component LearnerWorker_p0 stopped! [2025-03-28 17:56:10,820][2713170] Waiting for process learner_proc0 to stop... [2025-03-28 17:56:11,703][2713170] Waiting for process inference_proc0-0 to join... [2025-03-28 17:56:11,705][2713170] Waiting for process rollout_proc0 to join... [2025-03-28 17:56:11,706][2713170] Waiting for process rollout_proc1 to join... [2025-03-28 17:56:11,708][2713170] Waiting for process rollout_proc2 to join... [2025-03-28 17:56:11,709][2713170] Waiting for process rollout_proc3 to join... [2025-03-28 17:56:11,710][2713170] Waiting for process rollout_proc4 to join... [2025-03-28 17:56:11,711][2713170] Waiting for process rollout_proc5 to join... [2025-03-28 17:56:11,712][2713170] Waiting for process rollout_proc6 to join... [2025-03-28 17:56:11,713][2713170] Waiting for process rollout_proc7 to join... [2025-03-28 17:56:11,715][2713170] Batcher 0 profile tree view: batching: 14.1429, releasing_batches: 0.0182 [2025-03-28 17:56:11,716][2713170] InferenceWorker_p0-w0 profile tree view: wait_policy: 0.0000 wait_policy_total: 3.7221 update_model: 2.0099 weight_update: 0.0006 one_step: 0.0015 handle_policy_step: 137.5038 deserialize: 9.3191, stack: 0.7309, obs_to_device_normalize: 29.5231, forward: 65.4827, send_messages: 7.6935 prepare_outputs: 18.5681 to_cpu: 11.0246 [2025-03-28 17:56:11,717][2713170] Learner 0 profile tree view: misc: 0.0044, prepare_batch: 6.8311 train: 15.5300 epoch_init: 0.0049, minibatch_init: 0.0047, losses_postprocess: 0.2485, kl_divergence: 0.3022, after_optimizer: 2.1743 calculate_losses: 6.5099 losses_init: 0.0028, forward_head: 0.5498, bptt_initial: 3.4675, tail: 0.4757, advantages_returns: 0.1261, losses: 0.8851 bptt: 0.8547 bptt_forward_core: 0.8105 update: 5.9615 clip: 0.6831 [2025-03-28 17:56:11,718][2713170] RolloutWorker_w0 profile tree view: wait_for_trajectories: 0.1346, enqueue_policy_requests: 6.9249, env_step: 95.6459, overhead: 7.8198, complete_rollouts: 0.2001 save_policy_outputs: 7.4937 split_output_tensors: 3.5673 [2025-03-28 17:56:11,718][2713170] RolloutWorker_w7 profile tree view: wait_for_trajectories: 0.1319, enqueue_policy_requests: 7.9354, env_step: 94.6488, overhead: 7.8292, complete_rollouts: 0.2029 save_policy_outputs: 7.6534 split_output_tensors: 3.6544 [2025-03-28 17:56:11,720][2713170] Loop Runner_EvtLoop terminating... [2025-03-28 17:56:11,721][2713170] Runner profile tree view: main_loop: 161.0680 [2025-03-28 17:56:11,721][2713170] Collected {0: 4005888}, FPS: 24870.8 [2025-03-28 17:56:12,155][2713170] Loading existing experiment configuration from /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/config.json [2025-03-28 17:56:12,157][2713170] Overriding arg 'num_workers' with value 1 passed from command line [2025-03-28 17:56:12,158][2713170] Adding new argument 'no_render'=True that is not in the saved config file! [2025-03-28 17:56:12,159][2713170] Adding new argument 'save_video'=True that is not in the saved config file! [2025-03-28 17:56:12,160][2713170] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-03-28 17:56:12,161][2713170] Adding new argument 'video_name'=None that is not in the saved config file! [2025-03-28 17:56:12,162][2713170] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file! [2025-03-28 17:56:12,162][2713170] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-03-28 17:56:12,163][2713170] Adding new argument 'push_to_hub'=False that is not in the saved config file! [2025-03-28 17:56:12,164][2713170] Adding new argument 'hf_repository'=None that is not in the saved config file! [2025-03-28 17:56:12,165][2713170] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-03-28 17:56:12,166][2713170] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-03-28 17:56:12,166][2713170] Adding new argument 'train_script'=None that is not in the saved config file! [2025-03-28 17:56:12,167][2713170] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-03-28 17:56:12,168][2713170] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-03-28 17:56:12,192][2713170] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 17:56:12,196][2713170] RunningMeanStd input shape: (3, 72, 128) [2025-03-28 17:56:12,197][2713170] RunningMeanStd input shape: (1,) [2025-03-28 17:56:12,210][2713170] ConvEncoder: input_channels=3 [2025-03-28 17:56:12,294][2713170] Conv encoder output size: 512 [2025-03-28 17:56:12,295][2713170] Policy head output size: 512 [2025-03-28 17:56:14,346][2713170] Loading state from checkpoint /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 17:56:14,350][2713170] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/torch/serialization.py", line 1470, 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([scalar])` or the `torch.serialization.safe_globals([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. [2025-03-28 17:56:14,354][2713170] Loading state from checkpoint /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 17:56:14,355][2713170] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/torch/serialization.py", line 1470, 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([scalar])` or the `torch.serialization.safe_globals([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. [2025-03-28 17:56:14,356][2713170] Loading state from checkpoint /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 17:56:14,358][2713170] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/torch/serialization.py", line 1470, 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([scalar])` or the `torch.serialization.safe_globals([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. [2025-03-28 17:59:10,915][2713170] Loading existing experiment configuration from /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/config.json [2025-03-28 17:59:10,917][2713170] Overriding arg 'num_workers' with value 1 passed from command line [2025-03-28 17:59:10,918][2713170] Adding new argument 'no_render'=True that is not in the saved config file! [2025-03-28 17:59:10,918][2713170] Adding new argument 'save_video'=False that is not in the saved config file! [2025-03-28 17:59:10,919][2713170] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-03-28 17:59:10,920][2713170] Adding new argument 'video_name'=None that is not in the saved config file! [2025-03-28 17:59:10,920][2713170] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file! [2025-03-28 17:59:10,921][2713170] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-03-28 17:59:10,922][2713170] Adding new argument 'push_to_hub'=False that is not in the saved config file! [2025-03-28 17:59:10,923][2713170] Adding new argument 'hf_repository'=None that is not in the saved config file! [2025-03-28 17:59:10,923][2713170] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-03-28 17:59:10,924][2713170] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-03-28 17:59:10,925][2713170] Adding new argument 'train_script'=None that is not in the saved config file! [2025-03-28 17:59:10,926][2713170] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-03-28 17:59:10,927][2713170] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-03-28 17:59:10,965][2713170] RunningMeanStd input shape: (3, 72, 128) [2025-03-28 17:59:10,966][2713170] RunningMeanStd input shape: (1,) [2025-03-28 17:59:10,979][2713170] ConvEncoder: input_channels=3 [2025-03-28 17:59:11,015][2713170] Conv encoder output size: 512 [2025-03-28 17:59:11,016][2713170] Policy head output size: 512 [2025-03-28 17:59:11,049][2713170] Loading state from checkpoint /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 17:59:11,051][2713170] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/torch/serialization.py", line 1470, 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([scalar])` or the `torch.serialization.safe_globals([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. [2025-03-28 17:59:11,052][2713170] Loading state from checkpoint /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 17:59:11,054][2713170] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/torch/serialization.py", line 1470, 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([scalar])` or the `torch.serialization.safe_globals([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. [2025-03-28 17:59:11,055][2713170] Loading state from checkpoint /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 17:59:11,056][2713170] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/torch/serialization.py", line 1470, 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([scalar])` or the `torch.serialization.safe_globals([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. [2025-03-28 18:03:15,287][2713170] Loading existing experiment configuration from /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/config.json [2025-03-28 18:03:15,289][2713170] Overriding arg 'num_workers' with value 1 passed from command line [2025-03-28 18:03:15,289][2713170] Adding new argument 'no_render'=True that is not in the saved config file! [2025-03-28 18:03:15,290][2713170] Adding new argument 'save_video'=True that is not in the saved config file! [2025-03-28 18:03:15,291][2713170] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-03-28 18:03:15,291][2713170] Adding new argument 'video_name'=None that is not in the saved config file! [2025-03-28 18:03:15,292][2713170] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file! [2025-03-28 18:03:15,293][2713170] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-03-28 18:03:15,294][2713170] Adding new argument 'push_to_hub'=False that is not in the saved config file! [2025-03-28 18:03:15,294][2713170] Adding new argument 'hf_repository'=None that is not in the saved config file! [2025-03-28 18:03:15,295][2713170] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-03-28 18:03:15,296][2713170] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-03-28 18:03:15,296][2713170] Adding new argument 'train_script'=None that is not in the saved config file! [2025-03-28 18:03:15,297][2713170] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-03-28 18:03:15,298][2713170] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-03-28 18:03:15,317][2713170] RunningMeanStd input shape: (3, 72, 128) [2025-03-28 18:03:15,318][2713170] RunningMeanStd input shape: (1,) [2025-03-28 18:03:15,326][2713170] ConvEncoder: input_channels=3 [2025-03-28 18:03:15,355][2713170] Conv encoder output size: 512 [2025-03-28 18:03:15,355][2713170] Policy head output size: 512 [2025-03-28 18:03:16,020][2713170] Loading state from checkpoint /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 18:03:16,022][2713170] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/torch/serialization.py", line 1470, 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.dtype was not an allowed global by default. Please use `torch.serialization.add_safe_globals([dtype])` or the `torch.serialization.safe_globals([dtype])` 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. [2025-03-28 18:03:16,024][2713170] Loading state from checkpoint /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 18:03:16,025][2713170] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/torch/serialization.py", line 1470, 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.dtype was not an allowed global by default. Please use `torch.serialization.add_safe_globals([dtype])` or the `torch.serialization.safe_globals([dtype])` 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. [2025-03-28 18:03:16,026][2713170] Loading state from checkpoint /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 18:03:16,027][2713170] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/torch/serialization.py", line 1470, 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.dtype was not an allowed global by default. Please use `torch.serialization.add_safe_globals([dtype])` or the `torch.serialization.safe_globals([dtype])` 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. [2025-03-28 18:05:00,938][2713170] Loading existing experiment configuration from /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/config.json [2025-03-28 18:05:00,939][2713170] Overriding arg 'num_workers' with value 1 passed from command line [2025-03-28 18:05:00,940][2713170] Adding new argument 'no_render'=True that is not in the saved config file! [2025-03-28 18:05:00,941][2713170] Adding new argument 'save_video'=True that is not in the saved config file! [2025-03-28 18:05:00,941][2713170] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-03-28 18:05:00,942][2713170] Adding new argument 'video_name'=None that is not in the saved config file! [2025-03-28 18:05:00,943][2713170] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file! [2025-03-28 18:05:00,943][2713170] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-03-28 18:05:00,944][2713170] Adding new argument 'push_to_hub'=False that is not in the saved config file! [2025-03-28 18:05:00,945][2713170] Adding new argument 'hf_repository'=None that is not in the saved config file! [2025-03-28 18:05:00,946][2713170] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-03-28 18:05:00,946][2713170] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-03-28 18:05:00,947][2713170] Adding new argument 'train_script'=None that is not in the saved config file! [2025-03-28 18:05:00,948][2713170] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-03-28 18:05:00,948][2713170] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-03-28 18:05:00,966][2713170] RunningMeanStd input shape: (3, 72, 128) [2025-03-28 18:05:00,967][2713170] RunningMeanStd input shape: (1,) [2025-03-28 18:05:00,976][2713170] ConvEncoder: input_channels=3 [2025-03-28 18:05:01,005][2713170] Conv encoder output size: 512 [2025-03-28 18:05:01,006][2713170] Policy head output size: 512 [2025-03-28 18:05:01,959][2713170] Num frames 100... [2025-03-28 18:05:02,060][2713170] Num frames 200... [2025-03-28 18:05:02,157][2713170] Num frames 300... [2025-03-28 18:05:02,257][2713170] Num frames 400... [2025-03-28 18:05:02,356][2713170] Num frames 500... [2025-03-28 18:05:02,448][2713170] Num frames 600... [2025-03-28 18:05:02,550][2713170] Num frames 700... [2025-03-28 18:05:02,614][2713170] Avg episode rewards: #0: 14.060, true rewards: #0: 7.060 [2025-03-28 18:05:02,615][2713170] Avg episode reward: 14.060, avg true_objective: 7.060 [2025-03-28 18:05:02,736][2713170] Num frames 800... [2025-03-28 18:05:02,836][2713170] Num frames 900... [2025-03-28 18:05:02,931][2713170] Num frames 1000... [2025-03-28 18:05:03,017][2713170] Num frames 1100... [2025-03-28 18:05:03,110][2713170] Num frames 1200... [2025-03-28 18:05:03,206][2713170] Num frames 1300... [2025-03-28 18:05:03,301][2713170] Num frames 1400... [2025-03-28 18:05:03,396][2713170] Num frames 1500... [2025-03-28 18:05:03,485][2713170] Num frames 1600... [2025-03-28 18:05:03,576][2713170] Avg episode rewards: #0: 19.680, true rewards: #0: 8.180 [2025-03-28 18:05:03,577][2713170] Avg episode reward: 19.680, avg true_objective: 8.180 [2025-03-28 18:05:03,652][2713170] Num frames 1700... [2025-03-28 18:05:03,755][2713170] Num frames 1800... [2025-03-28 18:05:03,854][2713170] Num frames 1900... [2025-03-28 18:05:03,950][2713170] Num frames 2000... [2025-03-28 18:05:04,041][2713170] Num frames 2100... [2025-03-28 18:05:04,137][2713170] Num frames 2200... [2025-03-28 18:05:04,233][2713170] Num frames 2300... [2025-03-28 18:05:04,331][2713170] Num frames 2400... [2025-03-28 18:05:04,427][2713170] Num frames 2500... [2025-03-28 18:05:04,528][2713170] Num frames 2600... [2025-03-28 18:05:04,628][2713170] Num frames 2700... [2025-03-28 18:05:04,721][2713170] Num frames 2800... [2025-03-28 18:05:04,808][2713170] Num frames 2900... [2025-03-28 18:05:04,893][2713170] Num frames 3000... [2025-03-28 18:05:04,979][2713170] Num frames 3100... [2025-03-28 18:05:05,064][2713170] Num frames 3200... [2025-03-28 18:05:05,152][2713170] Num frames 3300... [2025-03-28 18:05:05,236][2713170] Num frames 3400... [2025-03-28 18:05:05,319][2713170] Num frames 3500... [2025-03-28 18:05:05,403][2713170] Num frames 3600... [2025-03-28 18:05:05,487][2713170] Num frames 3700... [2025-03-28 18:05:05,578][2713170] Avg episode rewards: #0: 32.120, true rewards: #0: 12.453 [2025-03-28 18:05:05,579][2713170] Avg episode reward: 32.120, avg true_objective: 12.453 [2025-03-28 18:05:05,632][2713170] Num frames 3800... [2025-03-28 18:05:05,714][2713170] Num frames 3900... [2025-03-28 18:05:05,797][2713170] Num frames 4000... [2025-03-28 18:05:05,881][2713170] Num frames 4100... [2025-03-28 18:05:05,965][2713170] Num frames 4200... [2025-03-28 18:05:06,049][2713170] Num frames 4300... [2025-03-28 18:05:06,131][2713170] Num frames 4400... [2025-03-28 18:05:06,214][2713170] Num frames 4500... [2025-03-28 18:05:06,298][2713170] Num frames 4600... [2025-03-28 18:05:06,380][2713170] Num frames 4700... [2025-03-28 18:05:06,464][2713170] Num frames 4800... [2025-03-28 18:05:06,549][2713170] Num frames 4900... [2025-03-28 18:05:06,624][2713170] Avg episode rewards: #0: 31.050, true rewards: #0: 12.300 [2025-03-28 18:05:06,625][2713170] Avg episode reward: 31.050, avg true_objective: 12.300 [2025-03-28 18:05:06,690][2713170] Num frames 5000... [2025-03-28 18:05:06,772][2713170] Num frames 5100... [2025-03-28 18:05:06,854][2713170] Num frames 5200... [2025-03-28 18:05:06,937][2713170] Num frames 5300... [2025-03-28 18:05:07,018][2713170] Num frames 5400... [2025-03-28 18:05:07,098][2713170] Num frames 5500... [2025-03-28 18:05:07,181][2713170] Num frames 5600... [2025-03-28 18:05:07,266][2713170] Num frames 5700... [2025-03-28 18:05:07,365][2713170] Avg episode rewards: #0: 27.904, true rewards: #0: 11.504 [2025-03-28 18:05:07,366][2713170] Avg episode reward: 27.904, avg true_objective: 11.504 [2025-03-28 18:05:07,418][2713170] Num frames 5800... [2025-03-28 18:05:07,501][2713170] Num frames 5900... [2025-03-28 18:05:07,580][2713170] Num frames 6000... [2025-03-28 18:05:07,662][2713170] Num frames 6100... [2025-03-28 18:05:07,744][2713170] Num frames 6200... [2025-03-28 18:05:07,826][2713170] Num frames 6300... [2025-03-28 18:05:07,905][2713170] Avg episode rewards: #0: 25.047, true rewards: #0: 10.547 [2025-03-28 18:05:07,906][2713170] Avg episode reward: 25.047, avg true_objective: 10.547 [2025-03-28 18:05:07,965][2713170] Num frames 6400... [2025-03-28 18:05:08,048][2713170] Num frames 6500... [2025-03-28 18:05:08,130][2713170] Num frames 6600... [2025-03-28 18:05:08,213][2713170] Num frames 6700... [2025-03-28 18:05:08,295][2713170] Num frames 6800... [2025-03-28 18:05:08,377][2713170] Num frames 6900... [2025-03-28 18:05:08,464][2713170] Num frames 7000... [2025-03-28 18:05:08,515][2713170] Avg episode rewards: #0: 23.571, true rewards: #0: 10.000 [2025-03-28 18:05:08,516][2713170] Avg episode reward: 23.571, avg true_objective: 10.000 [2025-03-28 18:05:08,617][2713170] Num frames 7100... [2025-03-28 18:05:08,700][2713170] Num frames 7200... [2025-03-28 18:05:08,783][2713170] Num frames 7300... [2025-03-28 18:05:08,866][2713170] Num frames 7400... [2025-03-28 18:05:08,948][2713170] Num frames 7500... [2025-03-28 18:05:09,030][2713170] Num frames 7600... [2025-03-28 18:05:09,094][2713170] Avg episode rewards: #0: 22.135, true rewards: #0: 9.510 [2025-03-28 18:05:09,095][2713170] Avg episode reward: 22.135, avg true_objective: 9.510 [2025-03-28 18:05:09,191][2713170] Num frames 7700... [2025-03-28 18:05:09,270][2713170] Num frames 7800... [2025-03-28 18:05:09,351][2713170] Num frames 7900... [2025-03-28 18:05:09,430][2713170] Num frames 8000... [2025-03-28 18:05:09,509][2713170] Num frames 8100... [2025-03-28 18:05:09,591][2713170] Num frames 8200... [2025-03-28 18:05:09,712][2713170] Avg episode rewards: #0: 21.089, true rewards: #0: 9.200 [2025-03-28 18:05:09,712][2713170] Avg episode reward: 21.089, avg true_objective: 9.200 [2025-03-28 18:05:09,730][2713170] Num frames 8300... [2025-03-28 18:05:09,812][2713170] Num frames 8400... [2025-03-28 18:05:09,897][2713170] Num frames 8500... [2025-03-28 18:05:09,978][2713170] Num frames 8600... [2025-03-28 18:05:10,061][2713170] Num frames 8700... [2025-03-28 18:05:10,144][2713170] Num frames 8800... [2025-03-28 18:05:10,227][2713170] Num frames 8900... [2025-03-28 18:05:10,299][2713170] Avg episode rewards: #0: 20.120, true rewards: #0: 8.920 [2025-03-28 18:05:10,300][2713170] Avg episode reward: 20.120, avg true_objective: 8.920 [2025-03-28 18:05:14,141][2713170] Replay video saved to /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/replay.mp4! [2025-03-28 18:06:26,046][2713170] Loading existing experiment configuration from /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/config.json [2025-03-28 18:06:26,047][2713170] Overriding arg 'num_workers' with value 1 passed from command line [2025-03-28 18:06:26,048][2713170] Adding new argument 'no_render'=True that is not in the saved config file! [2025-03-28 18:06:26,048][2713170] Adding new argument 'save_video'=True that is not in the saved config file! [2025-03-28 18:06:26,049][2713170] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-03-28 18:06:26,050][2713170] Adding new argument 'video_name'=None that is not in the saved config file! [2025-03-28 18:06:26,050][2713170] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-03-28 18:06:26,051][2713170] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-03-28 18:06:26,052][2713170] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-03-28 18:06:26,053][2713170] Adding new argument 'hf_repository'='stalaei/DeepRL_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-03-28 18:06:26,053][2713170] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-03-28 18:06:26,054][2713170] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-03-28 18:06:26,055][2713170] Adding new argument 'train_script'=None that is not in the saved config file! [2025-03-28 18:06:26,056][2713170] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-03-28 18:06:26,057][2713170] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-03-28 18:06:26,076][2713170] RunningMeanStd input shape: (3, 72, 128) [2025-03-28 18:06:26,077][2713170] RunningMeanStd input shape: (1,) [2025-03-28 18:06:26,087][2713170] ConvEncoder: input_channels=3 [2025-03-28 18:06:26,120][2713170] Conv encoder output size: 512 [2025-03-28 18:06:26,120][2713170] Policy head output size: 512 [2025-03-28 18:06:26,659][2713170] Num frames 100... [2025-03-28 18:06:26,742][2713170] Num frames 200... [2025-03-28 18:06:26,824][2713170] Num frames 300... [2025-03-28 18:06:26,906][2713170] Num frames 400... [2025-03-28 18:06:26,989][2713170] Num frames 500... [2025-03-28 18:06:27,081][2713170] Num frames 600... [2025-03-28 18:06:27,176][2713170] Num frames 700... [2025-03-28 18:06:27,276][2713170] Num frames 800... [2025-03-28 18:06:27,328][2713170] Avg episode rewards: #0: 14.000, true rewards: #0: 8.000 [2025-03-28 18:06:27,329][2713170] Avg episode reward: 14.000, avg true_objective: 8.000 [2025-03-28 18:06:27,455][2713170] Num frames 900... [2025-03-28 18:06:27,551][2713170] Num frames 1000... [2025-03-28 18:06:27,646][2713170] Num frames 1100... [2025-03-28 18:06:27,738][2713170] Num frames 1200... [2025-03-28 18:06:27,807][2713170] Avg episode rewards: #0: 10.080, true rewards: #0: 6.080 [2025-03-28 18:06:27,808][2713170] Avg episode reward: 10.080, avg true_objective: 6.080 [2025-03-28 18:06:27,915][2713170] Num frames 1300... [2025-03-28 18:06:28,010][2713170] Num frames 1400... [2025-03-28 18:06:28,106][2713170] Num frames 1500... [2025-03-28 18:06:28,203][2713170] Num frames 1600... [2025-03-28 18:06:28,296][2713170] Num frames 1700... [2025-03-28 18:06:28,393][2713170] Num frames 1800... [2025-03-28 18:06:28,487][2713170] Num frames 1900... [2025-03-28 18:06:28,583][2713170] Num frames 2000... [2025-03-28 18:06:28,679][2713170] Num frames 2100... [2025-03-28 18:06:28,781][2713170] Avg episode rewards: #0: 12.827, true rewards: #0: 7.160 [2025-03-28 18:06:28,781][2713170] Avg episode reward: 12.827, avg true_objective: 7.160 [2025-03-28 18:06:28,852][2713170] Num frames 2200... [2025-03-28 18:06:28,950][2713170] Num frames 2300... [2025-03-28 18:06:29,048][2713170] Num frames 2400... [2025-03-28 18:06:29,142][2713170] Num frames 2500... [2025-03-28 18:06:29,238][2713170] Num frames 2600... [2025-03-28 18:06:29,323][2713170] Num frames 2700... [2025-03-28 18:06:29,416][2713170] Num frames 2800... [2025-03-28 18:06:29,516][2713170] Num frames 2900... [2025-03-28 18:06:29,612][2713170] Num frames 3000... [2025-03-28 18:06:29,708][2713170] Num frames 3100... [2025-03-28 18:06:29,806][2713170] Num frames 3200... [2025-03-28 18:06:29,945][2713170] Avg episode rewards: #0: 16.683, true rewards: #0: 8.182 [2025-03-28 18:06:29,946][2713170] Avg episode reward: 16.683, avg true_objective: 8.182 [2025-03-28 18:06:29,994][2713170] Num frames 3300... [2025-03-28 18:06:30,112][2713170] Num frames 3400... [2025-03-28 18:06:30,208][2713170] Num frames 3500... [2025-03-28 18:06:30,296][2713170] Num frames 3600... [2025-03-28 18:06:30,390][2713170] Num frames 3700... [2025-03-28 18:06:30,497][2713170] Avg episode rewards: #0: 14.906, true rewards: #0: 7.506 [2025-03-28 18:06:30,498][2713170] Avg episode reward: 14.906, avg true_objective: 7.506 [2025-03-28 18:06:30,564][2713170] Num frames 3800... [2025-03-28 18:06:30,663][2713170] Num frames 3900... [2025-03-28 18:06:30,761][2713170] Num frames 4000... [2025-03-28 18:06:30,854][2713170] Num frames 4100... [2025-03-28 18:06:30,957][2713170] Num frames 4200... [2025-03-28 18:06:31,055][2713170] Num frames 4300... [2025-03-28 18:06:31,154][2713170] Num frames 4400... [2025-03-28 18:06:31,254][2713170] Num frames 4500... [2025-03-28 18:06:31,341][2713170] Num frames 4600... [2025-03-28 18:06:31,429][2713170] Num frames 4700... [2025-03-28 18:06:31,520][2713170] Num frames 4800... [2025-03-28 18:06:31,615][2713170] Num frames 4900... [2025-03-28 18:06:31,710][2713170] Num frames 5000... [2025-03-28 18:06:31,837][2713170] Avg episode rewards: #0: 17.132, true rewards: #0: 8.465 [2025-03-28 18:06:31,837][2713170] Avg episode reward: 17.132, avg true_objective: 8.465 [2025-03-28 18:06:31,858][2713170] Num frames 5100... [2025-03-28 18:06:31,946][2713170] Num frames 5200... [2025-03-28 18:06:32,030][2713170] Num frames 5300... [2025-03-28 18:06:32,115][2713170] Num frames 5400... [2025-03-28 18:06:32,200][2713170] Num frames 5500... [2025-03-28 18:06:32,285][2713170] Num frames 5600... [2025-03-28 18:06:32,382][2713170] Avg episode rewards: #0: 16.073, true rewards: #0: 8.073 [2025-03-28 18:06:32,383][2713170] Avg episode reward: 16.073, avg true_objective: 8.073 [2025-03-28 18:06:32,429][2713170] Num frames 5700... [2025-03-28 18:06:32,525][2713170] Num frames 5800... [2025-03-28 18:06:32,622][2713170] Num frames 5900... [2025-03-28 18:06:32,718][2713170] Num frames 6000... [2025-03-28 18:06:32,867][2713170] Avg episode rewards: #0: 14.749, true rewards: #0: 7.624 [2025-03-28 18:06:32,868][2713170] Avg episode reward: 14.749, avg true_objective: 7.624 [2025-03-28 18:06:32,870][2713170] Num frames 6100... [2025-03-28 18:06:32,969][2713170] Num frames 6200... [2025-03-28 18:06:33,066][2713170] Num frames 6300... [2025-03-28 18:06:33,162][2713170] Num frames 6400... [2025-03-28 18:06:33,264][2713170] Num frames 6500... [2025-03-28 18:06:33,357][2713170] Num frames 6600... [2025-03-28 18:06:33,454][2713170] Num frames 6700... [2025-03-28 18:06:33,548][2713170] Num frames 6800... [2025-03-28 18:06:33,654][2713170] Avg episode rewards: #0: 14.717, true rewards: #0: 7.606 [2025-03-28 18:06:33,655][2713170] Avg episode reward: 14.717, avg true_objective: 7.606 [2025-03-28 18:06:33,735][2713170] Num frames 6900... [2025-03-28 18:06:33,829][2713170] Num frames 7000... [2025-03-28 18:06:33,914][2713170] Num frames 7100... [2025-03-28 18:06:33,998][2713170] Num frames 7200... [2025-03-28 18:06:34,131][2713170] Avg episode rewards: #0: 14.093, true rewards: #0: 7.293 [2025-03-28 18:06:34,131][2713170] Avg episode reward: 14.093, avg true_objective: 7.293 [2025-03-28 18:06:37,271][2713170] Replay video saved to /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/replay.mp4! [2025-03-28 18:06:42,159][2713170] The model has been pushed to https://huggingface.co/stalaei/DeepRL_vizdoom_health_gathering_supreme [2025-03-28 18:07:55,657][2713170] Environment doom_basic already registered, overwriting... [2025-03-28 18:07:55,660][2713170] Environment doom_two_colors_easy already registered, overwriting... [2025-03-28 18:07:55,661][2713170] Environment doom_two_colors_hard already registered, overwriting... [2025-03-28 18:07:55,662][2713170] Environment doom_dm already registered, overwriting... [2025-03-28 18:07:55,663][2713170] Environment doom_dwango5 already registered, overwriting... [2025-03-28 18:07:55,664][2713170] Environment doom_my_way_home_flat_actions already registered, overwriting... [2025-03-28 18:07:55,665][2713170] Environment doom_defend_the_center_flat_actions already registered, overwriting... [2025-03-28 18:07:55,666][2713170] Environment doom_my_way_home already registered, overwriting... [2025-03-28 18:07:55,667][2713170] Environment doom_deadly_corridor already registered, overwriting... [2025-03-28 18:07:55,668][2713170] Environment doom_defend_the_center already registered, overwriting... [2025-03-28 18:07:55,669][2713170] Environment doom_defend_the_line already registered, overwriting... [2025-03-28 18:07:55,669][2713170] Environment doom_health_gathering already registered, overwriting... [2025-03-28 18:07:55,670][2713170] Environment doom_health_gathering_supreme already registered, overwriting... [2025-03-28 18:07:55,671][2713170] Environment doom_battle already registered, overwriting... [2025-03-28 18:07:55,672][2713170] Environment doom_battle2 already registered, overwriting... [2025-03-28 18:07:55,672][2713170] Environment doom_duel_bots already registered, overwriting... [2025-03-28 18:07:55,673][2713170] Environment doom_deathmatch_bots already registered, overwriting... [2025-03-28 18:07:55,674][2713170] Environment doom_duel already registered, overwriting... [2025-03-28 18:07:55,675][2713170] Environment doom_deathmatch_full already registered, overwriting... [2025-03-28 18:07:55,675][2713170] Environment doom_benchmark already registered, overwriting... [2025-03-28 18:07:55,676][2713170] register_encoder_factory: [2025-03-28 18:07:55,733][2713170] Loading existing experiment configuration from /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/config.json [2025-03-28 18:07:55,734][2713170] Overriding arg 'train_for_env_steps' with value 20000000 passed from command line [2025-03-28 18:07:55,742][2713170] Experiment dir /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment already exists! [2025-03-28 18:07:55,743][2713170] Resuming existing experiment from /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment... [2025-03-28 18:07:55,744][2713170] Weights and Biases integration disabled [2025-03-28 18:07:55,749][2713170] Environment var CUDA_VISIBLE_DEVICES is 0,1,2,3,4 [2025-03-28 18:07:59,954][2713170] Starting experiment with the following configuration: help=False algo=APPO env=doom_health_gathering_supreme experiment=default_experiment train_dir=/home/stalaei/projects/deep-rl-class/notebooks/unit8/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=20000000 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=be21bbf2a4a24818a3c258f91917092a29a01603 git_repo_name=https://github.com/ShayanTalaei/deep-rl-class.git [2025-03-28 18:07:59,957][2713170] Saving configuration to /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/config.json... [2025-03-28 18:08:00,030][2713170] Rollout worker 0 uses device cpu [2025-03-28 18:08:00,031][2713170] Rollout worker 1 uses device cpu [2025-03-28 18:08:00,032][2713170] Rollout worker 2 uses device cpu [2025-03-28 18:08:00,032][2713170] Rollout worker 3 uses device cpu [2025-03-28 18:08:00,033][2713170] Rollout worker 4 uses device cpu [2025-03-28 18:08:00,034][2713170] Rollout worker 5 uses device cpu [2025-03-28 18:08:00,034][2713170] Rollout worker 6 uses device cpu [2025-03-28 18:08:00,035][2713170] Rollout worker 7 uses device cpu [2025-03-28 18:08:00,075][2713170] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-03-28 18:08:00,076][2713170] InferenceWorker_p0-w0: min num requests: 2 [2025-03-28 18:08:00,671][2713170] Starting all processes... [2025-03-28 18:08:00,672][2713170] Starting process learner_proc0 [2025-03-28 18:08:00,746][2713170] Starting all processes... [2025-03-28 18:08:00,750][2713170] Starting process inference_proc0-0 [2025-03-28 18:08:00,751][2713170] Starting process rollout_proc0 [2025-03-28 18:08:00,751][2713170] Starting process rollout_proc1 [2025-03-28 18:08:00,753][2713170] Starting process rollout_proc2 [2025-03-28 18:08:00,754][2713170] Starting process rollout_proc3 [2025-03-28 18:08:00,756][2713170] Starting process rollout_proc4 [2025-03-28 18:08:00,758][2713170] Starting process rollout_proc5 [2025-03-28 18:08:00,759][2713170] Starting process rollout_proc6 [2025-03-28 18:08:00,763][2713170] Starting process rollout_proc7 [2025-03-28 18:08:03,436][2761553] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-03-28 18:08:03,436][2761553] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0 [2025-03-28 18:08:03,498][2761576] Worker 1 uses CPU cores [32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63] [2025-03-28 18:08:03,528][2761582] Worker 7 uses CPU cores [224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255] [2025-03-28 18:08:03,552][2761581] Worker 6 uses CPU cores [192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223] [2025-03-28 18:08:03,554][2761553] Num visible devices: 1 [2025-03-28 18:08:03,555][2761553] Starting seed is not provided [2025-03-28 18:08:03,555][2761553] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-03-28 18:08:03,556][2761553] Initializing actor-critic model on device cuda:0 [2025-03-28 18:08:03,556][2761553] RunningMeanStd input shape: (3, 72, 128) [2025-03-28 18:08:03,575][2761578] Worker 4 uses CPU cores [128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159] [2025-03-28 18:08:03,582][2761553] RunningMeanStd input shape: (1,) [2025-03-28 18:08:03,583][2761574] Worker 0 uses CPU cores [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31] [2025-03-28 18:08:03,597][2761553] ConvEncoder: input_channels=3 [2025-03-28 18:08:03,618][2761579] Worker 5 uses CPU cores [160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191] [2025-03-28 18:08:03,629][2761575] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-03-28 18:08:03,629][2761575] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0 [2025-03-28 18:08:03,629][2761580] Worker 3 uses CPU cores [96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127] [2025-03-28 18:08:03,679][2761577] Worker 2 uses CPU cores [64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95] [2025-03-28 18:08:03,688][2761575] Num visible devices: 1 [2025-03-28 18:08:03,695][2761553] Conv encoder output size: 512 [2025-03-28 18:08:03,695][2761553] Policy head output size: 512 [2025-03-28 18:08:03,707][2761553] Created Actor Critic model with architecture: [2025-03-28 18:08:03,707][2761553] 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) ) ) [2025-03-28 18:08:04,133][2761553] Using optimizer [2025-03-28 18:08:05,697][2761553] Loading state from checkpoint /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 18:08:05,699][2761553] Could not load from checkpoint, attempt 0 Traceback (most recent call last): File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/torch/serialization.py", line 1470, 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([scalar])` or the `torch.serialization.safe_globals([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. [2025-03-28 18:08:05,700][2761553] Loading state from checkpoint /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 18:08:05,701][2761553] Could not load from checkpoint, attempt 1 Traceback (most recent call last): File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/torch/serialization.py", line 1470, 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([scalar])` or the `torch.serialization.safe_globals([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. [2025-03-28 18:08:05,701][2761553] Loading state from checkpoint /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... [2025-03-28 18:08:05,701][2761553] Could not load from checkpoint, attempt 2 Traceback (most recent call last): File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/sample_factory/algo/learning/learner.py", line 281, in load_checkpoint checkpoint_dict = torch.load(latest_checkpoint, map_location=device) File "/home/stalaei/miniconda3/envs/DeepRL/lib/python3.10/site-packages/torch/serialization.py", line 1470, 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([scalar])` or the `torch.serialization.safe_globals([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. [2025-03-28 18:08:05,702][2761553] Did not load from checkpoint, starting from scratch! [2025-03-28 18:08:05,702][2761553] Initialized policy 0 weights for model version 0 [2025-03-28 18:08:06,198][2761553] LearnerWorker_p0 finished initialization! [2025-03-28 18:08:06,198][2761553] Using GPUs [0] for process 0 (actually maps to GPUs [0]) [2025-03-28 18:08:06,532][2761575] RunningMeanStd input shape: (3, 72, 128) [2025-03-28 18:08:06,539][2761575] RunningMeanStd input shape: (1,) [2025-03-28 18:08:06,548][2761575] ConvEncoder: input_channels=3 [2025-03-28 18:08:06,623][2761575] Conv encoder output size: 512 [2025-03-28 18:08:06,623][2761575] Policy head output size: 512 [2025-03-28 18:08:06,733][2713170] Inference worker 0-0 is ready! [2025-03-28 18:08:06,734][2713170] All inference workers are ready! Signal rollout workers to start! [2025-03-28 18:08:06,770][2761582] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 18:08:06,770][2761580] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 18:08:06,781][2761577] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 18:08:06,791][2761581] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 18:08:06,799][2761574] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 18:08:06,799][2761578] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 18:08:06,801][2761579] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 18:08:06,809][2761576] Doom resolution: 160x120, resize resolution: (128, 72) [2025-03-28 18:08:07,201][2761580] Decorrelating experience for 0 frames... [2025-03-28 18:08:07,217][2761582] Decorrelating experience for 0 frames... [2025-03-28 18:08:07,247][2761574] Decorrelating experience for 0 frames... [2025-03-28 18:08:07,248][2761578] Decorrelating experience for 0 frames... [2025-03-28 18:08:07,248][2761576] Decorrelating experience for 0 frames... [2025-03-28 18:08:07,250][2761577] Decorrelating experience for 0 frames... [2025-03-28 18:08:07,661][2761576] Decorrelating experience for 32 frames... [2025-03-28 18:08:07,662][2761578] Decorrelating experience for 32 frames... [2025-03-28 18:08:07,663][2761581] Decorrelating experience for 0 frames... [2025-03-28 18:08:07,676][2761580] Decorrelating experience for 32 frames... [2025-03-28 18:08:07,676][2761579] Decorrelating experience for 0 frames... [2025-03-28 18:08:08,085][2761579] Decorrelating experience for 32 frames... [2025-03-28 18:08:08,098][2761581] Decorrelating experience for 32 frames... [2025-03-28 18:08:08,107][2761574] Decorrelating experience for 32 frames... [2025-03-28 18:08:08,142][2761577] Decorrelating experience for 32 frames... [2025-03-28 18:08:08,147][2761576] Decorrelating experience for 64 frames... [2025-03-28 18:08:08,148][2761582] Decorrelating experience for 32 frames... [2025-03-28 18:08:08,153][2761580] Decorrelating experience for 64 frames... [2025-03-28 18:08:08,475][2761578] Decorrelating experience for 64 frames... [2025-03-28 18:08:08,548][2761579] Decorrelating experience for 64 frames... [2025-03-28 18:08:08,561][2761574] Decorrelating experience for 64 frames... [2025-03-28 18:08:08,562][2761576] Decorrelating experience for 96 frames... [2025-03-28 18:08:08,577][2761581] Decorrelating experience for 64 frames... [2025-03-28 18:08:08,876][2761578] Decorrelating experience for 96 frames... [2025-03-28 18:08:08,984][2761579] Decorrelating experience for 96 frames... [2025-03-28 18:08:08,998][2761582] Decorrelating experience for 64 frames... [2025-03-28 18:08:08,999][2761580] Decorrelating experience for 96 frames... [2025-03-28 18:08:09,004][2761581] Decorrelating experience for 96 frames... [2025-03-28 18:08:09,428][2761574] Decorrelating experience for 96 frames... [2025-03-28 18:08:09,429][2761577] Decorrelating experience for 64 frames... [2025-03-28 18:08:09,434][2761582] Decorrelating experience for 96 frames... [2025-03-28 18:08:09,835][2761553] Signal inference workers to stop experience collection... [2025-03-28 18:08:09,838][2761575] InferenceWorker_p0-w0: stopping experience collection [2025-03-28 18:08:09,851][2761577] Decorrelating experience for 96 frames... [2025-03-28 18:08:10,749][2713170] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 0. Throughput: 0: nan. Samples: 1940. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) [2025-03-28 18:08:10,750][2713170] Avg episode reward: [(0, '2.092')] [2025-03-28 18:08:12,053][2761553] Signal inference workers to resume experience collection... [2025-03-28 18:08:12,054][2761575] InferenceWorker_p0-w0: resuming experience collection [2025-03-28 18:08:13,204][2761575] Updated weights for policy 0, policy_version 10 (0.0062) [2025-03-28 18:08:14,532][2761575] Updated weights for policy 0, policy_version 20 (0.0006) [2025-03-28 18:08:15,749][2713170] Fps is (10 sec: 22118.3, 60 sec: 22118.3, 300 sec: 22118.3). Total num frames: 110592. Throughput: 0: 1034.4. Samples: 7112. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-03-28 18:08:15,751][2713170] Avg episode reward: [(0, '4.391')] [2025-03-28 18:08:15,756][2761553] Saving new best policy, reward=4.391! [2025-03-28 18:08:16,436][2761575] Updated weights for policy 0, policy_version 30 (0.0006) [2025-03-28 18:08:17,836][2761575] Updated weights for policy 0, policy_version 40 (0.0006) [2025-03-28 18:08:19,264][2761575] Updated weights for policy 0, policy_version 50 (0.0006) [2025-03-28 18:08:20,062][2713170] Heartbeat connected on Batcher_0 [2025-03-28 18:08:20,072][2713170] Heartbeat connected on LearnerWorker_p0 [2025-03-28 18:08:20,076][2713170] Heartbeat connected on InferenceWorker_p0-w0 [2025-03-28 18:08:20,088][2713170] Heartbeat connected on RolloutWorker_w1 [2025-03-28 18:08:20,089][2713170] Heartbeat connected on RolloutWorker_w0 [2025-03-28 18:08:20,093][2713170] Heartbeat connected on RolloutWorker_w2 [2025-03-28 18:08:20,098][2713170] Heartbeat connected on RolloutWorker_w3 [2025-03-28 18:08:20,100][2713170] Heartbeat connected on RolloutWorker_w4 [2025-03-28 18:08:20,664][2713170] Heartbeat connected on RolloutWorker_w5 [2025-03-28 18:08:20,669][2713170] Heartbeat connected on RolloutWorker_w6 [2025-03-28 18:08:20,680][2713170] Heartbeat connected on RolloutWorker_w7 [2025-03-28 18:08:20,719][2761575] Updated weights for policy 0, policy_version 60 (0.0006) [2025-03-28 18:08:20,749][2713170] Fps is (10 sec: 24576.4, 60 sec: 24576.4, 300 sec: 24576.4). Total num frames: 245760. Throughput: 0: 4571.7. Samples: 47656. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:08:20,750][2713170] Avg episode reward: [(0, '4.609')] [2025-03-28 18:08:20,751][2761553] Saving new best policy, reward=4.609! [2025-03-28 18:08:22,242][2761575] Updated weights for policy 0, policy_version 70 (0.0006) [2025-03-28 18:08:23,738][2761575] Updated weights for policy 0, policy_version 80 (0.0006) [2025-03-28 18:08:25,198][2761575] Updated weights for policy 0, policy_version 90 (0.0007) [2025-03-28 18:08:25,749][2713170] Fps is (10 sec: 27033.7, 60 sec: 25395.2, 300 sec: 25395.2). Total num frames: 380928. Throughput: 0: 5858.5. Samples: 89818. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:08:25,751][2713170] Avg episode reward: [(0, '4.515')] [2025-03-28 18:08:26,774][2761575] Updated weights for policy 0, policy_version 100 (0.0006) [2025-03-28 18:08:28,294][2761575] Updated weights for policy 0, policy_version 110 (0.0006) [2025-03-28 18:08:29,875][2761575] Updated weights for policy 0, policy_version 120 (0.0006) [2025-03-28 18:08:30,749][2713170] Fps is (10 sec: 25804.7, 60 sec: 25190.6, 300 sec: 25190.6). Total num frames: 503808. Throughput: 0: 6265.3. Samples: 127246. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:08:30,750][2713170] Avg episode reward: [(0, '4.567')] [2025-03-28 18:08:31,694][2761575] Updated weights for policy 0, policy_version 130 (0.0006) [2025-03-28 18:08:33,141][2761575] Updated weights for policy 0, policy_version 140 (0.0006) [2025-03-28 18:08:34,722][2761575] Updated weights for policy 0, policy_version 150 (0.0006) [2025-03-28 18:08:35,749][2713170] Fps is (10 sec: 26214.4, 60 sec: 25722.9, 300 sec: 25722.9). Total num frames: 643072. Throughput: 0: 5833.7. Samples: 147782. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:08:35,750][2713170] Avg episode reward: [(0, '4.442')] [2025-03-28 18:08:36,223][2761575] Updated weights for policy 0, policy_version 160 (0.0006) [2025-03-28 18:08:37,796][2761575] Updated weights for policy 0, policy_version 170 (0.0006) [2025-03-28 18:08:39,312][2761575] Updated weights for policy 0, policy_version 180 (0.0006) [2025-03-28 18:08:40,749][2713170] Fps is (10 sec: 27033.4, 60 sec: 25804.9, 300 sec: 25804.9). Total num frames: 774144. Throughput: 0: 6202.1. Samples: 188002. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:08:40,750][2713170] Avg episode reward: [(0, '4.381')] [2025-03-28 18:08:40,807][2761575] Updated weights for policy 0, policy_version 190 (0.0007) [2025-03-28 18:08:42,363][2761575] Updated weights for policy 0, policy_version 200 (0.0007) [2025-03-28 18:08:43,838][2761575] Updated weights for policy 0, policy_version 210 (0.0006) [2025-03-28 18:08:45,376][2761575] Updated weights for policy 0, policy_version 220 (0.0007) [2025-03-28 18:08:45,749][2713170] Fps is (10 sec: 26624.4, 60 sec: 25980.5, 300 sec: 25980.5). Total num frames: 909312. Throughput: 0: 6468.4. Samples: 228332. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:08:45,750][2713170] Avg episode reward: [(0, '4.580')] [2025-03-28 18:08:46,927][2761575] Updated weights for policy 0, policy_version 230 (0.0006) [2025-03-28 18:08:48,481][2761575] Updated weights for policy 0, policy_version 240 (0.0006) [2025-03-28 18:08:50,030][2761575] Updated weights for policy 0, policy_version 250 (0.0006) [2025-03-28 18:08:50,749][2713170] Fps is (10 sec: 26624.2, 60 sec: 26009.7, 300 sec: 26009.7). Total num frames: 1040384. Throughput: 0: 6154.3. Samples: 248110. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:08:50,751][2713170] Avg episode reward: [(0, '4.720')] [2025-03-28 18:08:50,752][2761553] Saving new best policy, reward=4.720! [2025-03-28 18:08:51,570][2761575] Updated weights for policy 0, policy_version 260 (0.0007) [2025-03-28 18:08:53,044][2761575] Updated weights for policy 0, policy_version 270 (0.0006) [2025-03-28 18:08:54,522][2761575] Updated weights for policy 0, policy_version 280 (0.0006) [2025-03-28 18:08:55,749][2713170] Fps is (10 sec: 26623.6, 60 sec: 26123.4, 300 sec: 26123.4). Total num frames: 1175552. Throughput: 0: 6373.9. Samples: 288764. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:08:55,750][2713170] Avg episode reward: [(0, '4.768')] [2025-03-28 18:08:55,769][2761553] Saving new best policy, reward=4.768! [2025-03-28 18:08:56,096][2761575] Updated weights for policy 0, policy_version 290 (0.0007) [2025-03-28 18:08:57,583][2761575] Updated weights for policy 0, policy_version 300 (0.0006) [2025-03-28 18:08:59,084][2761575] Updated weights for policy 0, policy_version 310 (0.0006) [2025-03-28 18:09:00,749][2713170] Fps is (10 sec: 26623.9, 60 sec: 26132.6, 300 sec: 26132.6). Total num frames: 1306624. Throughput: 0: 7108.7. Samples: 327002. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-03-28 18:09:00,750][2713170] Avg episode reward: [(0, '4.578')] [2025-03-28 18:09:00,892][2761575] Updated weights for policy 0, policy_version 320 (0.0008) [2025-03-28 18:09:02,440][2761575] Updated weights for policy 0, policy_version 330 (0.0006) [2025-03-28 18:09:03,890][2761575] Updated weights for policy 0, policy_version 340 (0.0007) [2025-03-28 18:09:05,437][2761575] Updated weights for policy 0, policy_version 350 (0.0006) [2025-03-28 18:09:05,749][2713170] Fps is (10 sec: 26624.2, 60 sec: 26214.5, 300 sec: 26214.5). Total num frames: 1441792. Throughput: 0: 6665.3. Samples: 347596. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-03-28 18:09:05,750][2713170] Avg episode reward: [(0, '4.792')] [2025-03-28 18:09:05,754][2761553] Saving new best policy, reward=4.792! [2025-03-28 18:09:07,011][2761575] Updated weights for policy 0, policy_version 360 (0.0006) [2025-03-28 18:09:08,543][2761575] Updated weights for policy 0, policy_version 370 (0.0006) [2025-03-28 18:09:10,066][2761575] Updated weights for policy 0, policy_version 380 (0.0006) [2025-03-28 18:09:10,749][2713170] Fps is (10 sec: 25394.9, 60 sec: 26009.6, 300 sec: 26009.6). Total num frames: 1560576. Throughput: 0: 6614.2. Samples: 387456. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-03-28 18:09:10,751][2713170] Avg episode reward: [(0, '5.040')] [2025-03-28 18:09:10,752][2761553] Saving new best policy, reward=5.040! [2025-03-28 18:09:12,754][2761575] Updated weights for policy 0, policy_version 390 (0.0006) [2025-03-28 18:09:14,116][2761575] Updated weights for policy 0, policy_version 400 (0.0006) [2025-03-28 18:09:15,665][2761575] Updated weights for policy 0, policy_version 410 (0.0006) [2025-03-28 18:09:15,749][2713170] Fps is (10 sec: 23756.9, 60 sec: 26146.2, 300 sec: 25836.4). Total num frames: 1679360. Throughput: 0: 6527.7. Samples: 420992. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:09:15,750][2713170] Avg episode reward: [(0, '4.568')] [2025-03-28 18:09:17,179][2761575] Updated weights for policy 0, policy_version 420 (0.0006) [2025-03-28 18:09:18,690][2761575] Updated weights for policy 0, policy_version 430 (0.0006) [2025-03-28 18:09:20,165][2761575] Updated weights for policy 0, policy_version 440 (0.0006) [2025-03-28 18:09:20,749][2713170] Fps is (10 sec: 25395.6, 60 sec: 26146.1, 300 sec: 25921.9). Total num frames: 1814528. Throughput: 0: 6523.8. Samples: 441350. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:09:20,750][2713170] Avg episode reward: [(0, '4.834')] [2025-03-28 18:09:21,640][2761575] Updated weights for policy 0, policy_version 450 (0.0007) [2025-03-28 18:09:23,138][2761575] Updated weights for policy 0, policy_version 460 (0.0006) [2025-03-28 18:09:24,567][2761575] Updated weights for policy 0, policy_version 470 (0.0006) [2025-03-28 18:09:25,749][2713170] Fps is (10 sec: 27852.7, 60 sec: 26282.7, 300 sec: 26105.2). Total num frames: 1957888. Throughput: 0: 6562.1. Samples: 483298. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:09:25,750][2713170] Avg episode reward: [(0, '4.767')] [2025-03-28 18:09:26,026][2761575] Updated weights for policy 0, policy_version 480 (0.0006) [2025-03-28 18:09:27,537][2761575] Updated weights for policy 0, policy_version 490 (0.0006) [2025-03-28 18:09:29,060][2761575] Updated weights for policy 0, policy_version 500 (0.0006) [2025-03-28 18:09:30,749][2713170] Fps is (10 sec: 26623.9, 60 sec: 26282.7, 300 sec: 26009.6). Total num frames: 2080768. Throughput: 0: 6491.6. Samples: 520454. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:09:30,750][2713170] Avg episode reward: [(0, '4.612')] [2025-03-28 18:09:31,055][2761575] Updated weights for policy 0, policy_version 510 (0.0008) [2025-03-28 18:09:32,617][2761575] Updated weights for policy 0, policy_version 520 (0.0006) [2025-03-28 18:09:34,152][2761575] Updated weights for policy 0, policy_version 530 (0.0006) [2025-03-28 18:09:35,656][2761575] Updated weights for policy 0, policy_version 540 (0.0006) [2025-03-28 18:09:35,749][2713170] Fps is (10 sec: 25395.3, 60 sec: 26146.2, 300 sec: 26021.7). Total num frames: 2211840. Throughput: 0: 6502.2. Samples: 540710. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:09:35,750][2713170] Avg episode reward: [(0, '4.914')] [2025-03-28 18:09:37,189][2761575] Updated weights for policy 0, policy_version 550 (0.0006) [2025-03-28 18:09:38,700][2761575] Updated weights for policy 0, policy_version 560 (0.0006) [2025-03-28 18:09:40,235][2761575] Updated weights for policy 0, policy_version 570 (0.0006) [2025-03-28 18:09:40,749][2713170] Fps is (10 sec: 26624.1, 60 sec: 26214.4, 300 sec: 26077.9). Total num frames: 2347008. Throughput: 0: 6497.7. Samples: 581158. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:09:40,750][2713170] Avg episode reward: [(0, '4.892')] [2025-03-28 18:09:41,772][2761575] Updated weights for policy 0, policy_version 580 (0.0006) [2025-03-28 18:09:43,302][2761575] Updated weights for policy 0, policy_version 590 (0.0006) [2025-03-28 18:09:44,866][2761575] Updated weights for policy 0, policy_version 600 (0.0006) [2025-03-28 18:09:45,749][2713170] Fps is (10 sec: 26623.8, 60 sec: 26146.1, 300 sec: 26085.1). Total num frames: 2478080. Throughput: 0: 6528.9. Samples: 620802. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-03-28 18:09:45,750][2713170] Avg episode reward: [(0, '4.685')] [2025-03-28 18:09:46,391][2761575] Updated weights for policy 0, policy_version 610 (0.0006) [2025-03-28 18:09:47,967][2761575] Updated weights for policy 0, policy_version 620 (0.0006) [2025-03-28 18:09:49,521][2761575] Updated weights for policy 0, policy_version 630 (0.0006) [2025-03-28 18:09:50,749][2713170] Fps is (10 sec: 26214.4, 60 sec: 26146.1, 300 sec: 26091.6). Total num frames: 2609152. Throughput: 0: 6511.8. Samples: 640626. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:09:50,750][2713170] Avg episode reward: [(0, '4.976')] [2025-03-28 18:09:51,091][2761575] Updated weights for policy 0, policy_version 640 (0.0006) [2025-03-28 18:09:52,602][2761575] Updated weights for policy 0, policy_version 650 (0.0006) [2025-03-28 18:09:54,165][2761575] Updated weights for policy 0, policy_version 660 (0.0006) [2025-03-28 18:09:55,601][2761575] Updated weights for policy 0, policy_version 670 (0.0006) [2025-03-28 18:09:55,749][2713170] Fps is (10 sec: 26624.1, 60 sec: 26146.2, 300 sec: 26136.4). Total num frames: 2744320. Throughput: 0: 6514.9. Samples: 680628. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-03-28 18:09:55,750][2713170] Avg episode reward: [(0, '5.383')] [2025-03-28 18:09:55,756][2761553] Saving /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000671_2748416.pth... [2025-03-28 18:09:55,867][2761553] Removing /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000659_2699264.pth [2025-03-28 18:09:55,874][2761553] Saving new best policy, reward=5.383! [2025-03-28 18:09:57,073][2761575] Updated weights for policy 0, policy_version 680 (0.0006) [2025-03-28 18:09:58,549][2761575] Updated weights for policy 0, policy_version 690 (0.0006) [2025-03-28 18:10:00,749][2713170] Fps is (10 sec: 25394.9, 60 sec: 25941.3, 300 sec: 26028.2). Total num frames: 2863104. Throughput: 0: 6597.7. Samples: 717890. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:10:00,750][2713170] Avg episode reward: [(0, '5.080')] [2025-03-28 18:10:00,758][2761575] Updated weights for policy 0, policy_version 700 (0.0007) [2025-03-28 18:10:02,217][2761575] Updated weights for policy 0, policy_version 710 (0.0007) [2025-03-28 18:10:03,811][2761575] Updated weights for policy 0, policy_version 720 (0.0006) [2025-03-28 18:10:05,314][2761575] Updated weights for policy 0, policy_version 730 (0.0006) [2025-03-28 18:10:05,749][2713170] Fps is (10 sec: 25804.9, 60 sec: 26009.6, 300 sec: 26107.6). Total num frames: 3002368. Throughput: 0: 6582.2. Samples: 737548. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:10:05,750][2713170] Avg episode reward: [(0, '4.590')] [2025-03-28 18:10:06,754][2761575] Updated weights for policy 0, policy_version 740 (0.0006) [2025-03-28 18:10:08,294][2761575] Updated weights for policy 0, policy_version 750 (0.0006) [2025-03-28 18:10:09,818][2761575] Updated weights for policy 0, policy_version 760 (0.0006) [2025-03-28 18:10:10,749][2713170] Fps is (10 sec: 25804.8, 60 sec: 26009.6, 300 sec: 26009.6). Total num frames: 3121152. Throughput: 0: 6559.1. Samples: 778456. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:10:10,750][2713170] Avg episode reward: [(0, '5.235')] [2025-03-28 18:10:12,635][2761575] Updated weights for policy 0, policy_version 770 (0.0006) [2025-03-28 18:10:14,193][2761575] Updated weights for policy 0, policy_version 780 (0.0006) [2025-03-28 18:10:15,750][2761575] Updated weights for policy 0, policy_version 790 (0.0006) [2025-03-28 18:10:15,749][2713170] Fps is (10 sec: 22937.3, 60 sec: 25873.0, 300 sec: 25854.0). Total num frames: 3231744. Throughput: 0: 6429.1. Samples: 809764. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:10:15,751][2713170] Avg episode reward: [(0, '5.369')] [2025-03-28 18:10:17,226][2761575] Updated weights for policy 0, policy_version 800 (0.0006) [2025-03-28 18:10:18,817][2761575] Updated weights for policy 0, policy_version 810 (0.0006) [2025-03-28 18:10:20,405][2761575] Updated weights for policy 0, policy_version 820 (0.0006) [2025-03-28 18:10:20,749][2713170] Fps is (10 sec: 24575.9, 60 sec: 25873.0, 300 sec: 25899.3). Total num frames: 3366912. Throughput: 0: 6417.5. Samples: 829498. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:10:20,750][2713170] Avg episode reward: [(0, '4.745')] [2025-03-28 18:10:21,864][2761575] Updated weights for policy 0, policy_version 830 (0.0006) [2025-03-28 18:10:23,400][2761575] Updated weights for policy 0, policy_version 840 (0.0006) [2025-03-28 18:10:25,086][2761575] Updated weights for policy 0, policy_version 850 (0.0006) [2025-03-28 18:10:25,749][2713170] Fps is (10 sec: 26624.1, 60 sec: 25668.2, 300 sec: 25911.0). Total num frames: 3497984. Throughput: 0: 6393.9. Samples: 868882. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:10:25,750][2713170] Avg episode reward: [(0, '5.201')] [2025-03-28 18:10:26,598][2761575] Updated weights for policy 0, policy_version 860 (0.0006) [2025-03-28 18:10:28,133][2761575] Updated weights for policy 0, policy_version 870 (0.0006) [2025-03-28 18:10:29,700][2761575] Updated weights for policy 0, policy_version 880 (0.0006) [2025-03-28 18:10:30,749][2713170] Fps is (10 sec: 25395.5, 60 sec: 25668.3, 300 sec: 25863.3). Total num frames: 3620864. Throughput: 0: 6330.9. Samples: 905692. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:10:30,751][2713170] Avg episode reward: [(0, '5.018')] [2025-03-28 18:10:31,686][2761575] Updated weights for policy 0, policy_version 890 (0.0007) [2025-03-28 18:10:33,272][2761575] Updated weights for policy 0, policy_version 900 (0.0006) [2025-03-28 18:10:34,752][2761575] Updated weights for policy 0, policy_version 910 (0.0007) [2025-03-28 18:10:35,749][2713170] Fps is (10 sec: 25395.4, 60 sec: 25668.3, 300 sec: 25875.5). Total num frames: 3751936. Throughput: 0: 6329.7. Samples: 925462. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-03-28 18:10:35,750][2713170] Avg episode reward: [(0, '5.452')] [2025-03-28 18:10:35,754][2761553] Saving new best policy, reward=5.452! [2025-03-28 18:10:36,266][2761575] Updated weights for policy 0, policy_version 920 (0.0006) [2025-03-28 18:10:37,811][2761575] Updated weights for policy 0, policy_version 930 (0.0006) [2025-03-28 18:10:39,288][2761575] Updated weights for policy 0, policy_version 940 (0.0006) [2025-03-28 18:10:40,749][2713170] Fps is (10 sec: 26624.0, 60 sec: 25668.3, 300 sec: 25914.1). Total num frames: 3887104. Throughput: 0: 6343.9. Samples: 966104. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-03-28 18:10:40,750][2713170] Avg episode reward: [(0, '5.350')] [2025-03-28 18:10:40,873][2761575] Updated weights for policy 0, policy_version 950 (0.0006) [2025-03-28 18:10:42,368][2761575] Updated weights for policy 0, policy_version 960 (0.0006) [2025-03-28 18:10:43,955][2761575] Updated weights for policy 0, policy_version 970 (0.0006) [2025-03-28 18:10:45,536][2761575] Updated weights for policy 0, policy_version 980 (0.0006) [2025-03-28 18:10:45,749][2713170] Fps is (10 sec: 26624.0, 60 sec: 25668.3, 300 sec: 25923.7). Total num frames: 4018176. Throughput: 0: 6390.7. Samples: 1005472. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:10:45,750][2713170] Avg episode reward: [(0, '5.500')] [2025-03-28 18:10:45,756][2761553] Saving new best policy, reward=5.500! [2025-03-28 18:10:47,015][2761575] Updated weights for policy 0, policy_version 990 (0.0007) [2025-03-28 18:10:48,538][2761575] Updated weights for policy 0, policy_version 1000 (0.0007) [2025-03-28 18:10:50,064][2761575] Updated weights for policy 0, policy_version 1010 (0.0007) [2025-03-28 18:10:50,749][2713170] Fps is (10 sec: 26623.7, 60 sec: 25736.5, 300 sec: 25958.4). Total num frames: 4153344. Throughput: 0: 6405.5. Samples: 1025796. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-03-28 18:10:50,750][2713170] Avg episode reward: [(0, '5.786')] [2025-03-28 18:10:50,752][2761553] Saving new best policy, reward=5.786! [2025-03-28 18:10:51,614][2761575] Updated weights for policy 0, policy_version 1020 (0.0006) [2025-03-28 18:10:53,076][2761575] Updated weights for policy 0, policy_version 1030 (0.0006) [2025-03-28 18:10:54,607][2761575] Updated weights for policy 0, policy_version 1040 (0.0007) [2025-03-28 18:10:55,749][2713170] Fps is (10 sec: 27032.7, 60 sec: 25736.4, 300 sec: 25991.0). Total num frames: 4288512. Throughput: 0: 6403.7. Samples: 1066626. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:10:55,751][2713170] Avg episode reward: [(0, '6.084')] [2025-03-28 18:10:55,761][2761553] Saving new best policy, reward=6.084! [2025-03-28 18:10:56,076][2761575] Updated weights for policy 0, policy_version 1050 (0.0006) [2025-03-28 18:10:57,553][2761575] Updated weights for policy 0, policy_version 1060 (0.0006) [2025-03-28 18:10:59,052][2761575] Updated weights for policy 0, policy_version 1070 (0.0006) [2025-03-28 18:11:00,749][2713170] Fps is (10 sec: 25395.5, 60 sec: 25736.6, 300 sec: 25925.3). Total num frames: 4407296. Throughput: 0: 6171.8. Samples: 1087492. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-03-28 18:11:00,750][2713170] Avg episode reward: [(0, '6.544')] [2025-03-28 18:11:00,847][2761553] Saving new best policy, reward=6.544! [2025-03-28 18:11:01,323][2761575] Updated weights for policy 0, policy_version 1080 (0.0008) [2025-03-28 18:11:02,906][2761575] Updated weights for policy 0, policy_version 1090 (0.0006) [2025-03-28 18:11:04,463][2761575] Updated weights for policy 0, policy_version 1100 (0.0006) [2025-03-28 18:11:05,749][2713170] Fps is (10 sec: 24986.4, 60 sec: 25600.0, 300 sec: 25933.6). Total num frames: 4538368. Throughput: 0: 6508.8. Samples: 1122394. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:11:05,750][2713170] Avg episode reward: [(0, '5.801')] [2025-03-28 18:11:06,019][2761575] Updated weights for policy 0, policy_version 1110 (0.0006) [2025-03-28 18:11:07,546][2761575] Updated weights for policy 0, policy_version 1120 (0.0006) [2025-03-28 18:11:09,087][2761575] Updated weights for policy 0, policy_version 1130 (0.0006) [2025-03-28 18:11:10,455][2761575] Updated weights for policy 0, policy_version 1140 (0.0006) [2025-03-28 18:11:10,749][2713170] Fps is (10 sec: 26624.0, 60 sec: 25873.1, 300 sec: 25964.1). Total num frames: 4673536. Throughput: 0: 6528.0. Samples: 1162640. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) [2025-03-28 18:11:10,750][2713170] Avg episode reward: [(0, '5.945')] [2025-03-28 18:11:11,886][2761575] Updated weights for policy 0, policy_version 1150 (0.0006) [2025-03-28 18:11:13,312][2761575] Updated weights for policy 0, policy_version 1160 (0.0006) [2025-03-28 18:11:14,832][2761575] Updated weights for policy 0, policy_version 1170 (0.0006) [2025-03-28 18:11:15,749][2713170] Fps is (10 sec: 27852.9, 60 sec: 26419.3, 300 sec: 26037.3). Total num frames: 4816896. Throughput: 0: 6649.0. Samples: 1204898. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:11:15,750][2713170] Avg episode reward: [(0, '6.605')] [2025-03-28 18:11:15,755][2761553] Saving new best policy, reward=6.605! [2025-03-28 18:11:16,358][2761575] Updated weights for policy 0, policy_version 1180 (0.0006) [2025-03-28 18:11:17,822][2761575] Updated weights for policy 0, policy_version 1190 (0.0006) [2025-03-28 18:11:19,335][2761575] Updated weights for policy 0, policy_version 1200 (0.0006) [2025-03-28 18:11:20,749][2713170] Fps is (10 sec: 27853.0, 60 sec: 26419.3, 300 sec: 26063.5). Total num frames: 4952064. Throughput: 0: 6661.7. Samples: 1225238. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:11:20,750][2713170] Avg episode reward: [(0, '6.253')] [2025-03-28 18:11:20,886][2761575] Updated weights for policy 0, policy_version 1210 (0.0006) [2025-03-28 18:11:22,395][2761575] Updated weights for policy 0, policy_version 1220 (0.0006) [2025-03-28 18:11:23,965][2761575] Updated weights for policy 0, policy_version 1230 (0.0006) [2025-03-28 18:11:25,388][2761575] Updated weights for policy 0, policy_version 1240 (0.0006) [2025-03-28 18:11:25,749][2713170] Fps is (10 sec: 27033.6, 60 sec: 26487.5, 300 sec: 26088.4). Total num frames: 5087232. Throughput: 0: 6654.6. Samples: 1265560. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:11:25,750][2713170] Avg episode reward: [(0, '5.850')] [2025-03-28 18:11:26,901][2761575] Updated weights for policy 0, policy_version 1250 (0.0006) [2025-03-28 18:11:28,407][2761575] Updated weights for policy 0, policy_version 1260 (0.0006) [2025-03-28 18:11:29,946][2761575] Updated weights for policy 0, policy_version 1270 (0.0006) [2025-03-28 18:11:30,749][2713170] Fps is (10 sec: 25395.0, 60 sec: 26419.2, 300 sec: 26030.1). Total num frames: 5206016. Throughput: 0: 6242.3. Samples: 1286376. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:11:30,750][2713170] Avg episode reward: [(0, '8.475')] [2025-03-28 18:11:30,974][2761553] Saving new best policy, reward=8.475! [2025-03-28 18:11:32,224][2761575] Updated weights for policy 0, policy_version 1280 (0.0008) [2025-03-28 18:11:33,736][2761575] Updated weights for policy 0, policy_version 1290 (0.0006) [2025-03-28 18:11:35,240][2761575] Updated weights for policy 0, policy_version 1300 (0.0006) [2025-03-28 18:11:35,749][2713170] Fps is (10 sec: 24985.6, 60 sec: 26419.2, 300 sec: 26034.6). Total num frames: 5337088. Throughput: 0: 6574.7. Samples: 1321656. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:11:35,750][2713170] Avg episode reward: [(0, '7.334')] [2025-03-28 18:11:36,624][2761575] Updated weights for policy 0, policy_version 1310 (0.0006) [2025-03-28 18:11:38,153][2761575] Updated weights for policy 0, policy_version 1320 (0.0006) [2025-03-28 18:11:39,706][2761575] Updated weights for policy 0, policy_version 1330 (0.0006) [2025-03-28 18:11:40,749][2713170] Fps is (10 sec: 26623.8, 60 sec: 26419.2, 300 sec: 26058.4). Total num frames: 5472256. Throughput: 0: 6581.9. Samples: 1362810. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:11:40,751][2713170] Avg episode reward: [(0, '8.143')] [2025-03-28 18:11:41,220][2761575] Updated weights for policy 0, policy_version 1340 (0.0006) [2025-03-28 18:11:42,746][2761575] Updated weights for policy 0, policy_version 1350 (0.0006) [2025-03-28 18:11:44,281][2761575] Updated weights for policy 0, policy_version 1360 (0.0006) [2025-03-28 18:11:45,730][2761575] Updated weights for policy 0, policy_version 1370 (0.0006) [2025-03-28 18:11:45,749][2713170] Fps is (10 sec: 27443.1, 60 sec: 26555.7, 300 sec: 26100.1). Total num frames: 5611520. Throughput: 0: 7024.0. Samples: 1403572. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:11:45,750][2713170] Avg episode reward: [(0, '9.668')] [2025-03-28 18:11:45,755][2761553] Saving new best policy, reward=9.668! [2025-03-28 18:11:47,185][2761575] Updated weights for policy 0, policy_version 1380 (0.0006) [2025-03-28 18:11:48,598][2761575] Updated weights for policy 0, policy_version 1390 (0.0006) [2025-03-28 18:11:50,058][2761575] Updated weights for policy 0, policy_version 1400 (0.0006) [2025-03-28 18:11:50,749][2713170] Fps is (10 sec: 27853.0, 60 sec: 26624.0, 300 sec: 26139.9). Total num frames: 5750784. Throughput: 0: 6724.3. Samples: 1424988. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:11:50,750][2713170] Avg episode reward: [(0, '9.729')] [2025-03-28 18:11:50,751][2761553] Saving new best policy, reward=9.729! [2025-03-28 18:11:51,453][2761575] Updated weights for policy 0, policy_version 1410 (0.0006) [2025-03-28 18:11:52,834][2761575] Updated weights for policy 0, policy_version 1420 (0.0006) [2025-03-28 18:11:54,310][2761575] Updated weights for policy 0, policy_version 1430 (0.0006) [2025-03-28 18:11:55,749][2713170] Fps is (10 sec: 28262.1, 60 sec: 26760.6, 300 sec: 26196.2). Total num frames: 5894144. Throughput: 0: 6784.7. Samples: 1467952. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:11:55,750][2713170] Avg episode reward: [(0, '12.318')] [2025-03-28 18:11:55,758][2761553] Saving /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000001439_5894144.pth... [2025-03-28 18:11:55,883][2761553] Removing /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000671_2748416.pth [2025-03-28 18:11:55,890][2761553] Saving new best policy, reward=12.318! [2025-03-28 18:11:56,000][2761575] Updated weights for policy 0, policy_version 1440 (0.0007) [2025-03-28 18:11:57,342][2761575] Updated weights for policy 0, policy_version 1450 (0.0006) [2025-03-28 18:11:58,849][2761575] Updated weights for policy 0, policy_version 1460 (0.0006) [2025-03-28 18:12:00,749][2713170] Fps is (10 sec: 26214.4, 60 sec: 26760.5, 300 sec: 26143.2). Total num frames: 6012928. Throughput: 0: 6296.2. Samples: 1488226. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:12:00,750][2713170] Avg episode reward: [(0, '12.366')] [2025-03-28 18:12:01,082][2761553] Saving new best policy, reward=12.366! [2025-03-28 18:12:01,232][2761575] Updated weights for policy 0, policy_version 1470 (0.0007) [2025-03-28 18:12:02,771][2761575] Updated weights for policy 0, policy_version 1480 (0.0006) [2025-03-28 18:12:04,254][2761575] Updated weights for policy 0, policy_version 1490 (0.0006) [2025-03-28 18:12:05,728][2761575] Updated weights for policy 0, policy_version 1500 (0.0006) [2025-03-28 18:12:05,749][2713170] Fps is (10 sec: 24986.0, 60 sec: 26760.5, 300 sec: 26144.7). Total num frames: 6144000. Throughput: 0: 6615.9. Samples: 1522952. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:12:05,750][2713170] Avg episode reward: [(0, '15.018')] [2025-03-28 18:12:05,754][2761553] Saving new best policy, reward=15.018! [2025-03-28 18:12:07,205][2761575] Updated weights for policy 0, policy_version 1510 (0.0006) [2025-03-28 18:12:08,691][2761575] Updated weights for policy 0, policy_version 1520 (0.0006) [2025-03-28 18:12:10,162][2761575] Updated weights for policy 0, policy_version 1530 (0.0006) [2025-03-28 18:12:10,749][2713170] Fps is (10 sec: 27033.6, 60 sec: 26828.8, 300 sec: 26180.3). Total num frames: 6283264. Throughput: 0: 6642.0. Samples: 1564452. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:12:10,750][2713170] Avg episode reward: [(0, '17.186')] [2025-03-28 18:12:10,751][2761553] Saving new best policy, reward=17.186! [2025-03-28 18:12:11,626][2761575] Updated weights for policy 0, policy_version 1540 (0.0006) [2025-03-28 18:12:13,102][2761575] Updated weights for policy 0, policy_version 1550 (0.0006) [2025-03-28 18:12:14,573][2761575] Updated weights for policy 0, policy_version 1560 (0.0006) [2025-03-28 18:12:15,749][2713170] Fps is (10 sec: 27852.8, 60 sec: 26760.5, 300 sec: 26214.4). Total num frames: 6422528. Throughput: 0: 7113.8. Samples: 1606498. Policy #0 lag: (min: 0.0, avg: 0.9, max: 2.0) [2025-03-28 18:12:15,750][2713170] Avg episode reward: [(0, '17.213')] [2025-03-28 18:12:15,754][2761553] Saving new best policy, reward=17.213! [2025-03-28 18:12:16,013][2761575] Updated weights for policy 0, policy_version 1570 (0.0006) [2025-03-28 18:12:17,431][2761575] Updated weights for policy 0, policy_version 1580 (0.0006) [2025-03-28 18:12:18,912][2761575] Updated weights for policy 0, policy_version 1590 (0.0006) [2025-03-28 18:12:20,330][2761575] Updated weights for policy 0, policy_version 1600 (0.0006) [2025-03-28 18:12:20,749][2713170] Fps is (10 sec: 27852.6, 60 sec: 26828.7, 300 sec: 26247.2). Total num frames: 6561792. Throughput: 0: 6797.4. Samples: 1627540. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:12:20,750][2713170] Avg episode reward: [(0, '17.337')] [2025-03-28 18:12:20,752][2761553] Saving new best policy, reward=17.337! [2025-03-28 18:12:21,768][2761575] Updated weights for policy 0, policy_version 1610 (0.0007) [2025-03-28 18:12:23,151][2761575] Updated weights for policy 0, policy_version 1620 (0.0007) [2025-03-28 18:12:24,578][2761575] Updated weights for policy 0, policy_version 1630 (0.0007) [2025-03-28 18:12:25,749][2713170] Fps is (10 sec: 28671.9, 60 sec: 27033.6, 300 sec: 26310.8). Total num frames: 6709248. Throughput: 0: 6849.9. Samples: 1671054. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:12:25,750][2713170] Avg episode reward: [(0, '17.790')] [2025-03-28 18:12:25,755][2761553] Saving new best policy, reward=17.790! [2025-03-28 18:12:25,923][2761575] Updated weights for policy 0, policy_version 1640 (0.0006) [2025-03-28 18:12:27,352][2761575] Updated weights for policy 0, policy_version 1650 (0.0006) [2025-03-28 18:12:28,870][2761575] Updated weights for policy 0, policy_version 1660 (0.0006) [2025-03-28 18:12:30,749][2713170] Fps is (10 sec: 27033.7, 60 sec: 27101.8, 300 sec: 26277.4). Total num frames: 6832128. Throughput: 0: 6430.9. Samples: 1692962. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:12:30,750][2713170] Avg episode reward: [(0, '21.064')] [2025-03-28 18:12:31,210][2761553] Saving new best policy, reward=21.064! [2025-03-28 18:12:31,332][2761575] Updated weights for policy 0, policy_version 1670 (0.0007) [2025-03-28 18:12:32,939][2761575] Updated weights for policy 0, policy_version 1680 (0.0006) [2025-03-28 18:12:34,426][2761575] Updated weights for policy 0, policy_version 1690 (0.0006) [2025-03-28 18:12:35,749][2713170] Fps is (10 sec: 24985.6, 60 sec: 27033.6, 300 sec: 26260.8). Total num frames: 6959104. Throughput: 0: 6698.4. Samples: 1726418. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:12:35,750][2713170] Avg episode reward: [(0, '21.138')] [2025-03-28 18:12:35,756][2761553] Saving new best policy, reward=21.138! [2025-03-28 18:12:35,864][2761575] Updated weights for policy 0, policy_version 1700 (0.0007) [2025-03-28 18:12:37,318][2761575] Updated weights for policy 0, policy_version 1710 (0.0006) [2025-03-28 18:12:38,729][2761575] Updated weights for policy 0, policy_version 1720 (0.0006) [2025-03-28 18:12:40,239][2761575] Updated weights for policy 0, policy_version 1730 (0.0006) [2025-03-28 18:12:40,749][2713170] Fps is (10 sec: 26624.1, 60 sec: 27101.9, 300 sec: 26290.3). Total num frames: 7098368. Throughput: 0: 6688.9. Samples: 1768952. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:12:40,750][2713170] Avg episode reward: [(0, '20.840')] [2025-03-28 18:12:41,700][2761575] Updated weights for policy 0, policy_version 1740 (0.0007) [2025-03-28 18:12:43,280][2761575] Updated weights for policy 0, policy_version 1750 (0.0006) [2025-03-28 18:12:44,930][2761575] Updated weights for policy 0, policy_version 1760 (0.0006) [2025-03-28 18:12:45,749][2713170] Fps is (10 sec: 26623.8, 60 sec: 26897.0, 300 sec: 26274.0). Total num frames: 7225344. Throughput: 0: 7103.0. Samples: 1807860. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-03-28 18:12:45,750][2713170] Avg episode reward: [(0, '18.316')] [2025-03-28 18:12:46,580][2761575] Updated weights for policy 0, policy_version 1770 (0.0006) [2025-03-28 18:12:48,104][2761575] Updated weights for policy 0, policy_version 1780 (0.0006) [2025-03-28 18:12:49,727][2761575] Updated weights for policy 0, policy_version 1790 (0.0006) [2025-03-28 18:12:50,749][2713170] Fps is (10 sec: 25804.6, 60 sec: 26760.5, 300 sec: 26272.9). Total num frames: 7356416. Throughput: 0: 6760.8. Samples: 1827188. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:12:50,750][2713170] Avg episode reward: [(0, '16.420')] [2025-03-28 18:12:51,277][2761575] Updated weights for policy 0, policy_version 1800 (0.0007) [2025-03-28 18:12:52,847][2761575] Updated weights for policy 0, policy_version 1810 (0.0006) [2025-03-28 18:12:54,402][2761575] Updated weights for policy 0, policy_version 1820 (0.0006) [2025-03-28 18:12:55,749][2713170] Fps is (10 sec: 26214.5, 60 sec: 26555.8, 300 sec: 26271.9). Total num frames: 7487488. Throughput: 0: 6712.7. Samples: 1866526. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:12:55,750][2713170] Avg episode reward: [(0, '20.495')] [2025-03-28 18:12:55,934][2761575] Updated weights for policy 0, policy_version 1830 (0.0006) [2025-03-28 18:12:57,542][2761575] Updated weights for policy 0, policy_version 1840 (0.0006) [2025-03-28 18:12:59,171][2761575] Updated weights for policy 0, policy_version 1850 (0.0006) [2025-03-28 18:13:00,749][2713170] Fps is (10 sec: 24166.6, 60 sec: 26419.2, 300 sec: 26200.3). Total num frames: 7598080. Throughput: 0: 6213.4. Samples: 1886100. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:13:00,750][2713170] Avg episode reward: [(0, '23.074')] [2025-03-28 18:13:01,220][2761553] Saving new best policy, reward=23.074! [2025-03-28 18:13:01,883][2761575] Updated weights for policy 0, policy_version 1860 (0.0008) [2025-03-28 18:13:03,466][2761575] Updated weights for policy 0, policy_version 1870 (0.0007) [2025-03-28 18:13:05,004][2761575] Updated weights for policy 0, policy_version 1880 (0.0006) [2025-03-28 18:13:05,749][2713170] Fps is (10 sec: 22937.6, 60 sec: 26214.4, 300 sec: 26158.9). Total num frames: 7716864. Throughput: 0: 6440.0. Samples: 1917340. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:13:05,750][2713170] Avg episode reward: [(0, '22.152')] [2025-03-28 18:13:06,591][2761575] Updated weights for policy 0, policy_version 1890 (0.0006) [2025-03-28 18:13:08,176][2761575] Updated weights for policy 0, policy_version 1900 (0.0006) [2025-03-28 18:13:09,721][2761575] Updated weights for policy 0, policy_version 1910 (0.0006) [2025-03-28 18:13:10,749][2713170] Fps is (10 sec: 24985.4, 60 sec: 26077.8, 300 sec: 26228.3). Total num frames: 7847936. Throughput: 0: 6346.0. Samples: 1956626. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:13:10,750][2713170] Avg episode reward: [(0, '19.936')] [2025-03-28 18:13:11,213][2761575] Updated weights for policy 0, policy_version 1920 (0.0006) [2025-03-28 18:13:12,629][2761575] Updated weights for policy 0, policy_version 1930 (0.0006) [2025-03-28 18:13:14,184][2761575] Updated weights for policy 0, policy_version 1940 (0.0006) [2025-03-28 18:13:15,626][2761575] Updated weights for policy 0, policy_version 1950 (0.0006) [2025-03-28 18:13:15,749][2713170] Fps is (10 sec: 27033.4, 60 sec: 26077.8, 300 sec: 26242.2). Total num frames: 7987200. Throughput: 0: 6781.3. Samples: 1998122. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:13:15,750][2713170] Avg episode reward: [(0, '23.233')] [2025-03-28 18:13:15,756][2761553] Saving new best policy, reward=23.233! [2025-03-28 18:13:17,190][2761575] Updated weights for policy 0, policy_version 1960 (0.0006) [2025-03-28 18:13:18,779][2761575] Updated weights for policy 0, policy_version 1970 (0.0007) [2025-03-28 18:13:20,284][2761575] Updated weights for policy 0, policy_version 1980 (0.0006) [2025-03-28 18:13:20,749][2713170] Fps is (10 sec: 27033.8, 60 sec: 25941.4, 300 sec: 26228.3). Total num frames: 8118272. Throughput: 0: 6474.8. Samples: 2017784. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:13:20,750][2713170] Avg episode reward: [(0, '23.993')] [2025-03-28 18:13:20,752][2761553] Saving new best policy, reward=23.993! [2025-03-28 18:13:21,840][2761575] Updated weights for policy 0, policy_version 1990 (0.0006) [2025-03-28 18:13:23,284][2761575] Updated weights for policy 0, policy_version 2000 (0.0006) [2025-03-28 18:13:24,791][2761575] Updated weights for policy 0, policy_version 2010 (0.0006) [2025-03-28 18:13:25,749][2713170] Fps is (10 sec: 27443.6, 60 sec: 25873.1, 300 sec: 26297.7). Total num frames: 8261632. Throughput: 0: 6433.2. Samples: 2058446. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-03-28 18:13:25,750][2713170] Avg episode reward: [(0, '20.369')] [2025-03-28 18:13:26,141][2761575] Updated weights for policy 0, policy_version 2020 (0.0006) [2025-03-28 18:13:27,482][2761575] Updated weights for policy 0, policy_version 2030 (0.0006) [2025-03-28 18:13:28,811][2761575] Updated weights for policy 0, policy_version 2040 (0.0006) [2025-03-28 18:13:30,749][2713170] Fps is (10 sec: 27033.4, 60 sec: 25941.3, 300 sec: 26256.1). Total num frames: 8388608. Throughput: 0: 6075.7. Samples: 2081266. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:13:30,750][2713170] Avg episode reward: [(0, '25.474')] [2025-03-28 18:13:31,235][2761553] Saving new best policy, reward=25.474! [2025-03-28 18:13:31,360][2761575] Updated weights for policy 0, policy_version 2050 (0.0007) [2025-03-28 18:13:32,911][2761575] Updated weights for policy 0, policy_version 2060 (0.0006) [2025-03-28 18:13:34,347][2761575] Updated weights for policy 0, policy_version 2070 (0.0006) [2025-03-28 18:13:35,749][2713170] Fps is (10 sec: 25395.2, 60 sec: 25941.4, 300 sec: 26242.2). Total num frames: 8515584. Throughput: 0: 6413.5. Samples: 2115796. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-03-28 18:13:35,750][2713170] Avg episode reward: [(0, '19.077')] [2025-03-28 18:13:35,857][2761575] Updated weights for policy 0, policy_version 2080 (0.0006) [2025-03-28 18:13:37,271][2761575] Updated weights for policy 0, policy_version 2090 (0.0006) [2025-03-28 18:13:38,754][2761575] Updated weights for policy 0, policy_version 2100 (0.0006) [2025-03-28 18:13:40,189][2761575] Updated weights for policy 0, policy_version 2110 (0.0006) [2025-03-28 18:13:40,749][2713170] Fps is (10 sec: 26624.3, 60 sec: 25941.4, 300 sec: 26256.1). Total num frames: 8654848. Throughput: 0: 6476.9. Samples: 2157988. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:13:40,750][2713170] Avg episode reward: [(0, '21.779')] [2025-03-28 18:13:41,644][2761575] Updated weights for policy 0, policy_version 2120 (0.0006) [2025-03-28 18:13:43,144][2761575] Updated weights for policy 0, policy_version 2130 (0.0006) [2025-03-28 18:13:44,616][2761575] Updated weights for policy 0, policy_version 2140 (0.0006) [2025-03-28 18:13:45,749][2713170] Fps is (10 sec: 27852.6, 60 sec: 26146.2, 300 sec: 26283.8). Total num frames: 8794112. Throughput: 0: 6969.5. Samples: 2199726. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:13:45,750][2713170] Avg episode reward: [(0, '21.749')] [2025-03-28 18:13:46,107][2761575] Updated weights for policy 0, policy_version 2150 (0.0006) [2025-03-28 18:13:47,532][2761575] Updated weights for policy 0, policy_version 2160 (0.0006) [2025-03-28 18:13:48,990][2761575] Updated weights for policy 0, policy_version 2170 (0.0006) [2025-03-28 18:13:50,455][2761575] Updated weights for policy 0, policy_version 2180 (0.0006) [2025-03-28 18:13:50,749][2713170] Fps is (10 sec: 28262.5, 60 sec: 26351.0, 300 sec: 26311.6). Total num frames: 8937472. Throughput: 0: 6750.1. Samples: 2221092. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:13:50,750][2713170] Avg episode reward: [(0, '24.903')] [2025-03-28 18:13:51,874][2761575] Updated weights for policy 0, policy_version 2190 (0.0007) [2025-03-28 18:13:53,272][2761575] Updated weights for policy 0, policy_version 2200 (0.0007) [2025-03-28 18:13:54,739][2761575] Updated weights for policy 0, policy_version 2210 (0.0006) [2025-03-28 18:13:55,749][2713170] Fps is (10 sec: 28672.0, 60 sec: 26555.8, 300 sec: 26353.2). Total num frames: 9080832. Throughput: 0: 6825.1. Samples: 2263756. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:13:55,750][2713170] Avg episode reward: [(0, '23.567')] [2025-03-28 18:13:55,754][2761553] Saving /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000002217_9080832.pth... [2025-03-28 18:13:55,849][2761553] Removing /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth [2025-03-28 18:13:56,126][2761575] Updated weights for policy 0, policy_version 2220 (0.0006) [2025-03-28 18:13:57,663][2761575] Updated weights for policy 0, policy_version 2230 (0.0006) [2025-03-28 18:13:59,141][2761575] Updated weights for policy 0, policy_version 2240 (0.0006) [2025-03-28 18:14:00,749][2713170] Fps is (10 sec: 26213.4, 60 sec: 26692.1, 300 sec: 26297.7). Total num frames: 9199616. Throughput: 0: 6369.3. Samples: 2284744. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:14:00,751][2713170] Avg episode reward: [(0, '24.604')] [2025-03-28 18:14:01,763][2761575] Updated weights for policy 0, policy_version 2250 (0.0006) [2025-03-28 18:14:03,310][2761575] Updated weights for policy 0, policy_version 2260 (0.0006) [2025-03-28 18:14:04,749][2761575] Updated weights for policy 0, policy_version 2270 (0.0006) [2025-03-28 18:14:05,749][2713170] Fps is (10 sec: 24166.2, 60 sec: 26760.5, 300 sec: 26311.6). Total num frames: 9322496. Throughput: 0: 6669.6. Samples: 2317916. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) [2025-03-28 18:14:05,750][2713170] Avg episode reward: [(0, '24.252')] [2025-03-28 18:14:06,297][2761575] Updated weights for policy 0, policy_version 2280 (0.0006) [2025-03-28 18:14:07,747][2761575] Updated weights for policy 0, policy_version 2290 (0.0006) [2025-03-28 18:14:09,184][2761575] Updated weights for policy 0, policy_version 2300 (0.0006) [2025-03-28 18:14:10,703][2761575] Updated weights for policy 0, policy_version 2310 (0.0006) [2025-03-28 18:14:10,749][2713170] Fps is (10 sec: 26215.0, 60 sec: 26897.1, 300 sec: 26381.0). Total num frames: 9461760. Throughput: 0: 6696.1. Samples: 2359772. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:14:10,750][2713170] Avg episode reward: [(0, '24.326')] [2025-03-28 18:14:12,216][2761575] Updated weights for policy 0, policy_version 2320 (0.0006) [2025-03-28 18:14:13,745][2761575] Updated weights for policy 0, policy_version 2330 (0.0006) [2025-03-28 18:14:15,209][2761575] Updated weights for policy 0, policy_version 2340 (0.0006) [2025-03-28 18:14:15,749][2713170] Fps is (10 sec: 27443.2, 60 sec: 26828.8, 300 sec: 26381.0). Total num frames: 9596928. Throughput: 0: 7091.2. Samples: 2400370. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:14:15,750][2713170] Avg episode reward: [(0, '20.018')] [2025-03-28 18:14:16,719][2761575] Updated weights for policy 0, policy_version 2350 (0.0006) [2025-03-28 18:14:18,155][2761575] Updated weights for policy 0, policy_version 2360 (0.0006) [2025-03-28 18:14:19,710][2761575] Updated weights for policy 0, policy_version 2370 (0.0006) [2025-03-28 18:14:20,749][2713170] Fps is (10 sec: 27443.5, 60 sec: 26965.4, 300 sec: 26367.1). Total num frames: 9736192. Throughput: 0: 6782.9. Samples: 2421028. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:14:20,750][2713170] Avg episode reward: [(0, '23.661')] [2025-03-28 18:14:21,185][2761575] Updated weights for policy 0, policy_version 2380 (0.0006) [2025-03-28 18:14:22,759][2761575] Updated weights for policy 0, policy_version 2390 (0.0006) [2025-03-28 18:14:24,221][2761575] Updated weights for policy 0, policy_version 2400 (0.0006) [2025-03-28 18:14:25,551][2761575] Updated weights for policy 0, policy_version 2410 (0.0006) [2025-03-28 18:14:25,749][2713170] Fps is (10 sec: 27853.1, 60 sec: 26897.0, 300 sec: 26422.7). Total num frames: 9875456. Throughput: 0: 6755.1. Samples: 2461968. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:14:25,750][2713170] Avg episode reward: [(0, '24.863')] [2025-03-28 18:14:26,917][2761575] Updated weights for policy 0, policy_version 2420 (0.0006) [2025-03-28 18:14:28,389][2761575] Updated weights for policy 0, policy_version 2430 (0.0006) [2025-03-28 18:14:29,879][2761575] Updated weights for policy 0, policy_version 2440 (0.0006) [2025-03-28 18:14:30,749][2713170] Fps is (10 sec: 26214.2, 60 sec: 26828.8, 300 sec: 26394.9). Total num frames: 9998336. Throughput: 0: 6324.1. Samples: 2484310. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:14:30,750][2713170] Avg episode reward: [(0, '25.162')] [2025-03-28 18:14:32,585][2761575] Updated weights for policy 0, policy_version 2450 (0.0008) [2025-03-28 18:14:34,124][2761575] Updated weights for policy 0, policy_version 2460 (0.0006) [2025-03-28 18:14:35,621][2761575] Updated weights for policy 0, policy_version 2470 (0.0006) [2025-03-28 18:14:35,749][2713170] Fps is (10 sec: 24166.1, 60 sec: 26692.2, 300 sec: 26339.4). Total num frames: 10117120. Throughput: 0: 6577.6. Samples: 2517086. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) [2025-03-28 18:14:35,750][2713170] Avg episode reward: [(0, '24.141')] [2025-03-28 18:14:37,120][2761575] Updated weights for policy 0, policy_version 2480 (0.0007) [2025-03-28 18:14:38,635][2761575] Updated weights for policy 0, policy_version 2490 (0.0006) [2025-03-28 18:14:40,105][2761575] Updated weights for policy 0, policy_version 2500 (0.0006) [2025-03-28 18:14:40,749][2713170] Fps is (10 sec: 25804.7, 60 sec: 26692.2, 300 sec: 26367.1). Total num frames: 10256384. Throughput: 0: 6539.5. Samples: 2558034. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:14:40,750][2713170] Avg episode reward: [(0, '25.604')] [2025-03-28 18:14:40,752][2761553] Saving new best policy, reward=25.604! [2025-03-28 18:14:41,632][2761575] Updated weights for policy 0, policy_version 2510 (0.0006) [2025-03-28 18:14:42,985][2761575] Updated weights for policy 0, policy_version 2520 (0.0006) [2025-03-28 18:14:44,513][2761575] Updated weights for policy 0, policy_version 2530 (0.0006) [2025-03-28 18:14:45,749][2713170] Fps is (10 sec: 27443.2, 60 sec: 26624.0, 300 sec: 26381.0). Total num frames: 10391552. Throughput: 0: 6995.8. Samples: 2599552. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-03-28 18:14:45,750][2713170] Avg episode reward: [(0, '24.231')] [2025-03-28 18:14:46,063][2761575] Updated weights for policy 0, policy_version 2540 (0.0006) [2025-03-28 18:14:47,590][2761575] Updated weights for policy 0, policy_version 2550 (0.0006) [2025-03-28 18:14:49,112][2761575] Updated weights for policy 0, policy_version 2560 (0.0006) [2025-03-28 18:14:50,602][2761575] Updated weights for policy 0, policy_version 2570 (0.0006) [2025-03-28 18:14:50,749][2713170] Fps is (10 sec: 27033.6, 60 sec: 26487.4, 300 sec: 26381.0). Total num frames: 10526720. Throughput: 0: 6705.0. Samples: 2619642. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:14:50,750][2713170] Avg episode reward: [(0, '25.343')] [2025-03-28 18:14:52,104][2761575] Updated weights for policy 0, policy_version 2580 (0.0006) [2025-03-28 18:14:53,578][2761575] Updated weights for policy 0, policy_version 2590 (0.0006) [2025-03-28 18:14:55,157][2761575] Updated weights for policy 0, policy_version 2600 (0.0006) [2025-03-28 18:14:55,749][2713170] Fps is (10 sec: 27443.2, 60 sec: 26419.2, 300 sec: 26450.4). Total num frames: 10665984. Throughput: 0: 6680.1. Samples: 2660378. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:14:55,750][2713170] Avg episode reward: [(0, '25.284')] [2025-03-28 18:14:56,629][2761575] Updated weights for policy 0, policy_version 2610 (0.0006) [2025-03-28 18:14:58,201][2761575] Updated weights for policy 0, policy_version 2620 (0.0006) [2025-03-28 18:14:59,737][2761575] Updated weights for policy 0, policy_version 2630 (0.0006) [2025-03-28 18:15:00,749][2713170] Fps is (10 sec: 25395.1, 60 sec: 26351.0, 300 sec: 26367.1). Total num frames: 10780672. Throughput: 0: 6222.7. Samples: 2680392. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-03-28 18:15:00,751][2713170] Avg episode reward: [(0, '22.368')] [2025-03-28 18:15:02,451][2761575] Updated weights for policy 0, policy_version 2640 (0.0008) [2025-03-28 18:15:03,935][2761575] Updated weights for policy 0, policy_version 2650 (0.0006) [2025-03-28 18:15:05,302][2761575] Updated weights for policy 0, policy_version 2660 (0.0006) [2025-03-28 18:15:05,749][2713170] Fps is (10 sec: 23756.8, 60 sec: 26350.9, 300 sec: 26381.0). Total num frames: 10903552. Throughput: 0: 6486.5. Samples: 2712922. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:15:05,750][2713170] Avg episode reward: [(0, '26.731')] [2025-03-28 18:15:05,773][2761553] Saving new best policy, reward=26.731! [2025-03-28 18:15:06,803][2761575] Updated weights for policy 0, policy_version 2670 (0.0006) [2025-03-28 18:15:08,239][2761575] Updated weights for policy 0, policy_version 2680 (0.0006) [2025-03-28 18:15:09,657][2761575] Updated weights for policy 0, policy_version 2690 (0.0006) [2025-03-28 18:15:10,749][2713170] Fps is (10 sec: 26624.0, 60 sec: 26419.2, 300 sec: 26492.1). Total num frames: 11046912. Throughput: 0: 6531.6. Samples: 2755890. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:15:10,751][2713170] Avg episode reward: [(0, '27.009')] [2025-03-28 18:15:10,767][2761553] Saving new best policy, reward=27.009! [2025-03-28 18:15:11,080][2761575] Updated weights for policy 0, policy_version 2700 (0.0006) [2025-03-28 18:15:12,574][2761575] Updated weights for policy 0, policy_version 2710 (0.0006) [2025-03-28 18:15:14,037][2761575] Updated weights for policy 0, policy_version 2720 (0.0006) [2025-03-28 18:15:15,360][2761575] Updated weights for policy 0, policy_version 2730 (0.0006) [2025-03-28 18:15:15,749][2713170] Fps is (10 sec: 28672.0, 60 sec: 26555.7, 300 sec: 26519.9). Total num frames: 11190272. Throughput: 0: 6990.3. Samples: 2798872. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:15:15,750][2713170] Avg episode reward: [(0, '25.415')] [2025-03-28 18:15:16,862][2761575] Updated weights for policy 0, policy_version 2740 (0.0006) [2025-03-28 18:15:18,246][2761575] Updated weights for policy 0, policy_version 2750 (0.0006) [2025-03-28 18:15:19,774][2761575] Updated weights for policy 0, policy_version 2760 (0.0006) [2025-03-28 18:15:20,749][2713170] Fps is (10 sec: 28262.5, 60 sec: 26555.7, 300 sec: 26547.6). Total num frames: 11329536. Throughput: 0: 6735.9. Samples: 2820200. Policy #0 lag: (min: 0.0, avg: 0.9, max: 2.0) [2025-03-28 18:15:20,750][2713170] Avg episode reward: [(0, '24.524')] [2025-03-28 18:15:21,229][2761575] Updated weights for policy 0, policy_version 2770 (0.0006) [2025-03-28 18:15:22,769][2761575] Updated weights for policy 0, policy_version 2780 (0.0006) [2025-03-28 18:15:24,212][2761575] Updated weights for policy 0, policy_version 2790 (0.0006) [2025-03-28 18:15:25,687][2761575] Updated weights for policy 0, policy_version 2800 (0.0006) [2025-03-28 18:15:25,749][2713170] Fps is (10 sec: 27852.8, 60 sec: 26555.7, 300 sec: 26603.2). Total num frames: 11468800. Throughput: 0: 6734.6. Samples: 2861092. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:15:25,750][2713170] Avg episode reward: [(0, '23.794')] [2025-03-28 18:15:27,128][2761575] Updated weights for policy 0, policy_version 2810 (0.0006) [2025-03-28 18:15:28,658][2761575] Updated weights for policy 0, policy_version 2820 (0.0007) [2025-03-28 18:15:30,115][2761575] Updated weights for policy 0, policy_version 2830 (0.0006) [2025-03-28 18:15:30,749][2713170] Fps is (10 sec: 27852.8, 60 sec: 26828.8, 300 sec: 26630.9). Total num frames: 11608064. Throughput: 0: 6735.0. Samples: 2902628. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:15:30,750][2713170] Avg episode reward: [(0, '26.945')] [2025-03-28 18:15:31,660][2761575] Updated weights for policy 0, policy_version 2840 (0.0006) [2025-03-28 18:15:33,175][2761575] Updated weights for policy 0, policy_version 2850 (0.0006) [2025-03-28 18:15:34,678][2761575] Updated weights for policy 0, policy_version 2860 (0.0006) [2025-03-28 18:15:35,749][2713170] Fps is (10 sec: 27033.6, 60 sec: 27033.6, 300 sec: 26617.0). Total num frames: 11739136. Throughput: 0: 6735.7. Samples: 2922748. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:15:35,750][2713170] Avg episode reward: [(0, '26.052')] [2025-03-28 18:15:36,210][2761575] Updated weights for policy 0, policy_version 2870 (0.0006) [2025-03-28 18:15:37,696][2761575] Updated weights for policy 0, policy_version 2880 (0.0006) [2025-03-28 18:15:39,233][2761575] Updated weights for policy 0, policy_version 2890 (0.0006) [2025-03-28 18:15:40,721][2761575] Updated weights for policy 0, policy_version 2900 (0.0006) [2025-03-28 18:15:40,749][2713170] Fps is (10 sec: 27033.6, 60 sec: 27033.6, 300 sec: 26644.8). Total num frames: 11878400. Throughput: 0: 6736.8. Samples: 2963532. Policy #0 lag: (min: 0.0, avg: 0.9, max: 2.0) [2025-03-28 18:15:40,750][2713170] Avg episode reward: [(0, '24.499')] [2025-03-28 18:15:42,186][2761575] Updated weights for policy 0, policy_version 2910 (0.0006) [2025-03-28 18:15:43,692][2761575] Updated weights for policy 0, policy_version 2920 (0.0006) [2025-03-28 18:15:45,231][2761575] Updated weights for policy 0, policy_version 2930 (0.0006) [2025-03-28 18:15:45,749][2713170] Fps is (10 sec: 27443.0, 60 sec: 27033.6, 300 sec: 26644.8). Total num frames: 12013568. Throughput: 0: 7199.3. Samples: 3004362. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:15:45,751][2713170] Avg episode reward: [(0, '26.088')] [2025-03-28 18:15:46,752][2761575] Updated weights for policy 0, policy_version 2940 (0.0006) [2025-03-28 18:15:48,292][2761575] Updated weights for policy 0, policy_version 2950 (0.0006) [2025-03-28 18:15:49,786][2761575] Updated weights for policy 0, policy_version 2960 (0.0006) [2025-03-28 18:15:50,749][2713170] Fps is (10 sec: 27033.6, 60 sec: 27033.6, 300 sec: 26644.8). Total num frames: 12148736. Throughput: 0: 6925.6. Samples: 3024574. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:15:50,751][2713170] Avg episode reward: [(0, '25.095')] [2025-03-28 18:15:51,213][2761575] Updated weights for policy 0, policy_version 2970 (0.0006) [2025-03-28 18:15:52,807][2761575] Updated weights for policy 0, policy_version 2980 (0.0006) [2025-03-28 18:15:54,287][2761575] Updated weights for policy 0, policy_version 2990 (0.0006) [2025-03-28 18:15:55,749][2713170] Fps is (10 sec: 27033.7, 60 sec: 26965.3, 300 sec: 26700.4). Total num frames: 12283904. Throughput: 0: 6880.2. Samples: 3065498. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:15:55,750][2713170] Avg episode reward: [(0, '25.314')] [2025-03-28 18:15:55,757][2761553] Saving /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000002999_12283904.pth... [2025-03-28 18:15:55,866][2761553] Removing /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000001439_5894144.pth [2025-03-28 18:15:55,886][2761575] Updated weights for policy 0, policy_version 3000 (0.0006) [2025-03-28 18:15:57,349][2761575] Updated weights for policy 0, policy_version 3010 (0.0006) [2025-03-28 18:15:58,817][2761575] Updated weights for policy 0, policy_version 3020 (0.0006) [2025-03-28 18:16:00,313][2761575] Updated weights for policy 0, policy_version 3030 (0.0006) [2025-03-28 18:16:00,749][2713170] Fps is (10 sec: 27443.2, 60 sec: 27374.9, 300 sec: 26728.1). Total num frames: 12423168. Throughput: 0: 6836.8. Samples: 3106526. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:16:00,750][2713170] Avg episode reward: [(0, '25.221')] [2025-03-28 18:16:01,761][2761575] Updated weights for policy 0, policy_version 3040 (0.0006) [2025-03-28 18:16:03,274][2761575] Updated weights for policy 0, policy_version 3050 (0.0006) [2025-03-28 18:16:04,795][2761575] Updated weights for policy 0, policy_version 3060 (0.0006) [2025-03-28 18:16:05,749][2713170] Fps is (10 sec: 27443.4, 60 sec: 27579.7, 300 sec: 26728.1). Total num frames: 12558336. Throughput: 0: 6814.5. Samples: 3126852. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:16:05,750][2713170] Avg episode reward: [(0, '27.344')] [2025-03-28 18:16:05,755][2761553] Saving new best policy, reward=27.344! [2025-03-28 18:16:06,259][2761575] Updated weights for policy 0, policy_version 3070 (0.0007) [2025-03-28 18:16:07,628][2761575] Updated weights for policy 0, policy_version 3080 (0.0006) [2025-03-28 18:16:09,155][2761575] Updated weights for policy 0, policy_version 3090 (0.0006) [2025-03-28 18:16:10,636][2761575] Updated weights for policy 0, policy_version 3100 (0.0006) [2025-03-28 18:16:10,749][2713170] Fps is (10 sec: 27443.3, 60 sec: 27511.5, 300 sec: 26714.2). Total num frames: 12697600. Throughput: 0: 6843.6. Samples: 3169056. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:16:10,750][2713170] Avg episode reward: [(0, '25.365')] [2025-03-28 18:16:12,131][2761575] Updated weights for policy 0, policy_version 3110 (0.0006) [2025-03-28 18:16:13,575][2761575] Updated weights for policy 0, policy_version 3120 (0.0006) [2025-03-28 18:16:14,983][2761575] Updated weights for policy 0, policy_version 3130 (0.0006) [2025-03-28 18:16:15,749][2713170] Fps is (10 sec: 28262.1, 60 sec: 27511.4, 300 sec: 26742.0). Total num frames: 12840960. Throughput: 0: 6868.2. Samples: 3211696. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:16:15,752][2713170] Avg episode reward: [(0, '23.905')] [2025-03-28 18:16:16,381][2761575] Updated weights for policy 0, policy_version 3140 (0.0006) [2025-03-28 18:16:17,845][2761575] Updated weights for policy 0, policy_version 3150 (0.0006) [2025-03-28 18:16:19,329][2761575] Updated weights for policy 0, policy_version 3160 (0.0006) [2025-03-28 18:16:20,749][2713170] Fps is (10 sec: 28262.4, 60 sec: 27511.5, 300 sec: 26755.9). Total num frames: 12980224. Throughput: 0: 6885.0. Samples: 3232574. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:16:20,750][2713170] Avg episode reward: [(0, '27.121')] [2025-03-28 18:16:20,813][2761575] Updated weights for policy 0, policy_version 3170 (0.0006) [2025-03-28 18:16:22,327][2761575] Updated weights for policy 0, policy_version 3180 (0.0006) [2025-03-28 18:16:23,849][2761575] Updated weights for policy 0, policy_version 3190 (0.0006) [2025-03-28 18:16:25,314][2761575] Updated weights for policy 0, policy_version 3200 (0.0006) [2025-03-28 18:16:25,749][2713170] Fps is (10 sec: 27853.1, 60 sec: 27511.5, 300 sec: 26825.3). Total num frames: 13119488. Throughput: 0: 6886.4. Samples: 3273418. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:16:25,750][2713170] Avg episode reward: [(0, '24.838')] [2025-03-28 18:16:26,604][2761575] Updated weights for policy 0, policy_version 3210 (0.0006) [2025-03-28 18:16:28,120][2761575] Updated weights for policy 0, policy_version 3220 (0.0006) [2025-03-28 18:16:29,514][2761575] Updated weights for policy 0, policy_version 3230 (0.0006) [2025-03-28 18:16:30,749][2713170] Fps is (10 sec: 28262.4, 60 sec: 27579.7, 300 sec: 26867.0). Total num frames: 13262848. Throughput: 0: 6941.4. Samples: 3316724. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:16:30,750][2713170] Avg episode reward: [(0, '23.754')] [2025-03-28 18:16:30,980][2761575] Updated weights for policy 0, policy_version 3240 (0.0007) [2025-03-28 18:16:32,367][2761575] Updated weights for policy 0, policy_version 3250 (0.0006) [2025-03-28 18:16:33,822][2761575] Updated weights for policy 0, policy_version 3260 (0.0006) [2025-03-28 18:16:35,145][2761575] Updated weights for policy 0, policy_version 3270 (0.0006) [2025-03-28 18:16:35,749][2713170] Fps is (10 sec: 29081.8, 60 sec: 27852.8, 300 sec: 26908.6). Total num frames: 13410304. Throughput: 0: 6971.9. Samples: 3338308. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) [2025-03-28 18:16:35,750][2713170] Avg episode reward: [(0, '29.695')] [2025-03-28 18:16:35,753][2761553] Saving new best policy, reward=29.695! [2025-03-28 18:16:36,567][2761575] Updated weights for policy 0, policy_version 3280 (0.0006) [2025-03-28 18:16:38,027][2761575] Updated weights for policy 0, policy_version 3290 (0.0006) [2025-03-28 18:16:39,514][2761575] Updated weights for policy 0, policy_version 3300 (0.0006) [2025-03-28 18:16:40,749][2713170] Fps is (10 sec: 28672.2, 60 sec: 27852.8, 300 sec: 26908.6). Total num frames: 13549568. Throughput: 0: 7021.8. Samples: 3381476. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:16:40,750][2713170] Avg episode reward: [(0, '25.651')] [2025-03-28 18:16:40,957][2761575] Updated weights for policy 0, policy_version 3310 (0.0007) [2025-03-28 18:16:42,424][2761575] Updated weights for policy 0, policy_version 3320 (0.0006) [2025-03-28 18:16:43,873][2761575] Updated weights for policy 0, policy_version 3330 (0.0006) [2025-03-28 18:16:45,328][2761575] Updated weights for policy 0, policy_version 3340 (0.0007) [2025-03-28 18:16:45,749][2713170] Fps is (10 sec: 27852.9, 60 sec: 27921.2, 300 sec: 26908.6). Total num frames: 13688832. Throughput: 0: 7049.4. Samples: 3423748. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:16:45,750][2713170] Avg episode reward: [(0, '29.948')] [2025-03-28 18:16:45,753][2761553] Saving new best policy, reward=29.948! [2025-03-28 18:16:46,803][2761575] Updated weights for policy 0, policy_version 3350 (0.0006) [2025-03-28 18:16:48,250][2761575] Updated weights for policy 0, policy_version 3360 (0.0007) [2025-03-28 18:16:49,685][2761575] Updated weights for policy 0, policy_version 3370 (0.0006) [2025-03-28 18:16:50,749][2713170] Fps is (10 sec: 28262.2, 60 sec: 28057.6, 300 sec: 26908.6). Total num frames: 13832192. Throughput: 0: 7068.4. Samples: 3444930. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:16:50,750][2713170] Avg episode reward: [(0, '25.852')] [2025-03-28 18:16:51,036][2761575] Updated weights for policy 0, policy_version 3380 (0.0006) [2025-03-28 18:16:52,457][2761575] Updated weights for policy 0, policy_version 3390 (0.0007) [2025-03-28 18:16:53,965][2761575] Updated weights for policy 0, policy_version 3400 (0.0007) [2025-03-28 18:16:55,414][2761575] Updated weights for policy 0, policy_version 3410 (0.0007) [2025-03-28 18:16:55,749][2713170] Fps is (10 sec: 28671.6, 60 sec: 28194.2, 300 sec: 26991.9). Total num frames: 13975552. Throughput: 0: 7082.7. Samples: 3487778. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:16:55,750][2713170] Avg episode reward: [(0, '24.025')] [2025-03-28 18:16:56,833][2761575] Updated weights for policy 0, policy_version 3420 (0.0007) [2025-03-28 18:16:58,327][2761575] Updated weights for policy 0, policy_version 3430 (0.0006) [2025-03-28 18:16:59,818][2761575] Updated weights for policy 0, policy_version 3440 (0.0006) [2025-03-28 18:17:00,749][2713170] Fps is (10 sec: 28262.5, 60 sec: 28194.2, 300 sec: 27019.7). Total num frames: 14114816. Throughput: 0: 7063.4. Samples: 3529550. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:17:00,750][2713170] Avg episode reward: [(0, '24.609')] [2025-03-28 18:17:01,296][2761575] Updated weights for policy 0, policy_version 3450 (0.0006) [2025-03-28 18:17:02,727][2761575] Updated weights for policy 0, policy_version 3460 (0.0006) [2025-03-28 18:17:04,201][2761575] Updated weights for policy 0, policy_version 3470 (0.0006) [2025-03-28 18:17:05,645][2761575] Updated weights for policy 0, policy_version 3480 (0.0006) [2025-03-28 18:17:05,749][2713170] Fps is (10 sec: 27852.9, 60 sec: 28262.4, 300 sec: 27019.7). Total num frames: 14254080. Throughput: 0: 7069.8. Samples: 3550716. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:17:05,750][2713170] Avg episode reward: [(0, '30.324')] [2025-03-28 18:17:05,757][2761553] Saving new best policy, reward=30.324! [2025-03-28 18:17:07,144][2761575] Updated weights for policy 0, policy_version 3490 (0.0007) [2025-03-28 18:17:08,662][2761575] Updated weights for policy 0, policy_version 3500 (0.0007) [2025-03-28 18:17:10,149][2761575] Updated weights for policy 0, policy_version 3510 (0.0007) [2025-03-28 18:17:10,749][2713170] Fps is (10 sec: 26214.6, 60 sec: 27989.4, 300 sec: 26964.2). Total num frames: 14376960. Throughput: 0: 7082.8. Samples: 3592142. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:17:10,749][2713170] Avg episode reward: [(0, '23.140')] [2025-03-28 18:17:12,128][2761575] Updated weights for policy 0, policy_version 3520 (0.0006) [2025-03-28 18:17:13,651][2761575] Updated weights for policy 0, policy_version 3530 (0.0006) [2025-03-28 18:17:15,108][2761575] Updated weights for policy 0, policy_version 3540 (0.0006) [2025-03-28 18:17:15,749][2713170] Fps is (10 sec: 26214.4, 60 sec: 27921.1, 300 sec: 26964.2). Total num frames: 14516224. Throughput: 0: 6958.9. Samples: 3629876. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:17:15,751][2713170] Avg episode reward: [(0, '27.505')] [2025-03-28 18:17:16,608][2761575] Updated weights for policy 0, policy_version 3550 (0.0007) [2025-03-28 18:17:18,048][2761575] Updated weights for policy 0, policy_version 3560 (0.0007) [2025-03-28 18:17:19,369][2761575] Updated weights for policy 0, policy_version 3570 (0.0006) [2025-03-28 18:17:20,749][2713170] Fps is (10 sec: 28262.1, 60 sec: 27989.3, 300 sec: 26950.3). Total num frames: 14659584. Throughput: 0: 6960.4. Samples: 3651528. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:17:20,751][2713170] Avg episode reward: [(0, '29.310')] [2025-03-28 18:17:20,791][2761575] Updated weights for policy 0, policy_version 3580 (0.0006) [2025-03-28 18:17:22,289][2761575] Updated weights for policy 0, policy_version 3590 (0.0006) [2025-03-28 18:17:23,740][2761575] Updated weights for policy 0, policy_version 3600 (0.0007) [2025-03-28 18:17:25,167][2761575] Updated weights for policy 0, policy_version 3610 (0.0006) [2025-03-28 18:17:25,749][2713170] Fps is (10 sec: 28672.0, 60 sec: 28057.6, 300 sec: 27019.7). Total num frames: 14802944. Throughput: 0: 6950.7. Samples: 3694256. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:17:25,750][2713170] Avg episode reward: [(0, '28.705')] [2025-03-28 18:17:26,530][2761575] Updated weights for policy 0, policy_version 3620 (0.0006) [2025-03-28 18:17:27,910][2761575] Updated weights for policy 0, policy_version 3630 (0.0006) [2025-03-28 18:17:29,310][2761575] Updated weights for policy 0, policy_version 3640 (0.0006) [2025-03-28 18:17:30,711][2761575] Updated weights for policy 0, policy_version 3650 (0.0006) [2025-03-28 18:17:30,749][2713170] Fps is (10 sec: 29081.6, 60 sec: 28125.9, 300 sec: 27089.1). Total num frames: 14950400. Throughput: 0: 6996.1. Samples: 3738574. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:17:30,750][2713170] Avg episode reward: [(0, '27.222')] [2025-03-28 18:17:32,132][2761575] Updated weights for policy 0, policy_version 3660 (0.0006) [2025-03-28 18:17:33,589][2761575] Updated weights for policy 0, policy_version 3670 (0.0006) [2025-03-28 18:17:35,105][2761575] Updated weights for policy 0, policy_version 3680 (0.0007) [2025-03-28 18:17:35,749][2713170] Fps is (10 sec: 28671.9, 60 sec: 27989.3, 300 sec: 27089.1). Total num frames: 15089664. Throughput: 0: 6997.4. Samples: 3759812. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:17:35,751][2713170] Avg episode reward: [(0, '28.245')] [2025-03-28 18:17:36,637][2761575] Updated weights for policy 0, policy_version 3690 (0.0006) [2025-03-28 18:17:38,191][2761575] Updated weights for policy 0, policy_version 3700 (0.0007) [2025-03-28 18:17:39,703][2761575] Updated weights for policy 0, policy_version 3710 (0.0007) [2025-03-28 18:17:40,749][2713170] Fps is (10 sec: 27443.2, 60 sec: 27921.0, 300 sec: 27116.9). Total num frames: 15224832. Throughput: 0: 6938.4. Samples: 3800006. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:17:40,750][2713170] Avg episode reward: [(0, '30.611')] [2025-03-28 18:17:40,752][2761553] Saving new best policy, reward=30.611! [2025-03-28 18:17:41,157][2761575] Updated weights for policy 0, policy_version 3720 (0.0006) [2025-03-28 18:17:42,693][2761575] Updated weights for policy 0, policy_version 3730 (0.0006) [2025-03-28 18:17:44,161][2761575] Updated weights for policy 0, policy_version 3740 (0.0006) [2025-03-28 18:17:45,710][2761575] Updated weights for policy 0, policy_version 3750 (0.0006) [2025-03-28 18:17:45,749][2713170] Fps is (10 sec: 27033.6, 60 sec: 27852.7, 300 sec: 27130.8). Total num frames: 15360000. Throughput: 0: 6917.6. Samples: 3840844. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:17:45,750][2713170] Avg episode reward: [(0, '27.316')] [2025-03-28 18:17:47,229][2761575] Updated weights for policy 0, policy_version 3760 (0.0006) [2025-03-28 18:17:48,757][2761575] Updated weights for policy 0, policy_version 3770 (0.0006) [2025-03-28 18:17:50,309][2761575] Updated weights for policy 0, policy_version 3780 (0.0006) [2025-03-28 18:17:50,749][2713170] Fps is (10 sec: 27033.6, 60 sec: 27716.3, 300 sec: 27144.7). Total num frames: 15495168. Throughput: 0: 6894.2. Samples: 3860954. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:17:50,750][2713170] Avg episode reward: [(0, '24.025')] [2025-03-28 18:17:51,869][2761575] Updated weights for policy 0, policy_version 3790 (0.0006) [2025-03-28 18:17:53,413][2761575] Updated weights for policy 0, policy_version 3800 (0.0006) [2025-03-28 18:17:54,896][2761575] Updated weights for policy 0, policy_version 3810 (0.0006) [2025-03-28 18:17:55,749][2713170] Fps is (10 sec: 26623.8, 60 sec: 27511.4, 300 sec: 27214.1). Total num frames: 15626240. Throughput: 0: 6866.6. Samples: 3901140. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:17:55,750][2713170] Avg episode reward: [(0, '27.859')] [2025-03-28 18:17:55,757][2761553] Saving /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000003815_15626240.pth... [2025-03-28 18:17:55,861][2761553] Removing /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000002217_9080832.pth [2025-03-28 18:17:56,538][2761575] Updated weights for policy 0, policy_version 3820 (0.0007) [2025-03-28 18:17:58,075][2761575] Updated weights for policy 0, policy_version 3830 (0.0007) [2025-03-28 18:17:59,667][2761575] Updated weights for policy 0, policy_version 3840 (0.0007) [2025-03-28 18:18:00,749][2713170] Fps is (10 sec: 25804.8, 60 sec: 27306.6, 300 sec: 27241.9). Total num frames: 15753216. Throughput: 0: 6866.9. Samples: 3938886. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:18:00,750][2713170] Avg episode reward: [(0, '28.161')] [2025-03-28 18:18:01,340][2761575] Updated weights for policy 0, policy_version 3850 (0.0007) [2025-03-28 18:18:02,968][2761575] Updated weights for policy 0, policy_version 3860 (0.0006) [2025-03-28 18:18:04,651][2761575] Updated weights for policy 0, policy_version 3870 (0.0007) [2025-03-28 18:18:05,749][2713170] Fps is (10 sec: 24985.7, 60 sec: 27033.6, 300 sec: 27214.1). Total num frames: 15876096. Throughput: 0: 6808.7. Samples: 3957918. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:18:05,751][2713170] Avg episode reward: [(0, '26.752')] [2025-03-28 18:18:06,272][2761575] Updated weights for policy 0, policy_version 3880 (0.0007) [2025-03-28 18:18:07,847][2761575] Updated weights for policy 0, policy_version 3890 (0.0007) [2025-03-28 18:18:09,400][2761575] Updated weights for policy 0, policy_version 3900 (0.0007) [2025-03-28 18:18:10,749][2713170] Fps is (10 sec: 24166.6, 60 sec: 26965.3, 300 sec: 27144.7). Total num frames: 15994880. Throughput: 0: 6714.1. Samples: 3996392. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:18:10,750][2713170] Avg episode reward: [(0, '28.720')] [2025-03-28 18:18:11,964][2761575] Updated weights for policy 0, policy_version 3910 (0.0007) [2025-03-28 18:18:13,441][2761575] Updated weights for policy 0, policy_version 3920 (0.0007) [2025-03-28 18:18:14,850][2761575] Updated weights for policy 0, policy_version 3930 (0.0006) [2025-03-28 18:18:15,749][2713170] Fps is (10 sec: 24166.4, 60 sec: 26692.2, 300 sec: 27116.9). Total num frames: 16117760. Throughput: 0: 6498.7. Samples: 4031014. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:18:15,751][2713170] Avg episode reward: [(0, '26.314')] [2025-03-28 18:18:16,341][2761575] Updated weights for policy 0, policy_version 3940 (0.0006) [2025-03-28 18:18:17,708][2761575] Updated weights for policy 0, policy_version 3950 (0.0006) [2025-03-28 18:18:18,992][2761575] Updated weights for policy 0, policy_version 3960 (0.0006) [2025-03-28 18:18:20,331][2761575] Updated weights for policy 0, policy_version 3970 (0.0007) [2025-03-28 18:18:20,749][2713170] Fps is (10 sec: 27442.9, 60 sec: 26828.8, 300 sec: 27144.7). Total num frames: 16269312. Throughput: 0: 6533.2. Samples: 4053804. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:18:20,751][2713170] Avg episode reward: [(0, '25.039')] [2025-03-28 18:18:21,876][2761575] Updated weights for policy 0, policy_version 3980 (0.0006) [2025-03-28 18:18:23,401][2761575] Updated weights for policy 0, policy_version 3990 (0.0006) [2025-03-28 18:18:24,797][2761575] Updated weights for policy 0, policy_version 4000 (0.0006) [2025-03-28 18:18:25,749][2713170] Fps is (10 sec: 29081.7, 60 sec: 26760.5, 300 sec: 27186.3). Total num frames: 16408576. Throughput: 0: 6586.3. Samples: 4096388. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:18:25,751][2713170] Avg episode reward: [(0, '26.266')] [2025-03-28 18:18:26,277][2761575] Updated weights for policy 0, policy_version 4010 (0.0006) [2025-03-28 18:18:27,871][2761575] Updated weights for policy 0, policy_version 4020 (0.0006) [2025-03-28 18:18:29,417][2761575] Updated weights for policy 0, policy_version 4030 (0.0007) [2025-03-28 18:18:30,749][2713170] Fps is (10 sec: 26624.2, 60 sec: 26419.2, 300 sec: 27186.3). Total num frames: 16535552. Throughput: 0: 6526.5. Samples: 4134534. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:18:30,750][2713170] Avg episode reward: [(0, '25.414')] [2025-03-28 18:18:31,171][2761575] Updated weights for policy 0, policy_version 4040 (0.0008) [2025-03-28 18:18:32,693][2761575] Updated weights for policy 0, policy_version 4050 (0.0007) [2025-03-28 18:18:34,165][2761575] Updated weights for policy 0, policy_version 4060 (0.0006) [2025-03-28 18:18:35,669][2761575] Updated weights for policy 0, policy_version 4070 (0.0006) [2025-03-28 18:18:35,749][2713170] Fps is (10 sec: 26214.3, 60 sec: 26350.9, 300 sec: 27172.4). Total num frames: 16670720. Throughput: 0: 6537.7. Samples: 4155152. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:18:35,750][2713170] Avg episode reward: [(0, '26.396')] [2025-03-28 18:18:37,201][2761575] Updated weights for policy 0, policy_version 4080 (0.0006) [2025-03-28 18:18:38,743][2761575] Updated weights for policy 0, policy_version 4090 (0.0006) [2025-03-28 18:18:40,314][2761575] Updated weights for policy 0, policy_version 4100 (0.0006) [2025-03-28 18:18:40,749][2713170] Fps is (10 sec: 26623.8, 60 sec: 26282.7, 300 sec: 27144.7). Total num frames: 16801792. Throughput: 0: 6534.5. Samples: 4195190. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:18:40,750][2713170] Avg episode reward: [(0, '27.750')] [2025-03-28 18:18:41,874][2761575] Updated weights for policy 0, policy_version 4110 (0.0006) [2025-03-28 18:18:43,477][2761575] Updated weights for policy 0, policy_version 4120 (0.0006) [2025-03-28 18:18:45,123][2761575] Updated weights for policy 0, policy_version 4130 (0.0006) [2025-03-28 18:18:45,749][2713170] Fps is (10 sec: 25804.6, 60 sec: 26146.1, 300 sec: 27089.1). Total num frames: 16928768. Throughput: 0: 6553.1. Samples: 4233776. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:18:45,750][2713170] Avg episode reward: [(0, '26.994')] [2025-03-28 18:18:46,696][2761575] Updated weights for policy 0, policy_version 4140 (0.0007) [2025-03-28 18:18:48,385][2761575] Updated weights for policy 0, policy_version 4150 (0.0007) [2025-03-28 18:18:50,041][2761575] Updated weights for policy 0, policy_version 4160 (0.0006) [2025-03-28 18:18:50,749][2713170] Fps is (10 sec: 25395.2, 60 sec: 26009.6, 300 sec: 27033.6). Total num frames: 17055744. Throughput: 0: 6546.2. Samples: 4252496. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:18:50,750][2713170] Avg episode reward: [(0, '27.719')] [2025-03-28 18:18:51,663][2761575] Updated weights for policy 0, policy_version 4170 (0.0006) [2025-03-28 18:18:53,235][2761575] Updated weights for policy 0, policy_version 4180 (0.0006) [2025-03-28 18:18:54,839][2761575] Updated weights for policy 0, policy_version 4190 (0.0006) [2025-03-28 18:18:55,749][2713170] Fps is (10 sec: 25395.5, 60 sec: 25941.4, 300 sec: 27061.4). Total num frames: 17182720. Throughput: 0: 6537.7. Samples: 4290588. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:18:55,750][2713170] Avg episode reward: [(0, '28.506')] [2025-03-28 18:18:56,455][2761575] Updated weights for policy 0, policy_version 4200 (0.0006) [2025-03-28 18:18:57,990][2761575] Updated weights for policy 0, policy_version 4210 (0.0006) [2025-03-28 18:18:59,561][2761575] Updated weights for policy 0, policy_version 4220 (0.0006) [2025-03-28 18:19:00,749][2713170] Fps is (10 sec: 25395.4, 60 sec: 25941.3, 300 sec: 27075.3). Total num frames: 17309696. Throughput: 0: 6593.1. Samples: 4327702. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) [2025-03-28 18:19:00,750][2713170] Avg episode reward: [(0, '30.917')] [2025-03-28 18:19:00,752][2761553] Saving new best policy, reward=30.917! [2025-03-28 18:19:01,397][2761575] Updated weights for policy 0, policy_version 4230 (0.0009) [2025-03-28 18:19:02,841][2761575] Updated weights for policy 0, policy_version 4240 (0.0007) [2025-03-28 18:19:04,253][2761575] Updated weights for policy 0, policy_version 4250 (0.0006) [2025-03-28 18:19:05,749][2713170] Fps is (10 sec: 26214.4, 60 sec: 26146.2, 300 sec: 27061.4). Total num frames: 17444864. Throughput: 0: 6556.7. Samples: 4348854. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:19:05,751][2713170] Avg episode reward: [(0, '26.407')] [2025-03-28 18:19:05,798][2761575] Updated weights for policy 0, policy_version 4260 (0.0006) [2025-03-28 18:19:07,289][2761575] Updated weights for policy 0, policy_version 4270 (0.0007) [2025-03-28 18:19:08,852][2761575] Updated weights for policy 0, policy_version 4280 (0.0006) [2025-03-28 18:19:10,749][2713170] Fps is (10 sec: 25804.8, 60 sec: 26214.4, 300 sec: 27019.7). Total num frames: 17567744. Throughput: 0: 6504.4. Samples: 4389084. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:19:10,750][2713170] Avg episode reward: [(0, '27.930')] [2025-03-28 18:19:11,644][2761575] Updated weights for policy 0, policy_version 4290 (0.0006) [2025-03-28 18:19:13,181][2761575] Updated weights for policy 0, policy_version 4300 (0.0006) [2025-03-28 18:19:14,513][2761575] Updated weights for policy 0, policy_version 4310 (0.0006) [2025-03-28 18:19:15,749][2713170] Fps is (10 sec: 24166.4, 60 sec: 26146.1, 300 sec: 26950.3). Total num frames: 17686528. Throughput: 0: 6405.9. Samples: 4422798. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-03-28 18:19:15,750][2713170] Avg episode reward: [(0, '26.808')] [2025-03-28 18:19:15,964][2761575] Updated weights for policy 0, policy_version 4320 (0.0007) [2025-03-28 18:19:17,450][2761575] Updated weights for policy 0, policy_version 4330 (0.0007) [2025-03-28 18:19:18,958][2761575] Updated weights for policy 0, policy_version 4340 (0.0006) [2025-03-28 18:19:20,313][2761575] Updated weights for policy 0, policy_version 4350 (0.0006) [2025-03-28 18:19:20,749][2713170] Fps is (10 sec: 25804.7, 60 sec: 25941.3, 300 sec: 26950.3). Total num frames: 17825792. Throughput: 0: 6404.9. Samples: 4443370. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:19:20,750][2713170] Avg episode reward: [(0, '28.902')] [2025-03-28 18:19:21,858][2761575] Updated weights for policy 0, policy_version 4360 (0.0006) [2025-03-28 18:19:23,357][2761575] Updated weights for policy 0, policy_version 4370 (0.0006) [2025-03-28 18:19:24,842][2761575] Updated weights for policy 0, policy_version 4380 (0.0007) [2025-03-28 18:19:25,749][2713170] Fps is (10 sec: 27443.2, 60 sec: 25873.1, 300 sec: 26991.9). Total num frames: 17960960. Throughput: 0: 6443.9. Samples: 4485166. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:19:25,750][2713170] Avg episode reward: [(0, '25.973')] [2025-03-28 18:19:26,422][2761575] Updated weights for policy 0, policy_version 4390 (0.0006) [2025-03-28 18:19:28,062][2761575] Updated weights for policy 0, policy_version 4400 (0.0007) [2025-03-28 18:19:29,587][2761575] Updated weights for policy 0, policy_version 4410 (0.0006) [2025-03-28 18:19:30,749][2713170] Fps is (10 sec: 24985.6, 60 sec: 25668.3, 300 sec: 26978.1). Total num frames: 18075648. Throughput: 0: 6383.6. Samples: 4521038. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:19:30,750][2713170] Avg episode reward: [(0, '26.948')] [2025-03-28 18:19:31,585][2761575] Updated weights for policy 0, policy_version 4420 (0.0007) [2025-03-28 18:19:33,028][2761575] Updated weights for policy 0, policy_version 4430 (0.0006) [2025-03-28 18:19:34,489][2761575] Updated weights for policy 0, policy_version 4440 (0.0006) [2025-03-28 18:19:35,749][2713170] Fps is (10 sec: 26214.4, 60 sec: 25873.1, 300 sec: 27005.8). Total num frames: 18223104. Throughput: 0: 6435.7. Samples: 4542104. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:19:35,750][2713170] Avg episode reward: [(0, '28.764')] [2025-03-28 18:19:35,869][2761575] Updated weights for policy 0, policy_version 4450 (0.0006) [2025-03-28 18:19:37,485][2761575] Updated weights for policy 0, policy_version 4460 (0.0006) [2025-03-28 18:19:38,969][2761575] Updated weights for policy 0, policy_version 4470 (0.0006) [2025-03-28 18:19:40,601][2761575] Updated weights for policy 0, policy_version 4480 (0.0007) [2025-03-28 18:19:40,749][2713170] Fps is (10 sec: 27443.3, 60 sec: 25804.8, 300 sec: 26978.1). Total num frames: 18350080. Throughput: 0: 6492.3. Samples: 4582742. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:19:40,750][2713170] Avg episode reward: [(0, '30.547')] [2025-03-28 18:19:42,068][2761575] Updated weights for policy 0, policy_version 4490 (0.0007) [2025-03-28 18:19:43,555][2761575] Updated weights for policy 0, policy_version 4500 (0.0007) [2025-03-28 18:19:44,986][2761575] Updated weights for policy 0, policy_version 4510 (0.0006) [2025-03-28 18:19:45,749][2713170] Fps is (10 sec: 26623.8, 60 sec: 26009.6, 300 sec: 26991.9). Total num frames: 18489344. Throughput: 0: 6582.0. Samples: 4623892. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:19:45,751][2713170] Avg episode reward: [(0, '26.035')] [2025-03-28 18:19:46,500][2761575] Updated weights for policy 0, policy_version 4520 (0.0006) [2025-03-28 18:19:47,963][2761575] Updated weights for policy 0, policy_version 4530 (0.0006) [2025-03-28 18:19:49,401][2761575] Updated weights for policy 0, policy_version 4540 (0.0006) [2025-03-28 18:19:50,749][2713170] Fps is (10 sec: 28262.1, 60 sec: 26282.7, 300 sec: 27005.8). Total num frames: 18632704. Throughput: 0: 6577.1. Samples: 4644824. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) [2025-03-28 18:19:50,751][2713170] Avg episode reward: [(0, '25.006')] [2025-03-28 18:19:50,881][2761575] Updated weights for policy 0, policy_version 4550 (0.0007) [2025-03-28 18:19:52,418][2761575] Updated weights for policy 0, policy_version 4560 (0.0007) [2025-03-28 18:19:54,046][2761575] Updated weights for policy 0, policy_version 4570 (0.0006) [2025-03-28 18:19:55,562][2761575] Updated weights for policy 0, policy_version 4580 (0.0007) [2025-03-28 18:19:55,749][2713170] Fps is (10 sec: 27443.3, 60 sec: 26350.9, 300 sec: 27061.4). Total num frames: 18763776. Throughput: 0: 6574.1. Samples: 4684920. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:19:55,750][2713170] Avg episode reward: [(0, '26.334')] [2025-03-28 18:19:55,757][2761553] Saving /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000004581_18763776.pth... [2025-03-28 18:19:55,859][2761553] Removing /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000002999_12283904.pth [2025-03-28 18:19:56,990][2761575] Updated weights for policy 0, policy_version 4590 (0.0007) [2025-03-28 18:19:58,372][2761575] Updated weights for policy 0, policy_version 4600 (0.0006) [2025-03-28 18:19:59,988][2761575] Updated weights for policy 0, policy_version 4610 (0.0007) [2025-03-28 18:20:00,749][2713170] Fps is (10 sec: 24985.7, 60 sec: 26214.4, 300 sec: 27047.5). Total num frames: 18882560. Throughput: 0: 6301.3. Samples: 4706356. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:20:00,751][2713170] Avg episode reward: [(0, '27.610')] [2025-03-28 18:20:02,122][2761575] Updated weights for policy 0, policy_version 4620 (0.0007) [2025-03-28 18:20:03,624][2761575] Updated weights for policy 0, policy_version 4630 (0.0007) [2025-03-28 18:20:05,258][2761575] Updated weights for policy 0, policy_version 4640 (0.0007) [2025-03-28 18:20:05,749][2713170] Fps is (10 sec: 25395.0, 60 sec: 26214.4, 300 sec: 27019.7). Total num frames: 19017728. Throughput: 0: 6645.8. Samples: 4742430. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:20:05,751][2713170] Avg episode reward: [(0, '30.438')] [2025-03-28 18:20:06,716][2761575] Updated weights for policy 0, policy_version 4650 (0.0007) [2025-03-28 18:20:08,237][2761575] Updated weights for policy 0, policy_version 4660 (0.0007) [2025-03-28 18:20:09,843][2761575] Updated weights for policy 0, policy_version 4670 (0.0007) [2025-03-28 18:20:10,749][2713170] Fps is (10 sec: 26624.1, 60 sec: 26350.9, 300 sec: 26978.1). Total num frames: 19148800. Throughput: 0: 6598.0. Samples: 4782074. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:20:10,751][2713170] Avg episode reward: [(0, '26.418')] [2025-03-28 18:20:11,499][2761575] Updated weights for policy 0, policy_version 4680 (0.0007) [2025-03-28 18:20:13,157][2761575] Updated weights for policy 0, policy_version 4690 (0.0006) [2025-03-28 18:20:14,452][2761575] Updated weights for policy 0, policy_version 4700 (0.0006) [2025-03-28 18:20:15,749][2713170] Fps is (10 sec: 26624.3, 60 sec: 26624.0, 300 sec: 26964.2). Total num frames: 19283968. Throughput: 0: 6695.6. Samples: 4822340. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:20:15,751][2713170] Avg episode reward: [(0, '26.633')] [2025-03-28 18:20:15,973][2761575] Updated weights for policy 0, policy_version 4710 (0.0007) [2025-03-28 18:20:17,517][2761575] Updated weights for policy 0, policy_version 4720 (0.0007) [2025-03-28 18:20:19,069][2761575] Updated weights for policy 0, policy_version 4730 (0.0006) [2025-03-28 18:20:20,611][2761575] Updated weights for policy 0, policy_version 4740 (0.0006) [2025-03-28 18:20:20,749][2713170] Fps is (10 sec: 26624.0, 60 sec: 26487.5, 300 sec: 26936.4). Total num frames: 19415040. Throughput: 0: 6666.3. Samples: 4842088. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:20:20,750][2713170] Avg episode reward: [(0, '26.869')] [2025-03-28 18:20:22,165][2761575] Updated weights for policy 0, policy_version 4750 (0.0006) [2025-03-28 18:20:23,673][2761575] Updated weights for policy 0, policy_version 4760 (0.0006) [2025-03-28 18:20:25,116][2761575] Updated weights for policy 0, policy_version 4770 (0.0006) [2025-03-28 18:20:25,749][2713170] Fps is (10 sec: 27033.6, 60 sec: 26555.7, 300 sec: 26936.4). Total num frames: 19554304. Throughput: 0: 6655.9. Samples: 4882256. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) [2025-03-28 18:20:25,751][2713170] Avg episode reward: [(0, '30.615')] [2025-03-28 18:20:26,629][2761575] Updated weights for policy 0, policy_version 4780 (0.0006) [2025-03-28 18:20:28,286][2761575] Updated weights for policy 0, policy_version 4790 (0.0006) [2025-03-28 18:20:29,932][2761575] Updated weights for policy 0, policy_version 4800 (0.0006) [2025-03-28 18:20:30,749][2713170] Fps is (10 sec: 24985.8, 60 sec: 26487.5, 300 sec: 26867.0). Total num frames: 19664896. Throughput: 0: 6188.8. Samples: 4902386. Policy #0 lag: (min: 0.0, avg: 0.8, max: 1.0) [2025-03-28 18:20:30,750][2713170] Avg episode reward: [(0, '28.399')] [2025-03-28 18:20:32,026][2761575] Updated weights for policy 0, policy_version 4810 (0.0006) [2025-03-28 18:20:33,578][2761575] Updated weights for policy 0, policy_version 4820 (0.0006) [2025-03-28 18:20:35,092][2761575] Updated weights for policy 0, policy_version 4830 (0.0006) [2025-03-28 18:20:35,749][2713170] Fps is (10 sec: 24576.0, 60 sec: 26282.7, 300 sec: 26853.1). Total num frames: 19800064. Throughput: 0: 6502.9. Samples: 4937454. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) [2025-03-28 18:20:35,750][2713170] Avg episode reward: [(0, '27.893')] [2025-03-28 18:20:36,608][2761575] Updated weights for policy 0, policy_version 4840 (0.0006) [2025-03-28 18:20:38,203][2761575] Updated weights for policy 0, policy_version 4850 (0.0006) [2025-03-28 18:20:39,719][2761575] Updated weights for policy 0, policy_version 4860 (0.0006) [2025-03-28 18:20:40,749][2713170] Fps is (10 sec: 26623.1, 60 sec: 26350.8, 300 sec: 26839.2). Total num frames: 19931136. Throughput: 0: 6502.6. Samples: 4977538. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) [2025-03-28 18:20:40,751][2713170] Avg episode reward: [(0, '27.686')] [2025-03-28 18:20:41,244][2761575] Updated weights for policy 0, policy_version 4870 (0.0007) [2025-03-28 18:20:42,789][2761575] Updated weights for policy 0, policy_version 4880 (0.0006) [2025-03-28 18:20:43,401][2761553] Stopping Batcher_0... [2025-03-28 18:20:43,402][2761553] Loop batcher_evt_loop terminating... [2025-03-28 18:20:43,401][2713170] Component Batcher_0 stopped! [2025-03-28 18:20:43,403][2761553] Saving /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000004884_20004864.pth... [2025-03-28 18:20:43,439][2761575] Weights refcount: 2 0 [2025-03-28 18:20:43,479][2761553] Removing /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000003815_15626240.pth [2025-03-28 18:20:43,482][2761553] Saving /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/checkpoint_p0/checkpoint_000004884_20004864.pth... [2025-03-28 18:20:43,488][2761574] Stopping RolloutWorker_w0... [2025-03-28 18:20:43,489][2761574] Loop rollout_proc0_evt_loop terminating... [2025-03-28 18:20:43,488][2713170] Component RolloutWorker_w0 stopped! [2025-03-28 18:20:43,496][2713170] Component RolloutWorker_w7 stopped! [2025-03-28 18:20:43,496][2761582] Stopping RolloutWorker_w7... [2025-03-28 18:20:43,497][2761582] Loop rollout_proc7_evt_loop terminating... [2025-03-28 18:20:43,501][2761576] Stopping RolloutWorker_w1... [2025-03-28 18:20:43,501][2713170] Component RolloutWorker_w1 stopped! [2025-03-28 18:20:43,502][2761576] Loop rollout_proc1_evt_loop terminating... [2025-03-28 18:20:43,502][2713170] Component RolloutWorker_w4 stopped! [2025-03-28 18:20:43,502][2761578] Stopping RolloutWorker_w4... [2025-03-28 18:20:43,503][2713170] Component RolloutWorker_w2 stopped! [2025-03-28 18:20:43,503][2761578] Loop rollout_proc4_evt_loop terminating... [2025-03-28 18:20:43,503][2761577] Stopping RolloutWorker_w2... [2025-03-28 18:20:43,504][2761577] Loop rollout_proc2_evt_loop terminating... [2025-03-28 18:20:43,506][2713170] Component RolloutWorker_w6 stopped! [2025-03-28 18:20:43,506][2713170] Component RolloutWorker_w3 stopped! [2025-03-28 18:20:43,506][2761581] Stopping RolloutWorker_w6... [2025-03-28 18:20:43,507][2761581] Loop rollout_proc6_evt_loop terminating... [2025-03-28 18:20:43,507][2761580] Stopping RolloutWorker_w3... [2025-03-28 18:20:43,508][2761580] Loop rollout_proc3_evt_loop terminating... [2025-03-28 18:20:43,509][2713170] Component RolloutWorker_w5 stopped! [2025-03-28 18:20:43,509][2761579] Stopping RolloutWorker_w5... [2025-03-28 18:20:43,510][2761579] Loop rollout_proc5_evt_loop terminating... [2025-03-28 18:20:43,567][2761553] Stopping LearnerWorker_p0... [2025-03-28 18:20:43,568][2761553] Loop learner_proc0_evt_loop terminating... [2025-03-28 18:20:43,567][2713170] Component LearnerWorker_p0 stopped! [2025-03-28 18:20:44,673][2761575] Stopping InferenceWorker_p0-w0... [2025-03-28 18:20:44,674][2761575] Loop inference_proc0-0_evt_loop terminating... [2025-03-28 18:20:44,673][2713170] Component InferenceWorker_p0-w0 stopped! [2025-03-28 18:20:44,675][2713170] Waiting for process learner_proc0 to stop... [2025-03-28 18:20:44,792][2713170] Waiting for process inference_proc0-0 to join... [2025-03-28 18:20:45,238][2713170] Waiting for process rollout_proc0 to join... [2025-03-28 18:20:45,239][2713170] Waiting for process rollout_proc1 to join... [2025-03-28 18:20:45,240][2713170] Waiting for process rollout_proc2 to join... [2025-03-28 18:20:45,241][2713170] Waiting for process rollout_proc3 to join... [2025-03-28 18:20:45,242][2713170] Waiting for process rollout_proc4 to join... [2025-03-28 18:20:45,243][2713170] Waiting for process rollout_proc5 to join... [2025-03-28 18:20:45,244][2713170] Waiting for process rollout_proc6 to join... [2025-03-28 18:20:45,245][2713170] Waiting for process rollout_proc7 to join... [2025-03-28 18:20:45,246][2713170] Batcher 0 profile tree view: batching: 77.5075, releasing_batches: 0.1343 [2025-03-28 18:20:45,247][2713170] InferenceWorker_p0-w0 profile tree view: wait_policy: 0.0000 wait_policy_total: 23.1149 update_model: 9.8765 weight_update: 0.0006 one_step: 0.0013 handle_policy_step: 682.7108 deserialize: 52.1600, stack: 3.4602, obs_to_device_normalize: 153.6694, forward: 309.7647, send_messages: 38.3735 prepare_outputs: 94.5573 to_cpu: 56.9584 [2025-03-28 18:20:45,247][2713170] Learner 0 profile tree view: misc: 0.0224, prepare_batch: 28.5087 train: 78.0478 epoch_init: 0.0258, minibatch_init: 0.0250, losses_postprocess: 1.4478, kl_divergence: 1.7219, after_optimizer: 9.8786 calculate_losses: 33.1913 losses_init: 0.0203, forward_head: 3.4984, bptt_initial: 16.6980, tail: 2.4455, advantages_returns: 0.6670, losses: 4.6291 bptt: 4.4250 bptt_forward_core: 4.1954 update: 30.0998 clip: 3.4030 [2025-03-28 18:20:45,248][2713170] RolloutWorker_w0 profile tree view: wait_for_trajectories: 0.6317, enqueue_policy_requests: 31.6726, env_step: 475.5426, overhead: 37.4830, complete_rollouts: 0.9234 save_policy_outputs: 36.5034 split_output_tensors: 17.4458 [2025-03-28 18:20:45,249][2713170] RolloutWorker_w7 profile tree view: wait_for_trajectories: 0.6550, enqueue_policy_requests: 38.5978, env_step: 468.0902, overhead: 38.3842, complete_rollouts: 0.9600 save_policy_outputs: 37.3333 split_output_tensors: 17.8236 [2025-03-28 18:20:45,249][2713170] Loop Runner_EvtLoop terminating... [2025-03-28 18:20:45,250][2713170] Runner profile tree view: main_loop: 764.5789 [2025-03-28 18:20:45,251][2713170] Collected {0: 20004864}, FPS: 26164.5 [2025-03-28 18:21:04,142][2713170] Loading existing experiment configuration from /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/config.json [2025-03-28 18:21:04,143][2713170] Overriding arg 'num_workers' with value 1 passed from command line [2025-03-28 18:21:04,144][2713170] Adding new argument 'no_render'=True that is not in the saved config file! [2025-03-28 18:21:04,145][2713170] Adding new argument 'save_video'=True that is not in the saved config file! [2025-03-28 18:21:04,145][2713170] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! [2025-03-28 18:21:04,146][2713170] Adding new argument 'video_name'=None that is not in the saved config file! [2025-03-28 18:21:04,148][2713170] Adding new argument 'max_num_frames'=100000 that is not in the saved config file! [2025-03-28 18:21:04,148][2713170] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! [2025-03-28 18:21:04,149][2713170] Adding new argument 'push_to_hub'=True that is not in the saved config file! [2025-03-28 18:21:04,150][2713170] Adding new argument 'hf_repository'='stalaei/DeepRL_vizdoom_health_gathering_supreme' that is not in the saved config file! [2025-03-28 18:21:04,151][2713170] Adding new argument 'policy_index'=0 that is not in the saved config file! [2025-03-28 18:21:04,152][2713170] Adding new argument 'eval_deterministic'=False that is not in the saved config file! [2025-03-28 18:21:04,153][2713170] Adding new argument 'train_script'=None that is not in the saved config file! [2025-03-28 18:21:04,153][2713170] Adding new argument 'enjoy_script'=None that is not in the saved config file! [2025-03-28 18:21:04,154][2713170] Using frameskip 1 and render_action_repeat=4 for evaluation [2025-03-28 18:21:04,192][2713170] RunningMeanStd input shape: (3, 72, 128) [2025-03-28 18:21:04,194][2713170] RunningMeanStd input shape: (1,) [2025-03-28 18:21:04,209][2713170] ConvEncoder: input_channels=3 [2025-03-28 18:21:04,240][2713170] Conv encoder output size: 512 [2025-03-28 18:21:04,241][2713170] Policy head output size: 512 [2025-03-28 18:21:04,908][2713170] Num frames 100... [2025-03-28 18:21:05,006][2713170] Num frames 200... [2025-03-28 18:21:05,103][2713170] Num frames 300... [2025-03-28 18:21:05,203][2713170] Num frames 400... [2025-03-28 18:21:05,335][2713170] Avg episode rewards: #0: 6.800, true rewards: #0: 4.800 [2025-03-28 18:21:05,336][2713170] Avg episode reward: 6.800, avg true_objective: 4.800 [2025-03-28 18:21:05,357][2713170] Num frames 500... [2025-03-28 18:21:05,455][2713170] Num frames 600... [2025-03-28 18:21:05,553][2713170] Num frames 700... [2025-03-28 18:21:05,652][2713170] Num frames 800... [2025-03-28 18:21:05,752][2713170] Num frames 900... [2025-03-28 18:21:05,851][2713170] Num frames 1000... [2025-03-28 18:21:05,951][2713170] Num frames 1100... [2025-03-28 18:21:06,053][2713170] Num frames 1200... [2025-03-28 18:21:06,160][2713170] Num frames 1300... [2025-03-28 18:21:06,271][2713170] Num frames 1400... [2025-03-28 18:21:06,381][2713170] Num frames 1500... [2025-03-28 18:21:06,493][2713170] Num frames 1600... [2025-03-28 18:21:06,602][2713170] Num frames 1700... [2025-03-28 18:21:06,709][2713170] Num frames 1800... [2025-03-28 18:21:06,820][2713170] Num frames 1900... [2025-03-28 18:21:06,930][2713170] Num frames 2000... [2025-03-28 18:21:07,031][2713170] Num frames 2100... [2025-03-28 18:21:07,118][2713170] Avg episode rewards: #0: 24.600, true rewards: #0: 10.600 [2025-03-28 18:21:07,119][2713170] Avg episode reward: 24.600, avg true_objective: 10.600 [2025-03-28 18:21:07,221][2713170] Num frames 2200... [2025-03-28 18:21:07,305][2713170] Num frames 2300... [2025-03-28 18:21:07,387][2713170] Num frames 2400... [2025-03-28 18:21:07,474][2713170] Num frames 2500... [2025-03-28 18:21:07,559][2713170] Num frames 2600... [2025-03-28 18:21:07,642][2713170] Num frames 2700... [2025-03-28 18:21:07,726][2713170] Num frames 2800... [2025-03-28 18:21:07,811][2713170] Num frames 2900... [2025-03-28 18:21:07,895][2713170] Num frames 3000... [2025-03-28 18:21:07,980][2713170] Num frames 3100... [2025-03-28 18:21:08,066][2713170] Num frames 3200... [2025-03-28 18:21:08,152][2713170] Num frames 3300... [2025-03-28 18:21:08,237][2713170] Num frames 3400... [2025-03-28 18:21:08,325][2713170] Num frames 3500... [2025-03-28 18:21:08,412][2713170] Num frames 3600... [2025-03-28 18:21:08,498][2713170] Num frames 3700... [2025-03-28 18:21:08,583][2713170] Num frames 3800... [2025-03-28 18:21:08,671][2713170] Num frames 3900... [2025-03-28 18:21:08,740][2713170] Avg episode rewards: #0: 32.706, true rewards: #0: 13.040 [2025-03-28 18:21:08,741][2713170] Avg episode reward: 32.706, avg true_objective: 13.040 [2025-03-28 18:21:08,835][2713170] Num frames 4000... [2025-03-28 18:21:08,918][2713170] Num frames 4100... [2025-03-28 18:21:09,000][2713170] Num frames 4200... [2025-03-28 18:21:09,080][2713170] Num frames 4300... [2025-03-28 18:21:09,161][2713170] Num frames 4400... [2025-03-28 18:21:09,240][2713170] Num frames 4500... [2025-03-28 18:21:09,321][2713170] Num frames 4600... [2025-03-28 18:21:09,400][2713170] Num frames 4700... [2025-03-28 18:21:09,484][2713170] Num frames 4800... [2025-03-28 18:21:09,567][2713170] Num frames 4900... [2025-03-28 18:21:09,652][2713170] Num frames 5000... [2025-03-28 18:21:09,739][2713170] Num frames 5100... [2025-03-28 18:21:09,828][2713170] Num frames 5200... [2025-03-28 18:21:09,919][2713170] Num frames 5300... [2025-03-28 18:21:10,008][2713170] Num frames 5400... [2025-03-28 18:21:10,092][2713170] Num frames 5500... [2025-03-28 18:21:10,177][2713170] Num frames 5600... [2025-03-28 18:21:10,262][2713170] Num frames 5700... [2025-03-28 18:21:10,348][2713170] Num frames 5800... [2025-03-28 18:21:10,432][2713170] Num frames 5900... [2025-03-28 18:21:10,518][2713170] Num frames 6000... [2025-03-28 18:21:10,584][2713170] Avg episode rewards: #0: 38.029, true rewards: #0: 15.030 [2025-03-28 18:21:10,585][2713170] Avg episode reward: 38.029, avg true_objective: 15.030 [2025-03-28 18:21:10,660][2713170] Num frames 6100... [2025-03-28 18:21:10,747][2713170] Num frames 6200... [2025-03-28 18:21:10,831][2713170] Num frames 6300... [2025-03-28 18:21:10,919][2713170] Avg episode rewards: #0: 31.877, true rewards: #0: 12.678 [2025-03-28 18:21:10,920][2713170] Avg episode reward: 31.877, avg true_objective: 12.678 [2025-03-28 18:21:10,975][2713170] Num frames 6400... [2025-03-28 18:21:11,059][2713170] Num frames 6500... [2025-03-28 18:21:11,146][2713170] Num frames 6600... [2025-03-28 18:21:11,235][2713170] Num frames 6700... [2025-03-28 18:21:11,321][2713170] Num frames 6800... [2025-03-28 18:21:11,406][2713170] Num frames 6900... [2025-03-28 18:21:11,491][2713170] Num frames 7000... [2025-03-28 18:21:11,579][2713170] Num frames 7100... [2025-03-28 18:21:11,663][2713170] Num frames 7200... [2025-03-28 18:21:11,750][2713170] Num frames 7300... [2025-03-28 18:21:11,832][2713170] Num frames 7400... [2025-03-28 18:21:11,914][2713170] Num frames 7500... [2025-03-28 18:21:11,999][2713170] Num frames 7600... [2025-03-28 18:21:12,098][2713170] Avg episode rewards: #0: 31.918, true rewards: #0: 12.752 [2025-03-28 18:21:12,099][2713170] Avg episode reward: 31.918, avg true_objective: 12.752 [2025-03-28 18:21:12,143][2713170] Num frames 7700... [2025-03-28 18:21:12,231][2713170] Num frames 7800... [2025-03-28 18:21:12,315][2713170] Num frames 7900... [2025-03-28 18:21:12,402][2713170] Num frames 8000... [2025-03-28 18:21:12,488][2713170] Num frames 8100... [2025-03-28 18:21:12,580][2713170] Num frames 8200... [2025-03-28 18:21:12,664][2713170] Num frames 8300... [2025-03-28 18:21:12,746][2713170] Num frames 8400... [2025-03-28 18:21:12,829][2713170] Num frames 8500... [2025-03-28 18:21:12,950][2713170] Avg episode rewards: #0: 29.967, true rewards: #0: 12.253 [2025-03-28 18:21:12,951][2713170] Avg episode reward: 29.967, avg true_objective: 12.253 [2025-03-28 18:21:12,982][2713170] Num frames 8600... [2025-03-28 18:21:13,075][2713170] Num frames 8700... [2025-03-28 18:21:13,160][2713170] Num frames 8800... [2025-03-28 18:21:13,245][2713170] Num frames 8900... [2025-03-28 18:21:13,335][2713170] Num frames 9000... [2025-03-28 18:21:13,421][2713170] Num frames 9100... [2025-03-28 18:21:13,509][2713170] Num frames 9200... [2025-03-28 18:21:13,598][2713170] Num frames 9300... [2025-03-28 18:21:13,686][2713170] Num frames 9400... [2025-03-28 18:21:13,778][2713170] Num frames 9500... [2025-03-28 18:21:13,867][2713170] Num frames 9600... [2025-03-28 18:21:13,955][2713170] Num frames 9700... [2025-03-28 18:21:14,044][2713170] Num frames 9800... [2025-03-28 18:21:14,132][2713170] Num frames 9900... [2025-03-28 18:21:14,222][2713170] Num frames 10000... [2025-03-28 18:21:14,309][2713170] Num frames 10100... [2025-03-28 18:21:14,397][2713170] Num frames 10200... [2025-03-28 18:21:14,485][2713170] Num frames 10300... [2025-03-28 18:21:14,575][2713170] Num frames 10400... [2025-03-28 18:21:14,665][2713170] Num frames 10500... [2025-03-28 18:21:14,753][2713170] Num frames 10600... [2025-03-28 18:21:14,879][2713170] Avg episode rewards: #0: 33.721, true rewards: #0: 13.346 [2025-03-28 18:21:14,880][2713170] Avg episode reward: 33.721, avg true_objective: 13.346 [2025-03-28 18:21:14,907][2713170] Num frames 10700... [2025-03-28 18:21:15,001][2713170] Num frames 10800... [2025-03-28 18:21:15,091][2713170] Num frames 10900... [2025-03-28 18:21:15,181][2713170] Num frames 11000... [2025-03-28 18:21:15,270][2713170] Num frames 11100... [2025-03-28 18:21:15,357][2713170] Num frames 11200... [2025-03-28 18:21:15,446][2713170] Num frames 11300... [2025-03-28 18:21:15,537][2713170] Num frames 11400... [2025-03-28 18:21:15,626][2713170] Num frames 11500... [2025-03-28 18:21:15,715][2713170] Num frames 11600... [2025-03-28 18:21:15,804][2713170] Num frames 11700... [2025-03-28 18:21:15,895][2713170] Num frames 11800... [2025-03-28 18:21:15,979][2713170] Avg episode rewards: #0: 32.810, true rewards: #0: 13.143 [2025-03-28 18:21:15,980][2713170] Avg episode reward: 32.810, avg true_objective: 13.143 [2025-03-28 18:21:16,062][2713170] Num frames 11900... [2025-03-28 18:21:16,153][2713170] Num frames 12000... [2025-03-28 18:21:16,243][2713170] Num frames 12100... [2025-03-28 18:21:16,332][2713170] Num frames 12200... [2025-03-28 18:21:16,422][2713170] Num frames 12300... [2025-03-28 18:21:16,511][2713170] Num frames 12400... [2025-03-28 18:21:16,600][2713170] Num frames 12500... [2025-03-28 18:21:16,688][2713170] Num frames 12600... [2025-03-28 18:21:16,774][2713170] Num frames 12700... [2025-03-28 18:21:16,859][2713170] Num frames 12800... [2025-03-28 18:21:16,944][2713170] Num frames 12900... [2025-03-28 18:21:17,018][2713170] Avg episode rewards: #0: 31.721, true rewards: #0: 12.921 [2025-03-28 18:21:17,019][2713170] Avg episode reward: 31.721, avg true_objective: 12.921 [2025-03-28 18:21:22,578][2713170] Replay video saved to /home/stalaei/projects/deep-rl-class/notebooks/unit8/train_dir/default_experiment/replay.mp4!