diff --git "a/sf_log.txt" "b/sf_log.txt" new file mode 100644--- /dev/null +++ "b/sf_log.txt" @@ -0,0 +1,1022 @@ +[2023-02-23 20:51:56,996][00363] Saving configuration to /content/train_dir/default_experiment/config.json... +[2023-02-23 20:51:56,999][00363] Rollout worker 0 uses device cpu +[2023-02-23 20:51:57,000][00363] Rollout worker 1 uses device cpu +[2023-02-23 20:51:57,002][00363] Rollout worker 2 uses device cpu +[2023-02-23 20:51:57,003][00363] Rollout worker 3 uses device cpu +[2023-02-23 20:51:57,004][00363] Rollout worker 4 uses device cpu +[2023-02-23 20:51:57,008][00363] Rollout worker 5 uses device cpu +[2023-02-23 20:51:57,009][00363] Rollout worker 6 uses device cpu +[2023-02-23 20:51:57,011][00363] Rollout worker 7 uses device cpu +[2023-02-23 20:51:57,211][00363] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-23 20:51:57,212][00363] InferenceWorker_p0-w0: min num requests: 2 +[2023-02-23 20:51:57,245][00363] Starting all processes... +[2023-02-23 20:51:57,246][00363] Starting process learner_proc0 +[2023-02-23 20:51:57,299][00363] Starting all processes... +[2023-02-23 20:51:57,307][00363] Starting process inference_proc0-0 +[2023-02-23 20:51:57,308][00363] Starting process rollout_proc0 +[2023-02-23 20:51:57,309][00363] Starting process rollout_proc1 +[2023-02-23 20:51:57,310][00363] Starting process rollout_proc2 +[2023-02-23 20:51:57,310][00363] Starting process rollout_proc3 +[2023-02-23 20:51:57,310][00363] Starting process rollout_proc4 +[2023-02-23 20:51:57,310][00363] Starting process rollout_proc5 +[2023-02-23 20:51:57,310][00363] Starting process rollout_proc6 +[2023-02-23 20:51:57,311][00363] Starting process rollout_proc7 +[2023-02-23 20:52:10,065][12009] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-23 20:52:10,065][12009] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0 +[2023-02-23 20:52:10,176][12030] Worker 2 uses CPU cores [0] +[2023-02-23 20:52:10,346][12033] Worker 5 uses CPU cores [1] +[2023-02-23 20:52:10,353][12035] Worker 7 uses CPU cores [1] +[2023-02-23 20:52:10,387][12023] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-23 20:52:10,387][12023] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0 +[2023-02-23 20:52:10,405][12031] Worker 3 uses CPU cores [1] +[2023-02-23 20:52:10,426][12024] Worker 0 uses CPU cores [0] +[2023-02-23 20:52:10,448][12032] Worker 4 uses CPU cores [0] +[2023-02-23 20:52:10,509][12026] Worker 1 uses CPU cores [1] +[2023-02-23 20:52:10,547][12034] Worker 6 uses CPU cores [0] +[2023-02-23 20:52:10,998][12009] Num visible devices: 1 +[2023-02-23 20:52:10,997][12023] Num visible devices: 1 +[2023-02-23 20:52:11,009][12009] Starting seed is not provided +[2023-02-23 20:52:11,011][12009] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-23 20:52:11,011][12009] Initializing actor-critic model on device cuda:0 +[2023-02-23 20:52:11,012][12009] RunningMeanStd input shape: (3, 72, 128) +[2023-02-23 20:52:11,015][12009] RunningMeanStd input shape: (1,) +[2023-02-23 20:52:11,034][12009] ConvEncoder: input_channels=3 +[2023-02-23 20:52:11,465][12009] Conv encoder output size: 512 +[2023-02-23 20:52:11,466][12009] Policy head output size: 512 +[2023-02-23 20:52:11,534][12009] Created Actor Critic model with architecture: +[2023-02-23 20:52:11,535][12009] ActorCriticSharedWeights( + (obs_normalizer): ObservationNormalizer( + (running_mean_std): RunningMeanStdDictInPlace( + (running_mean_std): ModuleDict( + (obs): RunningMeanStdInPlace() + ) + ) + ) + (returns_normalizer): RecursiveScriptModule(original_name=RunningMeanStdInPlace) + (encoder): VizdoomEncoder( + (basic_encoder): ConvEncoder( + (enc): RecursiveScriptModule( + original_name=ConvEncoderImpl + (conv_head): RecursiveScriptModule( + original_name=Sequential + (0): RecursiveScriptModule(original_name=Conv2d) + (1): RecursiveScriptModule(original_name=ELU) + (2): RecursiveScriptModule(original_name=Conv2d) + (3): RecursiveScriptModule(original_name=ELU) + (4): RecursiveScriptModule(original_name=Conv2d) + (5): RecursiveScriptModule(original_name=ELU) + ) + (mlp_layers): RecursiveScriptModule( + original_name=Sequential + (0): RecursiveScriptModule(original_name=Linear) + (1): RecursiveScriptModule(original_name=ELU) + ) + ) + ) + ) + (core): ModelCoreRNN( + (core): GRU(512, 512) + ) + (decoder): MlpDecoder( + (mlp): Identity() + ) + (critic_linear): Linear(in_features=512, out_features=1, bias=True) + (action_parameterization): ActionParameterizationDefault( + (distribution_linear): Linear(in_features=512, out_features=5, bias=True) + ) +) +[2023-02-23 20:52:17,203][00363] Heartbeat connected on Batcher_0 +[2023-02-23 20:52:17,211][00363] Heartbeat connected on InferenceWorker_p0-w0 +[2023-02-23 20:52:17,222][00363] Heartbeat connected on RolloutWorker_w0 +[2023-02-23 20:52:17,225][00363] Heartbeat connected on RolloutWorker_w1 +[2023-02-23 20:52:17,228][00363] Heartbeat connected on RolloutWorker_w2 +[2023-02-23 20:52:17,231][00363] Heartbeat connected on RolloutWorker_w3 +[2023-02-23 20:52:17,235][00363] Heartbeat connected on RolloutWorker_w4 +[2023-02-23 20:52:17,239][00363] Heartbeat connected on RolloutWorker_w5 +[2023-02-23 20:52:17,240][00363] Heartbeat connected on RolloutWorker_w6 +[2023-02-23 20:52:17,248][00363] Heartbeat connected on RolloutWorker_w7 +[2023-02-23 20:52:20,038][12009] Using optimizer +[2023-02-23 20:52:20,040][12009] No checkpoints found +[2023-02-23 20:52:20,041][12009] Did not load from checkpoint, starting from scratch! +[2023-02-23 20:52:20,042][12009] Initialized policy 0 weights for model version 0 +[2023-02-23 20:52:20,044][12009] Using GPUs [0] for process 0 (actually maps to GPUs [0]) +[2023-02-23 20:52:20,052][12009] LearnerWorker_p0 finished initialization! +[2023-02-23 20:52:20,053][00363] Heartbeat connected on LearnerWorker_p0 +[2023-02-23 20:52:20,156][12023] RunningMeanStd input shape: (3, 72, 128) +[2023-02-23 20:52:20,157][12023] RunningMeanStd input shape: (1,) +[2023-02-23 20:52:20,173][12023] ConvEncoder: input_channels=3 +[2023-02-23 20:52:20,271][12023] Conv encoder output size: 512 +[2023-02-23 20:52:20,272][12023] Policy head output size: 512 +[2023-02-23 20:52:22,504][00363] Inference worker 0-0 is ready! +[2023-02-23 20:52:22,506][00363] All inference workers are ready! Signal rollout workers to start! +[2023-02-23 20:52:22,629][12031] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-23 20:52:22,632][12033] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-23 20:52:22,641][12035] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-23 20:52:22,644][12030] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-23 20:52:22,680][12026] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-23 20:52:22,679][12032] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-23 20:52:22,684][12034] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-23 20:52:22,677][12024] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-23 20:52:22,815][00363] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 0. Throughput: 0: nan. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) +[2023-02-23 20:52:23,842][12024] Decorrelating experience for 0 frames... +[2023-02-23 20:52:23,843][12031] Decorrelating experience for 0 frames... +[2023-02-23 20:52:23,845][12033] Decorrelating experience for 0 frames... +[2023-02-23 20:52:23,842][12035] Decorrelating experience for 0 frames... +[2023-02-23 20:52:23,845][12030] Decorrelating experience for 0 frames... +[2023-02-23 20:52:23,845][12032] Decorrelating experience for 0 frames... +[2023-02-23 20:52:24,205][12032] Decorrelating experience for 32 frames... +[2023-02-23 20:52:24,914][12024] Decorrelating experience for 32 frames... +[2023-02-23 20:52:25,052][12032] Decorrelating experience for 64 frames... +[2023-02-23 20:52:25,175][12033] Decorrelating experience for 32 frames... +[2023-02-23 20:52:25,178][12035] Decorrelating experience for 32 frames... +[2023-02-23 20:52:25,180][12031] Decorrelating experience for 32 frames... +[2023-02-23 20:52:25,187][12026] Decorrelating experience for 0 frames... +[2023-02-23 20:52:25,845][12030] Decorrelating experience for 32 frames... +[2023-02-23 20:52:25,915][12032] Decorrelating experience for 96 frames... +[2023-02-23 20:52:26,678][12035] Decorrelating experience for 64 frames... +[2023-02-23 20:52:26,702][12031] Decorrelating experience for 64 frames... +[2023-02-23 20:52:26,749][12033] Decorrelating experience for 64 frames... +[2023-02-23 20:52:26,843][12034] Decorrelating experience for 0 frames... +[2023-02-23 20:52:26,867][12030] Decorrelating experience for 64 frames... +[2023-02-23 20:52:27,613][12034] Decorrelating experience for 32 frames... +[2023-02-23 20:52:27,631][12030] Decorrelating experience for 96 frames... +[2023-02-23 20:52:27,790][12026] Decorrelating experience for 32 frames... +[2023-02-23 20:52:27,815][00363] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 0.0. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) +[2023-02-23 20:52:28,400][12024] Decorrelating experience for 64 frames... +[2023-02-23 20:52:28,513][12031] Decorrelating experience for 96 frames... +[2023-02-23 20:52:28,543][12033] Decorrelating experience for 96 frames... +[2023-02-23 20:52:29,026][12034] Decorrelating experience for 64 frames... +[2023-02-23 20:52:29,443][12034] Decorrelating experience for 96 frames... +[2023-02-23 20:52:29,479][12026] Decorrelating experience for 64 frames... +[2023-02-23 20:52:29,926][12024] Decorrelating experience for 96 frames... +[2023-02-23 20:52:30,117][12035] Decorrelating experience for 96 frames... +[2023-02-23 20:52:30,501][12026] Decorrelating experience for 96 frames... +[2023-02-23 20:52:32,815][00363] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 59.4. Samples: 594. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0) +[2023-02-23 20:52:32,821][00363] Avg episode reward: [(0, '1.024')] +[2023-02-23 20:52:35,063][12009] Signal inference workers to stop experience collection... +[2023-02-23 20:52:35,096][12023] InferenceWorker_p0-w0: stopping experience collection +[2023-02-23 20:52:37,662][12009] Signal inference workers to resume experience collection... +[2023-02-23 20:52:37,663][12023] InferenceWorker_p0-w0: resuming experience collection +[2023-02-23 20:52:37,821][00363] Fps is (10 sec: 409.4, 60 sec: 273.0, 300 sec: 273.0). Total num frames: 4096. Throughput: 0: 156.2. Samples: 2344. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0) +[2023-02-23 20:52:37,825][00363] Avg episode reward: [(0, '2.021')] +[2023-02-23 20:52:42,816][00363] Fps is (10 sec: 2457.6, 60 sec: 1228.8, 300 sec: 1228.8). Total num frames: 24576. Throughput: 0: 228.6. Samples: 4572. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:52:42,822][00363] Avg episode reward: [(0, '3.617')] +[2023-02-23 20:52:45,895][12023] Updated weights for policy 0, policy_version 10 (0.0031) +[2023-02-23 20:52:47,819][00363] Fps is (10 sec: 4506.5, 60 sec: 1965.8, 300 sec: 1965.8). Total num frames: 49152. Throughput: 0: 450.7. Samples: 11270. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) +[2023-02-23 20:52:47,833][00363] Avg episode reward: [(0, '4.322')] +[2023-02-23 20:52:52,819][00363] Fps is (10 sec: 3685.1, 60 sec: 2047.7, 300 sec: 2047.7). Total num frames: 61440. Throughput: 0: 531.3. Samples: 15942. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 20:52:52,822][00363] Avg episode reward: [(0, '4.425')] +[2023-02-23 20:52:57,816][00363] Fps is (10 sec: 2458.4, 60 sec: 2106.5, 300 sec: 2106.5). Total num frames: 73728. Throughput: 0: 510.3. Samples: 17860. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:52:57,818][00363] Avg episode reward: [(0, '4.408')] +[2023-02-23 20:52:59,810][12023] Updated weights for policy 0, policy_version 20 (0.0016) +[2023-02-23 20:53:02,816][00363] Fps is (10 sec: 3277.9, 60 sec: 2355.2, 300 sec: 2355.2). Total num frames: 94208. Throughput: 0: 575.4. Samples: 23016. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:53:02,825][00363] Avg episode reward: [(0, '4.428')] +[2023-02-23 20:53:07,827][00363] Fps is (10 sec: 4091.5, 60 sec: 2548.0, 300 sec: 2548.0). Total num frames: 114688. Throughput: 0: 662.3. Samples: 29812. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:53:07,835][00363] Avg episode reward: [(0, '4.336')] +[2023-02-23 20:53:07,909][12009] Saving new best policy, reward=4.336! +[2023-02-23 20:53:09,338][12023] Updated weights for policy 0, policy_version 30 (0.0015) +[2023-02-23 20:53:12,815][00363] Fps is (10 sec: 3686.5, 60 sec: 2621.4, 300 sec: 2621.4). Total num frames: 131072. Throughput: 0: 713.6. Samples: 32110. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:53:12,819][00363] Avg episode reward: [(0, '4.330')] +[2023-02-23 20:53:17,815][00363] Fps is (10 sec: 3280.5, 60 sec: 2681.0, 300 sec: 2681.0). Total num frames: 147456. Throughput: 0: 796.4. Samples: 36430. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:53:17,818][00363] Avg episode reward: [(0, '4.413')] +[2023-02-23 20:53:17,822][12009] Saving new best policy, reward=4.413! +[2023-02-23 20:53:21,851][12023] Updated weights for policy 0, policy_version 40 (0.0025) +[2023-02-23 20:53:22,817][00363] Fps is (10 sec: 3685.7, 60 sec: 2798.9, 300 sec: 2798.9). Total num frames: 167936. Throughput: 0: 887.4. Samples: 42272. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:53:22,821][00363] Avg episode reward: [(0, '4.484')] +[2023-02-23 20:53:22,829][12009] Saving new best policy, reward=4.484! +[2023-02-23 20:53:27,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3140.3, 300 sec: 2898.7). Total num frames: 188416. Throughput: 0: 911.2. Samples: 45578. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:53:27,822][00363] Avg episode reward: [(0, '4.532')] +[2023-02-23 20:53:27,825][12009] Saving new best policy, reward=4.532! +[2023-02-23 20:53:32,455][12023] Updated weights for policy 0, policy_version 50 (0.0023) +[2023-02-23 20:53:32,815][00363] Fps is (10 sec: 3687.1, 60 sec: 3413.3, 300 sec: 2925.7). Total num frames: 204800. Throughput: 0: 884.3. Samples: 51058. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:53:32,818][00363] Avg episode reward: [(0, '4.393')] +[2023-02-23 20:53:37,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3550.2, 300 sec: 2894.5). Total num frames: 217088. Throughput: 0: 875.9. Samples: 55354. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:53:37,819][00363] Avg episode reward: [(0, '4.374')] +[2023-02-23 20:53:42,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 2969.6). Total num frames: 237568. Throughput: 0: 893.3. Samples: 58060. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:53:42,817][00363] Avg episode reward: [(0, '4.321')] +[2023-02-23 20:53:44,063][12023] Updated weights for policy 0, policy_version 60 (0.0017) +[2023-02-23 20:53:47,815][00363] Fps is (10 sec: 4505.6, 60 sec: 3550.1, 300 sec: 3084.0). Total num frames: 262144. Throughput: 0: 929.9. Samples: 64862. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:53:47,824][00363] Avg episode reward: [(0, '4.349')] +[2023-02-23 20:53:52,817][00363] Fps is (10 sec: 4095.2, 60 sec: 3618.2, 300 sec: 3094.7). Total num frames: 278528. Throughput: 0: 897.3. Samples: 70184. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:53:52,819][00363] Avg episode reward: [(0, '4.521')] +[2023-02-23 20:53:52,834][12009] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000068_278528.pth... +[2023-02-23 20:53:55,481][12023] Updated weights for policy 0, policy_version 70 (0.0022) +[2023-02-23 20:53:57,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3618.2, 300 sec: 3061.2). Total num frames: 290816. Throughput: 0: 892.5. Samples: 72272. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:53:57,818][00363] Avg episode reward: [(0, '4.504')] +[2023-02-23 20:54:02,815][00363] Fps is (10 sec: 3277.5, 60 sec: 3618.2, 300 sec: 3113.0). Total num frames: 311296. Throughput: 0: 908.4. Samples: 77306. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:54:02,825][00363] Avg episode reward: [(0, '4.408')] +[2023-02-23 20:54:06,135][12023] Updated weights for policy 0, policy_version 80 (0.0019) +[2023-02-23 20:54:07,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3618.8, 300 sec: 3159.8). Total num frames: 331776. Throughput: 0: 927.9. Samples: 84028. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:54:07,823][00363] Avg episode reward: [(0, '4.449')] +[2023-02-23 20:54:12,819][00363] Fps is (10 sec: 4094.4, 60 sec: 3686.2, 300 sec: 3202.2). Total num frames: 352256. Throughput: 0: 921.4. Samples: 87044. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:54:12,822][00363] Avg episode reward: [(0, '4.458')] +[2023-02-23 20:54:17,819][00363] Fps is (10 sec: 3275.6, 60 sec: 3617.9, 300 sec: 3169.8). Total num frames: 364544. Throughput: 0: 891.8. Samples: 91194. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 20:54:17,822][00363] Avg episode reward: [(0, '4.467')] +[2023-02-23 20:54:18,785][12023] Updated weights for policy 0, policy_version 90 (0.0037) +[2023-02-23 20:54:22,815][00363] Fps is (10 sec: 2868.3, 60 sec: 3550.0, 300 sec: 3174.4). Total num frames: 380928. Throughput: 0: 906.5. Samples: 96148. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:54:22,821][00363] Avg episode reward: [(0, '4.376')] +[2023-02-23 20:54:27,815][00363] Fps is (10 sec: 3687.8, 60 sec: 3549.9, 300 sec: 3211.3). Total num frames: 401408. Throughput: 0: 918.8. Samples: 99408. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:54:27,825][00363] Avg episode reward: [(0, '4.672')] +[2023-02-23 20:54:27,828][12009] Saving new best policy, reward=4.672! +[2023-02-23 20:54:28,958][12023] Updated weights for policy 0, policy_version 100 (0.0015) +[2023-02-23 20:54:32,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3618.1, 300 sec: 3245.3). Total num frames: 421888. Throughput: 0: 897.8. Samples: 105264. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 20:54:32,820][00363] Avg episode reward: [(0, '4.712')] +[2023-02-23 20:54:32,832][12009] Saving new best policy, reward=4.712! +[2023-02-23 20:54:37,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3618.1, 300 sec: 3216.1). Total num frames: 434176. Throughput: 0: 869.2. Samples: 109298. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 20:54:37,822][00363] Avg episode reward: [(0, '4.502')] +[2023-02-23 20:54:42,251][12023] Updated weights for policy 0, policy_version 110 (0.0032) +[2023-02-23 20:54:42,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3549.9, 300 sec: 3218.3). Total num frames: 450560. Throughput: 0: 869.3. Samples: 111390. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:54:42,826][00363] Avg episode reward: [(0, '4.465')] +[2023-02-23 20:54:47,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3549.9, 300 sec: 3276.8). Total num frames: 475136. Throughput: 0: 904.0. Samples: 117984. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:54:47,818][00363] Avg episode reward: [(0, '4.331')] +[2023-02-23 20:54:51,597][12023] Updated weights for policy 0, policy_version 120 (0.0012) +[2023-02-23 20:54:52,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3550.0, 300 sec: 3276.8). Total num frames: 491520. Throughput: 0: 891.1. Samples: 124126. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:54:52,819][00363] Avg episode reward: [(0, '4.595')] +[2023-02-23 20:54:57,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3618.1, 300 sec: 3276.8). Total num frames: 507904. Throughput: 0: 871.3. Samples: 126250. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:54:57,821][00363] Avg episode reward: [(0, '4.641')] +[2023-02-23 20:55:02,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3276.8). Total num frames: 524288. Throughput: 0: 873.5. Samples: 130500. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) +[2023-02-23 20:55:02,822][00363] Avg episode reward: [(0, '4.547')] +[2023-02-23 20:55:04,302][12023] Updated weights for policy 0, policy_version 130 (0.0022) +[2023-02-23 20:55:07,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3301.6). Total num frames: 544768. Throughput: 0: 910.0. Samples: 137096. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 20:55:07,819][00363] Avg episode reward: [(0, '4.512')] +[2023-02-23 20:55:12,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3550.1, 300 sec: 3325.0). Total num frames: 565248. Throughput: 0: 910.3. Samples: 140370. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:55:12,818][00363] Avg episode reward: [(0, '4.434')] +[2023-02-23 20:55:15,019][12023] Updated weights for policy 0, policy_version 140 (0.0031) +[2023-02-23 20:55:17,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3618.4, 300 sec: 3323.6). Total num frames: 581632. Throughput: 0: 883.9. Samples: 145038. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 20:55:17,818][00363] Avg episode reward: [(0, '4.478')] +[2023-02-23 20:55:22,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3549.9, 300 sec: 3299.6). Total num frames: 593920. Throughput: 0: 894.0. Samples: 149526. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:55:22,821][00363] Avg episode reward: [(0, '4.564')] +[2023-02-23 20:55:26,735][12023] Updated weights for policy 0, policy_version 150 (0.0017) +[2023-02-23 20:55:27,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3618.1, 300 sec: 3343.2). Total num frames: 618496. Throughput: 0: 921.9. Samples: 152874. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 20:55:27,823][00363] Avg episode reward: [(0, '4.660')] +[2023-02-23 20:55:32,815][00363] Fps is (10 sec: 4505.6, 60 sec: 3618.1, 300 sec: 3363.0). Total num frames: 638976. Throughput: 0: 921.4. Samples: 159448. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) +[2023-02-23 20:55:32,818][00363] Avg episode reward: [(0, '4.633')] +[2023-02-23 20:55:37,822][00363] Fps is (10 sec: 3274.6, 60 sec: 3617.7, 300 sec: 3339.7). Total num frames: 651264. Throughput: 0: 882.8. Samples: 163858. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:55:37,831][00363] Avg episode reward: [(0, '4.706')] +[2023-02-23 20:55:38,012][12023] Updated weights for policy 0, policy_version 160 (0.0012) +[2023-02-23 20:55:42,815][00363] Fps is (10 sec: 2457.6, 60 sec: 3549.9, 300 sec: 3317.8). Total num frames: 663552. Throughput: 0: 880.6. Samples: 165878. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0) +[2023-02-23 20:55:42,824][00363] Avg episode reward: [(0, '4.641')] +[2023-02-23 20:55:47,815][00363] Fps is (10 sec: 2869.1, 60 sec: 3413.3, 300 sec: 3316.8). Total num frames: 679936. Throughput: 0: 871.4. Samples: 169714. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 20:55:47,820][00363] Avg episode reward: [(0, '4.729')] +[2023-02-23 20:55:47,824][12009] Saving new best policy, reward=4.729! +[2023-02-23 20:55:52,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3345.1, 300 sec: 3296.3). Total num frames: 692224. Throughput: 0: 817.5. Samples: 173884. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:55:52,817][00363] Avg episode reward: [(0, '4.810')] +[2023-02-23 20:55:52,831][12009] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000169_692224.pth... +[2023-02-23 20:55:52,952][12009] Saving new best policy, reward=4.810! +[2023-02-23 20:55:53,539][12023] Updated weights for policy 0, policy_version 170 (0.0026) +[2023-02-23 20:55:57,819][00363] Fps is (10 sec: 2456.7, 60 sec: 3276.6, 300 sec: 3276.7). Total num frames: 704512. Throughput: 0: 792.2. Samples: 176024. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:55:57,824][00363] Avg episode reward: [(0, '4.744')] +[2023-02-23 20:56:02,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3276.8, 300 sec: 3276.8). Total num frames: 720896. Throughput: 0: 778.3. Samples: 180060. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:56:02,819][00363] Avg episode reward: [(0, '4.664')] +[2023-02-23 20:56:06,812][12023] Updated weights for policy 0, policy_version 180 (0.0015) +[2023-02-23 20:56:07,815][00363] Fps is (10 sec: 3687.8, 60 sec: 3276.8, 300 sec: 3295.0). Total num frames: 741376. Throughput: 0: 803.6. Samples: 185686. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:56:07,818][00363] Avg episode reward: [(0, '4.629')] +[2023-02-23 20:56:12,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3276.8, 300 sec: 3312.4). Total num frames: 761856. Throughput: 0: 800.2. Samples: 188884. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 20:56:12,822][00363] Avg episode reward: [(0, '4.890')] +[2023-02-23 20:56:12,831][12009] Saving new best policy, reward=4.890! +[2023-02-23 20:56:17,565][12023] Updated weights for policy 0, policy_version 190 (0.0019) +[2023-02-23 20:56:17,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3276.8, 300 sec: 3311.7). Total num frames: 778240. Throughput: 0: 771.2. Samples: 194152. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 20:56:17,818][00363] Avg episode reward: [(0, '4.987')] +[2023-02-23 20:56:17,820][12009] Saving new best policy, reward=4.987! +[2023-02-23 20:56:22,818][00363] Fps is (10 sec: 2866.4, 60 sec: 3276.6, 300 sec: 3293.8). Total num frames: 790528. Throughput: 0: 765.8. Samples: 198314. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:56:22,826][00363] Avg episode reward: [(0, '5.575')] +[2023-02-23 20:56:22,836][12009] Saving new best policy, reward=5.575! +[2023-02-23 20:56:27,816][00363] Fps is (10 sec: 3276.7, 60 sec: 3208.5, 300 sec: 3310.2). Total num frames: 811008. Throughput: 0: 779.5. Samples: 200954. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:56:27,818][00363] Avg episode reward: [(0, '5.413')] +[2023-02-23 20:56:29,429][12023] Updated weights for policy 0, policy_version 200 (0.0018) +[2023-02-23 20:56:32,815][00363] Fps is (10 sec: 4097.2, 60 sec: 3208.5, 300 sec: 3326.0). Total num frames: 831488. Throughput: 0: 844.6. Samples: 207722. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 20:56:32,818][00363] Avg episode reward: [(0, '5.442')] +[2023-02-23 20:56:37,815][00363] Fps is (10 sec: 3686.5, 60 sec: 3277.2, 300 sec: 3325.0). Total num frames: 847872. Throughput: 0: 870.3. Samples: 213048. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 20:56:37,818][00363] Avg episode reward: [(0, '5.670')] +[2023-02-23 20:56:37,821][12009] Saving new best policy, reward=5.670! +[2023-02-23 20:56:41,354][12023] Updated weights for policy 0, policy_version 210 (0.0023) +[2023-02-23 20:56:42,816][00363] Fps is (10 sec: 2867.1, 60 sec: 3276.8, 300 sec: 3308.3). Total num frames: 860160. Throughput: 0: 868.9. Samples: 215120. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 20:56:42,819][00363] Avg episode reward: [(0, '5.954')] +[2023-02-23 20:56:42,890][12009] Saving new best policy, reward=5.954! +[2023-02-23 20:56:47,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3345.1, 300 sec: 3323.2). Total num frames: 880640. Throughput: 0: 890.9. Samples: 220150. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 20:56:47,818][00363] Avg episode reward: [(0, '5.811')] +[2023-02-23 20:56:51,632][12023] Updated weights for policy 0, policy_version 220 (0.0029) +[2023-02-23 20:56:52,815][00363] Fps is (10 sec: 4505.8, 60 sec: 3549.9, 300 sec: 3352.7). Total num frames: 905216. Throughput: 0: 912.8. Samples: 226760. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:56:52,819][00363] Avg episode reward: [(0, '6.194')] +[2023-02-23 20:56:52,832][12009] Saving new best policy, reward=6.194! +[2023-02-23 20:56:57,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3618.4, 300 sec: 3351.3). Total num frames: 921600. Throughput: 0: 906.9. Samples: 229696. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 20:56:57,819][00363] Avg episode reward: [(0, '6.458')] +[2023-02-23 20:56:57,824][12009] Saving new best policy, reward=6.458! +[2023-02-23 20:57:02,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3549.9, 300 sec: 3335.3). Total num frames: 933888. Throughput: 0: 881.6. Samples: 233824. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 20:57:02,823][00363] Avg episode reward: [(0, '6.619')] +[2023-02-23 20:57:02,831][12009] Saving new best policy, reward=6.619! +[2023-02-23 20:57:04,604][12023] Updated weights for policy 0, policy_version 230 (0.0022) +[2023-02-23 20:57:07,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3348.7). Total num frames: 954368. Throughput: 0: 907.7. Samples: 239156. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:57:07,817][00363] Avg episode reward: [(0, '6.178')] +[2023-02-23 20:57:12,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3549.9, 300 sec: 3361.5). Total num frames: 974848. Throughput: 0: 923.3. Samples: 242502. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:57:12,819][00363] Avg episode reward: [(0, '5.889')] +[2023-02-23 20:57:14,090][12023] Updated weights for policy 0, policy_version 240 (0.0018) +[2023-02-23 20:57:17,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3360.1). Total num frames: 991232. Throughput: 0: 907.9. Samples: 248578. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) +[2023-02-23 20:57:17,818][00363] Avg episode reward: [(0, '6.257')] +[2023-02-23 20:57:22,821][00363] Fps is (10 sec: 3274.9, 60 sec: 3618.0, 300 sec: 3415.6). Total num frames: 1007616. Throughput: 0: 884.0. Samples: 252832. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:57:22,823][00363] Avg episode reward: [(0, '6.522')] +[2023-02-23 20:57:26,996][12023] Updated weights for policy 0, policy_version 250 (0.0026) +[2023-02-23 20:57:27,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3471.2). Total num frames: 1024000. Throughput: 0: 887.2. Samples: 255044. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) +[2023-02-23 20:57:27,824][00363] Avg episode reward: [(0, '6.558')] +[2023-02-23 20:57:32,815][00363] Fps is (10 sec: 4098.4, 60 sec: 3618.1, 300 sec: 3540.7). Total num frames: 1048576. Throughput: 0: 926.2. Samples: 261830. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) +[2023-02-23 20:57:32,825][00363] Avg episode reward: [(0, '6.900')] +[2023-02-23 20:57:32,839][12009] Saving new best policy, reward=6.900! +[2023-02-23 20:57:36,921][12023] Updated weights for policy 0, policy_version 260 (0.0026) +[2023-02-23 20:57:37,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3618.1, 300 sec: 3526.7). Total num frames: 1064960. Throughput: 0: 903.2. Samples: 267402. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 20:57:37,820][00363] Avg episode reward: [(0, '7.478')] +[2023-02-23 20:57:37,826][12009] Saving new best policy, reward=7.478! +[2023-02-23 20:57:42,816][00363] Fps is (10 sec: 3276.7, 60 sec: 3686.4, 300 sec: 3499.0). Total num frames: 1081344. Throughput: 0: 883.9. Samples: 269474. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) +[2023-02-23 20:57:42,818][00363] Avg episode reward: [(0, '8.033')] +[2023-02-23 20:57:42,832][12009] Saving new best policy, reward=8.033! +[2023-02-23 20:57:47,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3618.1, 300 sec: 3512.9). Total num frames: 1097728. Throughput: 0: 890.9. Samples: 273916. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:57:47,826][00363] Avg episode reward: [(0, '7.817')] +[2023-02-23 20:57:49,216][12023] Updated weights for policy 0, policy_version 270 (0.0021) +[2023-02-23 20:57:52,816][00363] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3540.6). Total num frames: 1118208. Throughput: 0: 922.4. Samples: 280664. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 20:57:52,823][00363] Avg episode reward: [(0, '8.000')] +[2023-02-23 20:57:52,836][12009] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000273_1118208.pth... +[2023-02-23 20:57:52,955][12009] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000068_278528.pth +[2023-02-23 20:57:57,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3618.1, 300 sec: 3540.6). Total num frames: 1138688. Throughput: 0: 921.3. Samples: 283960. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 20:57:57,825][00363] Avg episode reward: [(0, '8.352')] +[2023-02-23 20:57:57,828][12009] Saving new best policy, reward=8.352! +[2023-02-23 20:58:00,095][12023] Updated weights for policy 0, policy_version 280 (0.0021) +[2023-02-23 20:58:02,815][00363] Fps is (10 sec: 3276.9, 60 sec: 3618.1, 300 sec: 3513.0). Total num frames: 1150976. Throughput: 0: 879.2. Samples: 288144. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 20:58:02,820][00363] Avg episode reward: [(0, '8.589')] +[2023-02-23 20:58:02,836][12009] Saving new best policy, reward=8.589! +[2023-02-23 20:58:07,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3549.9, 300 sec: 3512.8). Total num frames: 1167360. Throughput: 0: 887.8. Samples: 292780. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:58:07,817][00363] Avg episode reward: [(0, '8.426')] +[2023-02-23 20:58:11,806][12023] Updated weights for policy 0, policy_version 290 (0.0033) +[2023-02-23 20:58:12,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3618.1, 300 sec: 3540.6). Total num frames: 1191936. Throughput: 0: 912.9. Samples: 296124. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:58:12,822][00363] Avg episode reward: [(0, '8.164')] +[2023-02-23 20:58:17,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3618.1, 300 sec: 3526.7). Total num frames: 1208320. Throughput: 0: 907.6. Samples: 302674. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:58:17,822][00363] Avg episode reward: [(0, '8.164')] +[2023-02-23 20:58:22,819][00363] Fps is (10 sec: 3275.6, 60 sec: 3618.3, 300 sec: 3512.8). Total num frames: 1224704. Throughput: 0: 876.5. Samples: 306850. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:58:22,824][00363] Avg episode reward: [(0, '8.002')] +[2023-02-23 20:58:23,852][12023] Updated weights for policy 0, policy_version 300 (0.0020) +[2023-02-23 20:58:27,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3618.1, 300 sec: 3512.8). Total num frames: 1241088. Throughput: 0: 877.8. Samples: 308974. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 20:58:27,817][00363] Avg episode reward: [(0, '8.155')] +[2023-02-23 20:58:32,815][00363] Fps is (10 sec: 3687.8, 60 sec: 3549.9, 300 sec: 3540.6). Total num frames: 1261568. Throughput: 0: 919.8. Samples: 315308. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 20:58:32,817][00363] Avg episode reward: [(0, '8.331')] +[2023-02-23 20:58:33,987][12023] Updated weights for policy 0, policy_version 310 (0.0017) +[2023-02-23 20:58:37,820][00363] Fps is (10 sec: 4094.1, 60 sec: 3617.8, 300 sec: 3540.6). Total num frames: 1282048. Throughput: 0: 913.0. Samples: 321752. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:58:37,823][00363] Avg episode reward: [(0, '8.862')] +[2023-02-23 20:58:37,831][12009] Saving new best policy, reward=8.862! +[2023-02-23 20:58:42,816][00363] Fps is (10 sec: 3686.3, 60 sec: 3618.1, 300 sec: 3512.8). Total num frames: 1298432. Throughput: 0: 885.2. Samples: 323796. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 20:58:42,824][00363] Avg episode reward: [(0, '8.964')] +[2023-02-23 20:58:42,844][12009] Saving new best policy, reward=8.964! +[2023-02-23 20:58:47,046][12023] Updated weights for policy 0, policy_version 320 (0.0012) +[2023-02-23 20:58:47,815][00363] Fps is (10 sec: 2868.6, 60 sec: 3549.9, 300 sec: 3499.0). Total num frames: 1310720. Throughput: 0: 884.4. Samples: 327940. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:58:47,818][00363] Avg episode reward: [(0, '9.256')] +[2023-02-23 20:58:47,823][12009] Saving new best policy, reward=9.256! +[2023-02-23 20:58:52,815][00363] Fps is (10 sec: 3686.5, 60 sec: 3618.1, 300 sec: 3540.6). Total num frames: 1335296. Throughput: 0: 920.0. Samples: 334180. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 20:58:52,818][00363] Avg episode reward: [(0, '9.663')] +[2023-02-23 20:58:52,834][12009] Saving new best policy, reward=9.663! +[2023-02-23 20:58:56,410][12023] Updated weights for policy 0, policy_version 330 (0.0027) +[2023-02-23 20:58:57,815][00363] Fps is (10 sec: 4505.6, 60 sec: 3618.1, 300 sec: 3540.6). Total num frames: 1355776. Throughput: 0: 919.0. Samples: 337480. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 20:58:57,823][00363] Avg episode reward: [(0, '9.561')] +[2023-02-23 20:59:02,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3618.1, 300 sec: 3512.8). Total num frames: 1368064. Throughput: 0: 884.8. Samples: 342490. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:59:02,821][00363] Avg episode reward: [(0, '9.981')] +[2023-02-23 20:59:02,837][12009] Saving new best policy, reward=9.981! +[2023-02-23 20:59:07,816][00363] Fps is (10 sec: 2867.0, 60 sec: 3618.1, 300 sec: 3499.0). Total num frames: 1384448. Throughput: 0: 883.4. Samples: 346602. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 20:59:07,820][00363] Avg episode reward: [(0, '11.324')] +[2023-02-23 20:59:07,823][12009] Saving new best policy, reward=11.324! +[2023-02-23 20:59:09,683][12023] Updated weights for policy 0, policy_version 340 (0.0020) +[2023-02-23 20:59:12,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3526.8). Total num frames: 1404928. Throughput: 0: 906.9. Samples: 349784. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:59:12,821][00363] Avg episode reward: [(0, '11.446')] +[2023-02-23 20:59:12,832][12009] Saving new best policy, reward=11.446! +[2023-02-23 20:59:17,815][00363] Fps is (10 sec: 4096.3, 60 sec: 3618.1, 300 sec: 3540.6). Total num frames: 1425408. Throughput: 0: 911.5. Samples: 356326. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:59:17,818][00363] Avg episode reward: [(0, '12.221')] +[2023-02-23 20:59:17,821][12009] Saving new best policy, reward=12.221! +[2023-02-23 20:59:19,685][12023] Updated weights for policy 0, policy_version 350 (0.0015) +[2023-02-23 20:59:22,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3618.4, 300 sec: 3526.7). Total num frames: 1441792. Throughput: 0: 873.2. Samples: 361042. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:59:22,821][00363] Avg episode reward: [(0, '12.428')] +[2023-02-23 20:59:22,832][12009] Saving new best policy, reward=12.428! +[2023-02-23 20:59:27,816][00363] Fps is (10 sec: 2867.0, 60 sec: 3549.8, 300 sec: 3498.9). Total num frames: 1454080. Throughput: 0: 873.2. Samples: 363092. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 20:59:27,818][00363] Avg episode reward: [(0, '12.207')] +[2023-02-23 20:59:32,166][12023] Updated weights for policy 0, policy_version 360 (0.0023) +[2023-02-23 20:59:32,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3526.7). Total num frames: 1474560. Throughput: 0: 901.7. Samples: 368518. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:59:32,817][00363] Avg episode reward: [(0, '13.501')] +[2023-02-23 20:59:32,831][12009] Saving new best policy, reward=13.501! +[2023-02-23 20:59:37,815][00363] Fps is (10 sec: 4506.0, 60 sec: 3618.4, 300 sec: 3554.5). Total num frames: 1499136. Throughput: 0: 910.6. Samples: 375158. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:59:37,819][00363] Avg episode reward: [(0, '13.218')] +[2023-02-23 20:59:42,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3512.8). Total num frames: 1511424. Throughput: 0: 891.8. Samples: 377612. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:59:42,820][00363] Avg episode reward: [(0, '14.271')] +[2023-02-23 20:59:42,840][12009] Saving new best policy, reward=14.271! +[2023-02-23 20:59:43,134][12023] Updated weights for policy 0, policy_version 370 (0.0018) +[2023-02-23 20:59:47,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3618.1, 300 sec: 3512.8). Total num frames: 1527808. Throughput: 0: 872.8. Samples: 381764. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:59:47,822][00363] Avg episode reward: [(0, '14.790')] +[2023-02-23 20:59:47,826][12009] Saving new best policy, reward=14.790! +[2023-02-23 20:59:52,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3526.7). Total num frames: 1548288. Throughput: 0: 904.6. Samples: 387308. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 20:59:52,818][00363] Avg episode reward: [(0, '15.559')] +[2023-02-23 20:59:52,836][12009] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000378_1548288.pth... +[2023-02-23 20:59:52,968][12009] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000169_692224.pth +[2023-02-23 20:59:52,976][12009] Saving new best policy, reward=15.559! +[2023-02-23 20:59:54,803][12023] Updated weights for policy 0, policy_version 380 (0.0016) +[2023-02-23 20:59:57,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3549.9, 300 sec: 3540.6). Total num frames: 1568768. Throughput: 0: 905.9. Samples: 390550. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 20:59:57,824][00363] Avg episode reward: [(0, '15.601')] +[2023-02-23 20:59:57,827][12009] Saving new best policy, reward=15.601! +[2023-02-23 21:00:02,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3618.1, 300 sec: 3526.7). Total num frames: 1585152. Throughput: 0: 887.1. Samples: 396244. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:00:02,822][00363] Avg episode reward: [(0, '15.205')] +[2023-02-23 21:00:06,780][12023] Updated weights for policy 0, policy_version 390 (0.0038) +[2023-02-23 21:00:07,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3549.9, 300 sec: 3499.0). Total num frames: 1597440. Throughput: 0: 873.9. Samples: 400368. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:00:07,822][00363] Avg episode reward: [(0, '14.461')] +[2023-02-23 21:00:12,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3512.8). Total num frames: 1617920. Throughput: 0: 883.0. Samples: 402828. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 21:00:12,825][00363] Avg episode reward: [(0, '13.332')] +[2023-02-23 21:00:17,082][12023] Updated weights for policy 0, policy_version 400 (0.0017) +[2023-02-23 21:00:17,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3549.9, 300 sec: 3540.6). Total num frames: 1638400. Throughput: 0: 910.0. Samples: 409466. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:00:17,817][00363] Avg episode reward: [(0, '13.522')] +[2023-02-23 21:00:22,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3512.8). Total num frames: 1654784. Throughput: 0: 880.1. Samples: 414764. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:00:22,818][00363] Avg episode reward: [(0, '13.281')] +[2023-02-23 21:00:27,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3549.9, 300 sec: 3485.1). Total num frames: 1667072. Throughput: 0: 861.5. Samples: 416378. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:00:27,820][00363] Avg episode reward: [(0, '13.028')] +[2023-02-23 21:00:32,658][12023] Updated weights for policy 0, policy_version 410 (0.0015) +[2023-02-23 21:00:32,815][00363] Fps is (10 sec: 2457.6, 60 sec: 3413.3, 300 sec: 3485.2). Total num frames: 1679360. Throughput: 0: 843.2. Samples: 419708. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 21:00:32,822][00363] Avg episode reward: [(0, '12.657')] +[2023-02-23 21:00:37,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3276.8, 300 sec: 3499.0). Total num frames: 1695744. Throughput: 0: 823.2. Samples: 424350. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:00:37,822][00363] Avg episode reward: [(0, '13.529')] +[2023-02-23 21:00:42,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3413.3, 300 sec: 3512.8). Total num frames: 1716224. Throughput: 0: 822.5. Samples: 427564. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:00:42,818][00363] Avg episode reward: [(0, '13.719')] +[2023-02-23 21:00:43,255][12023] Updated weights for policy 0, policy_version 420 (0.0017) +[2023-02-23 21:00:47,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3413.3, 300 sec: 3526.7). Total num frames: 1732608. Throughput: 0: 832.5. Samples: 433706. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:00:47,817][00363] Avg episode reward: [(0, '15.855')] +[2023-02-23 21:00:47,831][12009] Saving new best policy, reward=15.855! +[2023-02-23 21:00:52,820][00363] Fps is (10 sec: 3275.3, 60 sec: 3344.8, 300 sec: 3540.6). Total num frames: 1748992. Throughput: 0: 830.8. Samples: 437758. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:00:52,824][00363] Avg episode reward: [(0, '16.258')] +[2023-02-23 21:00:52,848][12009] Saving new best policy, reward=16.258! +[2023-02-23 21:00:56,736][12023] Updated weights for policy 0, policy_version 430 (0.0026) +[2023-02-23 21:00:57,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3276.8, 300 sec: 3540.6). Total num frames: 1765376. Throughput: 0: 820.8. Samples: 439764. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:00:57,818][00363] Avg episode reward: [(0, '16.142')] +[2023-02-23 21:01:02,815][00363] Fps is (10 sec: 3688.1, 60 sec: 3345.1, 300 sec: 3540.6). Total num frames: 1785856. Throughput: 0: 812.7. Samples: 446036. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:01:02,818][00363] Avg episode reward: [(0, '15.398')] +[2023-02-23 21:01:06,370][12023] Updated weights for policy 0, policy_version 440 (0.0016) +[2023-02-23 21:01:07,816][00363] Fps is (10 sec: 3686.3, 60 sec: 3413.3, 300 sec: 3526.7). Total num frames: 1802240. Throughput: 0: 825.3. Samples: 451904. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:01:07,819][00363] Avg episode reward: [(0, '14.919')] +[2023-02-23 21:01:12,816][00363] Fps is (10 sec: 3276.7, 60 sec: 3345.0, 300 sec: 3526.7). Total num frames: 1818624. Throughput: 0: 834.9. Samples: 453950. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 21:01:12,823][00363] Avg episode reward: [(0, '14.957')] +[2023-02-23 21:01:17,815][00363] Fps is (10 sec: 3276.9, 60 sec: 3276.8, 300 sec: 3540.6). Total num frames: 1835008. Throughput: 0: 856.3. Samples: 458240. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 21:01:17,822][00363] Avg episode reward: [(0, '16.141')] +[2023-02-23 21:01:19,087][12023] Updated weights for policy 0, policy_version 450 (0.0033) +[2023-02-23 21:01:22,815][00363] Fps is (10 sec: 3686.5, 60 sec: 3345.1, 300 sec: 3540.6). Total num frames: 1855488. Throughput: 0: 901.7. Samples: 464928. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 21:01:22,822][00363] Avg episode reward: [(0, '16.709')] +[2023-02-23 21:01:22,835][12009] Saving new best policy, reward=16.709! +[2023-02-23 21:01:27,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3481.6, 300 sec: 3540.6). Total num frames: 1875968. Throughput: 0: 903.2. Samples: 468210. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 21:01:27,818][00363] Avg episode reward: [(0, '18.481')] +[2023-02-23 21:01:27,820][12009] Saving new best policy, reward=18.481! +[2023-02-23 21:01:30,011][12023] Updated weights for policy 0, policy_version 460 (0.0012) +[2023-02-23 21:01:32,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3540.6). Total num frames: 1892352. Throughput: 0: 867.6. Samples: 472750. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 21:01:32,822][00363] Avg episode reward: [(0, '19.018')] +[2023-02-23 21:01:32,838][12009] Saving new best policy, reward=19.018! +[2023-02-23 21:01:37,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3481.6, 300 sec: 3540.6). Total num frames: 1904640. Throughput: 0: 877.7. Samples: 477250. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:01:37,817][00363] Avg episode reward: [(0, '21.044')] +[2023-02-23 21:01:37,828][12009] Saving new best policy, reward=21.044! +[2023-02-23 21:01:41,998][12023] Updated weights for policy 0, policy_version 470 (0.0015) +[2023-02-23 21:01:42,816][00363] Fps is (10 sec: 3276.7, 60 sec: 3481.6, 300 sec: 3540.6). Total num frames: 1925120. Throughput: 0: 902.2. Samples: 480362. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:01:42,825][00363] Avg episode reward: [(0, '21.303')] +[2023-02-23 21:01:42,933][12009] Saving new best policy, reward=21.303! +[2023-02-23 21:01:47,817][00363] Fps is (10 sec: 4095.4, 60 sec: 3549.8, 300 sec: 3526.7). Total num frames: 1945600. Throughput: 0: 906.6. Samples: 486834. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:01:47,824][00363] Avg episode reward: [(0, '20.406')] +[2023-02-23 21:01:52,819][00363] Fps is (10 sec: 3685.1, 60 sec: 3549.9, 300 sec: 3526.7). Total num frames: 1961984. Throughput: 0: 871.6. Samples: 491128. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:01:52,826][00363] Avg episode reward: [(0, '20.399')] +[2023-02-23 21:01:52,845][12009] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000479_1961984.pth... +[2023-02-23 21:01:52,994][12009] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000273_1118208.pth +[2023-02-23 21:01:53,843][12023] Updated weights for policy 0, policy_version 480 (0.0022) +[2023-02-23 21:01:57,815][00363] Fps is (10 sec: 2867.6, 60 sec: 3481.6, 300 sec: 3526.7). Total num frames: 1974272. Throughput: 0: 870.8. Samples: 493136. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 21:01:57,818][00363] Avg episode reward: [(0, '20.018')] +[2023-02-23 21:02:02,816][00363] Fps is (10 sec: 3687.7, 60 sec: 3549.9, 300 sec: 3540.6). Total num frames: 1998848. Throughput: 0: 909.2. Samples: 499154. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:02:02,818][00363] Avg episode reward: [(0, '17.759')] +[2023-02-23 21:02:04,421][12023] Updated weights for policy 0, policy_version 490 (0.0015) +[2023-02-23 21:02:07,815][00363] Fps is (10 sec: 4505.6, 60 sec: 3618.2, 300 sec: 3540.6). Total num frames: 2019328. Throughput: 0: 909.1. Samples: 505838. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:02:07,818][00363] Avg episode reward: [(0, '17.673')] +[2023-02-23 21:02:12,815][00363] Fps is (10 sec: 3686.5, 60 sec: 3618.1, 300 sec: 3540.6). Total num frames: 2035712. Throughput: 0: 882.9. Samples: 507942. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:02:12,822][00363] Avg episode reward: [(0, '17.816')] +[2023-02-23 21:02:17,198][12023] Updated weights for policy 0, policy_version 500 (0.0020) +[2023-02-23 21:02:17,816][00363] Fps is (10 sec: 2867.0, 60 sec: 3549.8, 300 sec: 3526.8). Total num frames: 2048000. Throughput: 0: 876.1. Samples: 512174. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:02:17,824][00363] Avg episode reward: [(0, '17.659')] +[2023-02-23 21:02:22,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3540.6). Total num frames: 2068480. Throughput: 0: 907.3. Samples: 518080. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:02:22,823][00363] Avg episode reward: [(0, '17.208')] +[2023-02-23 21:02:26,791][12023] Updated weights for policy 0, policy_version 510 (0.0013) +[2023-02-23 21:02:27,815][00363] Fps is (10 sec: 4505.9, 60 sec: 3618.1, 300 sec: 3540.6). Total num frames: 2093056. Throughput: 0: 912.4. Samples: 521418. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:02:27,818][00363] Avg episode reward: [(0, '17.594')] +[2023-02-23 21:02:32,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3526.7). Total num frames: 2105344. Throughput: 0: 883.0. Samples: 526566. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:02:32,820][00363] Avg episode reward: [(0, '17.631')] +[2023-02-23 21:02:37,815][00363] Fps is (10 sec: 2457.6, 60 sec: 3549.9, 300 sec: 3512.8). Total num frames: 2117632. Throughput: 0: 875.4. Samples: 530516. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:02:37,819][00363] Avg episode reward: [(0, '18.515')] +[2023-02-23 21:02:40,386][12023] Updated weights for policy 0, policy_version 520 (0.0017) +[2023-02-23 21:02:42,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3526.7). Total num frames: 2138112. Throughput: 0: 892.1. Samples: 533282. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:02:42,822][00363] Avg episode reward: [(0, '18.422')] +[2023-02-23 21:02:47,817][00363] Fps is (10 sec: 4095.5, 60 sec: 3549.9, 300 sec: 3526.7). Total num frames: 2158592. Throughput: 0: 897.3. Samples: 539532. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:02:47,822][00363] Avg episode reward: [(0, '19.856')] +[2023-02-23 21:02:50,930][12023] Updated weights for policy 0, policy_version 530 (0.0018) +[2023-02-23 21:02:52,821][00363] Fps is (10 sec: 3684.3, 60 sec: 3549.8, 300 sec: 3512.8). Total num frames: 2174976. Throughput: 0: 852.6. Samples: 544208. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:02:52,832][00363] Avg episode reward: [(0, '19.042')] +[2023-02-23 21:02:57,816][00363] Fps is (10 sec: 2867.3, 60 sec: 3549.8, 300 sec: 3512.8). Total num frames: 2187264. Throughput: 0: 850.6. Samples: 546220. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:02:57,823][00363] Avg episode reward: [(0, '19.314')] +[2023-02-23 21:03:02,815][00363] Fps is (10 sec: 3278.6, 60 sec: 3481.6, 300 sec: 3526.7). Total num frames: 2207744. Throughput: 0: 873.0. Samples: 551460. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:03:02,817][00363] Avg episode reward: [(0, '18.617')] +[2023-02-23 21:03:03,419][12023] Updated weights for policy 0, policy_version 540 (0.0013) +[2023-02-23 21:03:07,815][00363] Fps is (10 sec: 4096.3, 60 sec: 3481.6, 300 sec: 3512.8). Total num frames: 2228224. Throughput: 0: 885.5. Samples: 557926. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:03:07,821][00363] Avg episode reward: [(0, '18.695')] +[2023-02-23 21:03:12,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3481.6, 300 sec: 3512.8). Total num frames: 2244608. Throughput: 0: 868.7. Samples: 560510. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:03:12,820][00363] Avg episode reward: [(0, '17.831')] +[2023-02-23 21:03:15,153][12023] Updated weights for policy 0, policy_version 550 (0.0032) +[2023-02-23 21:03:17,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3481.6, 300 sec: 3499.0). Total num frames: 2256896. Throughput: 0: 846.1. Samples: 564642. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:03:17,824][00363] Avg episode reward: [(0, '18.482')] +[2023-02-23 21:03:22,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3481.6, 300 sec: 3512.8). Total num frames: 2277376. Throughput: 0: 875.5. Samples: 569912. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:03:22,818][00363] Avg episode reward: [(0, '18.438')] +[2023-02-23 21:03:26,207][12023] Updated weights for policy 0, policy_version 560 (0.0026) +[2023-02-23 21:03:27,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3413.3, 300 sec: 3512.8). Total num frames: 2297856. Throughput: 0: 888.8. Samples: 573278. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:03:27,818][00363] Avg episode reward: [(0, '20.339')] +[2023-02-23 21:03:32,816][00363] Fps is (10 sec: 3686.2, 60 sec: 3481.6, 300 sec: 3499.0). Total num frames: 2314240. Throughput: 0: 877.3. Samples: 579010. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:03:32,819][00363] Avg episode reward: [(0, '21.736')] +[2023-02-23 21:03:32,840][12009] Saving new best policy, reward=21.736! +[2023-02-23 21:03:37,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3499.0). Total num frames: 2330624. Throughput: 0: 864.9. Samples: 583122. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:03:37,820][00363] Avg episode reward: [(0, '22.424')] +[2023-02-23 21:03:37,824][12009] Saving new best policy, reward=22.424! +[2023-02-23 21:03:39,293][12023] Updated weights for policy 0, policy_version 570 (0.0021) +[2023-02-23 21:03:42,815][00363] Fps is (10 sec: 3276.9, 60 sec: 3481.6, 300 sec: 3512.8). Total num frames: 2347008. Throughput: 0: 868.3. Samples: 585292. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 21:03:42,818][00363] Avg episode reward: [(0, '22.275')] +[2023-02-23 21:03:47,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3481.7, 300 sec: 3499.0). Total num frames: 2367488. Throughput: 0: 895.4. Samples: 591754. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:03:47,820][00363] Avg episode reward: [(0, '22.664')] +[2023-02-23 21:03:47,825][12009] Saving new best policy, reward=22.664! +[2023-02-23 21:03:49,140][12023] Updated weights for policy 0, policy_version 580 (0.0026) +[2023-02-23 21:03:52,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3481.9, 300 sec: 3485.1). Total num frames: 2383872. Throughput: 0: 876.5. Samples: 597370. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:03:52,821][00363] Avg episode reward: [(0, '22.869')] +[2023-02-23 21:03:52,832][12009] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000582_2383872.pth... +[2023-02-23 21:03:53,032][12009] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000378_1548288.pth +[2023-02-23 21:03:53,081][12009] Saving new best policy, reward=22.869! +[2023-02-23 21:03:57,816][00363] Fps is (10 sec: 3276.7, 60 sec: 3549.9, 300 sec: 3499.0). Total num frames: 2400256. Throughput: 0: 862.0. Samples: 599300. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:03:57,825][00363] Avg episode reward: [(0, '22.339')] +[2023-02-23 21:04:02,172][12023] Updated weights for policy 0, policy_version 590 (0.0017) +[2023-02-23 21:04:02,816][00363] Fps is (10 sec: 3276.8, 60 sec: 3481.6, 300 sec: 3499.0). Total num frames: 2416640. Throughput: 0: 873.9. Samples: 603966. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:04:02,818][00363] Avg episode reward: [(0, '21.208')] +[2023-02-23 21:04:07,815][00363] Fps is (10 sec: 4096.1, 60 sec: 3549.9, 300 sec: 3512.8). Total num frames: 2441216. Throughput: 0: 908.4. Samples: 610788. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:04:07,818][00363] Avg episode reward: [(0, '22.677')] +[2023-02-23 21:04:11,846][12023] Updated weights for policy 0, policy_version 600 (0.0020) +[2023-02-23 21:04:12,818][00363] Fps is (10 sec: 4094.9, 60 sec: 3549.7, 300 sec: 3498.9). Total num frames: 2457600. Throughput: 0: 908.4. Samples: 614158. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) +[2023-02-23 21:04:12,823][00363] Avg episode reward: [(0, '23.105')] +[2023-02-23 21:04:12,841][12009] Saving new best policy, reward=23.105! +[2023-02-23 21:04:17,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3618.1, 300 sec: 3499.0). Total num frames: 2473984. Throughput: 0: 873.8. Samples: 618330. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:04:17,823][00363] Avg episode reward: [(0, '22.581')] +[2023-02-23 21:04:22,815][00363] Fps is (10 sec: 3277.7, 60 sec: 3549.9, 300 sec: 3512.8). Total num frames: 2490368. Throughput: 0: 889.9. Samples: 623168. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:04:22,823][00363] Avg episode reward: [(0, '22.271')] +[2023-02-23 21:04:24,524][12023] Updated weights for policy 0, policy_version 610 (0.0056) +[2023-02-23 21:04:27,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3549.9, 300 sec: 3512.8). Total num frames: 2510848. Throughput: 0: 914.5. Samples: 626444. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:04:27,822][00363] Avg episode reward: [(0, '22.226')] +[2023-02-23 21:04:32,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3618.2, 300 sec: 3499.0). Total num frames: 2531328. Throughput: 0: 911.4. Samples: 632768. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:04:32,819][00363] Avg episode reward: [(0, '22.507')] +[2023-02-23 21:04:35,653][12023] Updated weights for policy 0, policy_version 620 (0.0015) +[2023-02-23 21:04:37,817][00363] Fps is (10 sec: 3276.3, 60 sec: 3549.8, 300 sec: 3498.9). Total num frames: 2543616. Throughput: 0: 875.8. Samples: 636784. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:04:37,820][00363] Avg episode reward: [(0, '22.034')] +[2023-02-23 21:04:42,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3549.9, 300 sec: 3499.0). Total num frames: 2560000. Throughput: 0: 877.1. Samples: 638770. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:04:42,824][00363] Avg episode reward: [(0, '21.780')] +[2023-02-23 21:04:47,244][12023] Updated weights for policy 0, policy_version 630 (0.0021) +[2023-02-23 21:04:47,815][00363] Fps is (10 sec: 3687.0, 60 sec: 3549.9, 300 sec: 3499.0). Total num frames: 2580480. Throughput: 0: 912.4. Samples: 645024. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:04:47,824][00363] Avg episode reward: [(0, '21.043')] +[2023-02-23 21:04:52,816][00363] Fps is (10 sec: 4096.0, 60 sec: 3618.1, 300 sec: 3499.0). Total num frames: 2600960. Throughput: 0: 899.2. Samples: 651252. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:04:52,820][00363] Avg episode reward: [(0, '22.405')] +[2023-02-23 21:04:57,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3485.1). Total num frames: 2613248. Throughput: 0: 870.1. Samples: 653312. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:04:57,822][00363] Avg episode reward: [(0, '22.644')] +[2023-02-23 21:04:59,611][12023] Updated weights for policy 0, policy_version 640 (0.0014) +[2023-02-23 21:05:02,820][00363] Fps is (10 sec: 2865.9, 60 sec: 3549.6, 300 sec: 3498.9). Total num frames: 2629632. Throughput: 0: 870.8. Samples: 657520. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:05:02,830][00363] Avg episode reward: [(0, '21.696')] +[2023-02-23 21:05:07,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3345.1, 300 sec: 3471.2). Total num frames: 2641920. Throughput: 0: 852.8. Samples: 661544. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0) +[2023-02-23 21:05:07,819][00363] Avg episode reward: [(0, '22.326')] +[2023-02-23 21:05:12,819][00363] Fps is (10 sec: 2867.5, 60 sec: 3345.0, 300 sec: 3457.3). Total num frames: 2658304. Throughput: 0: 825.4. Samples: 663588. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:05:12,822][00363] Avg episode reward: [(0, '22.026')] +[2023-02-23 21:05:14,235][12023] Updated weights for policy 0, policy_version 650 (0.0017) +[2023-02-23 21:05:17,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3276.8, 300 sec: 3443.4). Total num frames: 2670592. Throughput: 0: 773.9. Samples: 667594. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 21:05:17,817][00363] Avg episode reward: [(0, '22.082')] +[2023-02-23 21:05:22,815][00363] Fps is (10 sec: 2458.5, 60 sec: 3208.5, 300 sec: 3443.4). Total num frames: 2682880. Throughput: 0: 774.7. Samples: 671644. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 21:05:22,819][00363] Avg episode reward: [(0, '22.608')] +[2023-02-23 21:05:26,976][12023] Updated weights for policy 0, policy_version 660 (0.0012) +[2023-02-23 21:05:27,816][00363] Fps is (10 sec: 3276.5, 60 sec: 3208.5, 300 sec: 3471.2). Total num frames: 2703360. Throughput: 0: 801.7. Samples: 674848. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 21:05:27,821][00363] Avg episode reward: [(0, '21.305')] +[2023-02-23 21:05:32,815][00363] Fps is (10 sec: 4505.6, 60 sec: 3276.8, 300 sec: 3499.0). Total num frames: 2727936. Throughput: 0: 812.0. Samples: 681564. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 21:05:32,820][00363] Avg episode reward: [(0, '22.227')] +[2023-02-23 21:05:37,815][00363] Fps is (10 sec: 3686.7, 60 sec: 3276.9, 300 sec: 3471.2). Total num frames: 2740224. Throughput: 0: 770.8. Samples: 685938. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:05:37,821][00363] Avg episode reward: [(0, '20.908')] +[2023-02-23 21:05:38,483][12023] Updated weights for policy 0, policy_version 670 (0.0013) +[2023-02-23 21:05:42,815][00363] Fps is (10 sec: 2457.6, 60 sec: 3208.5, 300 sec: 3457.3). Total num frames: 2752512. Throughput: 0: 769.7. Samples: 687948. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:05:42,822][00363] Avg episode reward: [(0, '21.685')] +[2023-02-23 21:05:47,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3276.8, 300 sec: 3485.1). Total num frames: 2777088. Throughput: 0: 799.6. Samples: 693498. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:05:47,824][00363] Avg episode reward: [(0, '21.883')] +[2023-02-23 21:05:49,852][12023] Updated weights for policy 0, policy_version 680 (0.0031) +[2023-02-23 21:05:52,817][00363] Fps is (10 sec: 4504.9, 60 sec: 3276.7, 300 sec: 3498.9). Total num frames: 2797568. Throughput: 0: 852.6. Samples: 699914. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:05:52,820][00363] Avg episode reward: [(0, '21.693')] +[2023-02-23 21:05:52,828][12009] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000683_2797568.pth... +[2023-02-23 21:05:52,983][12009] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000479_1961984.pth +[2023-02-23 21:05:57,818][00363] Fps is (10 sec: 3275.9, 60 sec: 3276.7, 300 sec: 3471.2). Total num frames: 2809856. Throughput: 0: 855.5. Samples: 702084. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:05:57,821][00363] Avg episode reward: [(0, '21.848')] +[2023-02-23 21:06:02,816][00363] Fps is (10 sec: 2458.0, 60 sec: 3208.8, 300 sec: 3457.3). Total num frames: 2822144. Throughput: 0: 856.6. Samples: 706140. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:06:02,819][00363] Avg episode reward: [(0, '21.313')] +[2023-02-23 21:06:02,972][12023] Updated weights for policy 0, policy_version 690 (0.0017) +[2023-02-23 21:06:07,816][00363] Fps is (10 sec: 3277.6, 60 sec: 3345.0, 300 sec: 3471.2). Total num frames: 2842624. Throughput: 0: 889.9. Samples: 711690. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:06:07,823][00363] Avg episode reward: [(0, '21.567')] +[2023-02-23 21:06:12,816][00363] Fps is (10 sec: 4095.7, 60 sec: 3413.5, 300 sec: 3485.1). Total num frames: 2863104. Throughput: 0: 888.8. Samples: 714844. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:06:12,823][00363] Avg episode reward: [(0, '21.930')] +[2023-02-23 21:06:13,085][12023] Updated weights for policy 0, policy_version 700 (0.0013) +[2023-02-23 21:06:17,815][00363] Fps is (10 sec: 3686.5, 60 sec: 3481.6, 300 sec: 3471.2). Total num frames: 2879488. Throughput: 0: 853.8. Samples: 719984. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:06:17,823][00363] Avg episode reward: [(0, '21.843')] +[2023-02-23 21:06:22,816][00363] Fps is (10 sec: 2867.4, 60 sec: 3481.6, 300 sec: 3443.4). Total num frames: 2891776. Throughput: 0: 844.9. Samples: 723960. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:06:22,823][00363] Avg episode reward: [(0, '21.058')] +[2023-02-23 21:06:26,394][12023] Updated weights for policy 0, policy_version 710 (0.0027) +[2023-02-23 21:06:27,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3481.6, 300 sec: 3457.3). Total num frames: 2912256. Throughput: 0: 861.2. Samples: 726704. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 21:06:27,823][00363] Avg episode reward: [(0, '21.898')] +[2023-02-23 21:06:32,815][00363] Fps is (10 sec: 4096.1, 60 sec: 3413.3, 300 sec: 3485.1). Total num frames: 2932736. Throughput: 0: 885.6. Samples: 733352. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:06:32,820][00363] Avg episode reward: [(0, '21.922')] +[2023-02-23 21:06:36,721][12023] Updated weights for policy 0, policy_version 720 (0.0030) +[2023-02-23 21:06:37,816][00363] Fps is (10 sec: 3686.1, 60 sec: 3481.6, 300 sec: 3471.2). Total num frames: 2949120. Throughput: 0: 856.5. Samples: 738454. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:06:37,822][00363] Avg episode reward: [(0, '22.026')] +[2023-02-23 21:06:42,817][00363] Fps is (10 sec: 3276.2, 60 sec: 3549.8, 300 sec: 3457.3). Total num frames: 2965504. Throughput: 0: 854.5. Samples: 740534. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:06:42,822][00363] Avg episode reward: [(0, '22.009')] +[2023-02-23 21:06:47,815][00363] Fps is (10 sec: 3686.7, 60 sec: 3481.6, 300 sec: 3471.2). Total num frames: 2985984. Throughput: 0: 878.3. Samples: 745664. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:06:47,817][00363] Avg episode reward: [(0, '22.488')] +[2023-02-23 21:06:48,835][12023] Updated weights for policy 0, policy_version 730 (0.0013) +[2023-02-23 21:06:52,815][00363] Fps is (10 sec: 4096.8, 60 sec: 3481.7, 300 sec: 3499.0). Total num frames: 3006464. Throughput: 0: 900.1. Samples: 752192. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:06:52,818][00363] Avg episode reward: [(0, '21.720')] +[2023-02-23 21:06:57,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3550.0, 300 sec: 3471.2). Total num frames: 3022848. Throughput: 0: 893.7. Samples: 755058. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:06:57,819][00363] Avg episode reward: [(0, '20.712')] +[2023-02-23 21:07:00,158][12023] Updated weights for policy 0, policy_version 740 (0.0013) +[2023-02-23 21:07:02,816][00363] Fps is (10 sec: 2867.1, 60 sec: 3549.9, 300 sec: 3443.4). Total num frames: 3035136. Throughput: 0: 872.3. Samples: 759238. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:07:02,819][00363] Avg episode reward: [(0, '21.773')] +[2023-02-23 21:07:07,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3549.9, 300 sec: 3457.3). Total num frames: 3055616. Throughput: 0: 901.4. Samples: 764524. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 21:07:07,825][00363] Avg episode reward: [(0, '21.344')] +[2023-02-23 21:07:11,053][12023] Updated weights for policy 0, policy_version 750 (0.0013) +[2023-02-23 21:07:12,815][00363] Fps is (10 sec: 4096.1, 60 sec: 3549.9, 300 sec: 3485.1). Total num frames: 3076096. Throughput: 0: 915.9. Samples: 767920. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 21:07:12,823][00363] Avg episode reward: [(0, '22.061')] +[2023-02-23 21:07:17,823][00363] Fps is (10 sec: 4092.7, 60 sec: 3617.7, 300 sec: 3485.0). Total num frames: 3096576. Throughput: 0: 898.2. Samples: 773776. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:07:17,828][00363] Avg episode reward: [(0, '22.970')] +[2023-02-23 21:07:22,820][00363] Fps is (10 sec: 3275.2, 60 sec: 3617.9, 300 sec: 3443.4). Total num frames: 3108864. Throughput: 0: 877.5. Samples: 777946. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:07:22,823][00363] Avg episode reward: [(0, '22.584')] +[2023-02-23 21:07:23,727][12023] Updated weights for policy 0, policy_version 760 (0.0023) +[2023-02-23 21:07:27,815][00363] Fps is (10 sec: 2869.5, 60 sec: 3549.9, 300 sec: 3457.3). Total num frames: 3125248. Throughput: 0: 878.7. Samples: 780072. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:07:27,822][00363] Avg episode reward: [(0, '20.470')] +[2023-02-23 21:07:32,815][00363] Fps is (10 sec: 4098.0, 60 sec: 3618.1, 300 sec: 3499.0). Total num frames: 3149824. Throughput: 0: 910.6. Samples: 786640. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:07:32,822][00363] Avg episode reward: [(0, '20.601')] +[2023-02-23 21:07:33,739][12023] Updated weights for policy 0, policy_version 770 (0.0017) +[2023-02-23 21:07:37,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3618.2, 300 sec: 3485.1). Total num frames: 3166208. Throughput: 0: 890.6. Samples: 792270. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:07:37,823][00363] Avg episode reward: [(0, '21.124')] +[2023-02-23 21:07:42,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3550.0, 300 sec: 3457.3). Total num frames: 3178496. Throughput: 0: 870.2. Samples: 794218. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:07:42,820][00363] Avg episode reward: [(0, '20.695')] +[2023-02-23 21:07:47,627][12023] Updated weights for policy 0, policy_version 780 (0.0023) +[2023-02-23 21:07:47,816][00363] Fps is (10 sec: 2867.2, 60 sec: 3481.6, 300 sec: 3457.4). Total num frames: 3194880. Throughput: 0: 865.5. Samples: 798184. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:07:47,818][00363] Avg episode reward: [(0, '21.169')] +[2023-02-23 21:07:52,818][00363] Fps is (10 sec: 3685.5, 60 sec: 3481.5, 300 sec: 3485.1). Total num frames: 3215360. Throughput: 0: 888.0. Samples: 804488. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:07:52,820][00363] Avg episode reward: [(0, '21.860')] +[2023-02-23 21:07:52,836][12009] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000785_3215360.pth... +[2023-02-23 21:07:52,953][12009] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000582_2383872.pth +[2023-02-23 21:07:57,732][12023] Updated weights for policy 0, policy_version 790 (0.0013) +[2023-02-23 21:07:57,817][00363] Fps is (10 sec: 4095.3, 60 sec: 3549.8, 300 sec: 3485.1). Total num frames: 3235840. Throughput: 0: 883.9. Samples: 807698. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:07:57,820][00363] Avg episode reward: [(0, '22.931')] +[2023-02-23 21:08:02,817][00363] Fps is (10 sec: 3277.0, 60 sec: 3549.8, 300 sec: 3457.3). Total num frames: 3248128. Throughput: 0: 850.1. Samples: 812026. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:08:02,823][00363] Avg episode reward: [(0, '23.858')] +[2023-02-23 21:08:02,839][12009] Saving new best policy, reward=23.858! +[2023-02-23 21:08:07,815][00363] Fps is (10 sec: 2867.7, 60 sec: 3481.6, 300 sec: 3457.3). Total num frames: 3264512. Throughput: 0: 854.6. Samples: 816398. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:08:07,817][00363] Avg episode reward: [(0, '22.958')] +[2023-02-23 21:08:10,816][12023] Updated weights for policy 0, policy_version 800 (0.0041) +[2023-02-23 21:08:12,815][00363] Fps is (10 sec: 3687.1, 60 sec: 3481.6, 300 sec: 3485.1). Total num frames: 3284992. Throughput: 0: 877.5. Samples: 819560. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:08:12,818][00363] Avg episode reward: [(0, '22.664')] +[2023-02-23 21:08:17,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3482.1, 300 sec: 3485.1). Total num frames: 3305472. Throughput: 0: 872.2. Samples: 825888. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:08:17,818][00363] Avg episode reward: [(0, '21.384')] +[2023-02-23 21:08:22,402][12023] Updated weights for policy 0, policy_version 810 (0.0026) +[2023-02-23 21:08:22,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3481.9, 300 sec: 3457.3). Total num frames: 3317760. Throughput: 0: 840.8. Samples: 830104. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 21:08:22,825][00363] Avg episode reward: [(0, '21.394')] +[2023-02-23 21:08:27,815][00363] Fps is (10 sec: 2457.6, 60 sec: 3413.3, 300 sec: 3443.4). Total num frames: 3330048. Throughput: 0: 844.0. Samples: 832198. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:08:27,817][00363] Avg episode reward: [(0, '22.132')] +[2023-02-23 21:08:32,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3413.3, 300 sec: 3471.2). Total num frames: 3354624. Throughput: 0: 884.7. Samples: 837994. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 21:08:32,818][00363] Avg episode reward: [(0, '21.478')] +[2023-02-23 21:08:33,738][12023] Updated weights for policy 0, policy_version 820 (0.0017) +[2023-02-23 21:08:37,815][00363] Fps is (10 sec: 4505.6, 60 sec: 3481.6, 300 sec: 3485.1). Total num frames: 3375104. Throughput: 0: 889.7. Samples: 844524. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 21:08:37,822][00363] Avg episode reward: [(0, '22.892')] +[2023-02-23 21:08:42,816][00363] Fps is (10 sec: 3276.7, 60 sec: 3481.6, 300 sec: 3457.3). Total num frames: 3387392. Throughput: 0: 865.9. Samples: 846662. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:08:42,818][00363] Avg episode reward: [(0, '23.063')] +[2023-02-23 21:08:46,185][12023] Updated weights for policy 0, policy_version 830 (0.0021) +[2023-02-23 21:08:47,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3481.6, 300 sec: 3457.3). Total num frames: 3403776. Throughput: 0: 859.3. Samples: 850692. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:08:47,822][00363] Avg episode reward: [(0, '23.779')] +[2023-02-23 21:08:52,815][00363] Fps is (10 sec: 3276.9, 60 sec: 3413.5, 300 sec: 3457.3). Total num frames: 3420160. Throughput: 0: 887.6. Samples: 856338. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:08:52,823][00363] Avg episode reward: [(0, '24.822')] +[2023-02-23 21:08:52,866][12009] Saving new best policy, reward=24.822! +[2023-02-23 21:08:56,823][12023] Updated weights for policy 0, policy_version 840 (0.0044) +[2023-02-23 21:08:57,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3481.7, 300 sec: 3485.1). Total num frames: 3444736. Throughput: 0: 886.9. Samples: 859472. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:08:57,824][00363] Avg episode reward: [(0, '24.325')] +[2023-02-23 21:09:02,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3481.7, 300 sec: 3443.4). Total num frames: 3457024. Throughput: 0: 861.7. Samples: 864664. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:09:02,818][00363] Avg episode reward: [(0, '24.860')] +[2023-02-23 21:09:02,835][12009] Saving new best policy, reward=24.860! +[2023-02-23 21:09:07,817][00363] Fps is (10 sec: 2457.2, 60 sec: 3413.2, 300 sec: 3429.5). Total num frames: 3469312. Throughput: 0: 860.4. Samples: 868824. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:09:07,819][00363] Avg episode reward: [(0, '24.293')] +[2023-02-23 21:09:10,052][12023] Updated weights for policy 0, policy_version 850 (0.0019) +[2023-02-23 21:09:12,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3481.6, 300 sec: 3457.3). Total num frames: 3493888. Throughput: 0: 876.4. Samples: 871636. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:09:12,825][00363] Avg episode reward: [(0, '24.844')] +[2023-02-23 21:09:17,816][00363] Fps is (10 sec: 4506.2, 60 sec: 3481.6, 300 sec: 3471.2). Total num frames: 3514368. Throughput: 0: 892.1. Samples: 878140. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) +[2023-02-23 21:09:17,818][00363] Avg episode reward: [(0, '26.452')] +[2023-02-23 21:09:17,821][12009] Saving new best policy, reward=26.452! +[2023-02-23 21:09:19,729][12023] Updated weights for policy 0, policy_version 860 (0.0016) +[2023-02-23 21:09:22,819][00363] Fps is (10 sec: 3275.7, 60 sec: 3481.4, 300 sec: 3443.4). Total num frames: 3526656. Throughput: 0: 858.1. Samples: 883140. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:09:22,824][00363] Avg episode reward: [(0, '26.859')] +[2023-02-23 21:09:22,852][12009] Saving new best policy, reward=26.859! +[2023-02-23 21:09:27,815][00363] Fps is (10 sec: 2867.3, 60 sec: 3549.9, 300 sec: 3429.5). Total num frames: 3543040. Throughput: 0: 856.3. Samples: 885194. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:09:27,821][00363] Avg episode reward: [(0, '27.755')] +[2023-02-23 21:09:27,829][12009] Saving new best policy, reward=27.755! +[2023-02-23 21:09:32,815][00363] Fps is (10 sec: 3277.9, 60 sec: 3413.3, 300 sec: 3443.4). Total num frames: 3559424. Throughput: 0: 873.9. Samples: 890016. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 21:09:32,818][00363] Avg episode reward: [(0, '28.632')] +[2023-02-23 21:09:32,844][12009] Saving new best policy, reward=28.632! +[2023-02-23 21:09:32,846][12023] Updated weights for policy 0, policy_version 870 (0.0029) +[2023-02-23 21:09:37,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3481.6, 300 sec: 3471.2). Total num frames: 3584000. Throughput: 0: 892.8. Samples: 896512. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:09:37,818][00363] Avg episode reward: [(0, '27.156')] +[2023-02-23 21:09:42,818][00363] Fps is (10 sec: 3685.3, 60 sec: 3481.5, 300 sec: 3443.4). Total num frames: 3596288. Throughput: 0: 866.1. Samples: 898448. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:09:42,821][00363] Avg episode reward: [(0, '26.670')] +[2023-02-23 21:09:46,381][12023] Updated weights for policy 0, policy_version 880 (0.0026) +[2023-02-23 21:09:47,815][00363] Fps is (10 sec: 2048.0, 60 sec: 3345.1, 300 sec: 3401.8). Total num frames: 3604480. Throughput: 0: 822.1. Samples: 901658. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 21:09:47,820][00363] Avg episode reward: [(0, '26.413')] +[2023-02-23 21:09:52,816][00363] Fps is (10 sec: 2048.5, 60 sec: 3276.8, 300 sec: 3401.8). Total num frames: 3616768. Throughput: 0: 800.4. Samples: 904842. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:09:52,820][00363] Avg episode reward: [(0, '26.756')] +[2023-02-23 21:09:52,832][12009] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000883_3616768.pth... +[2023-02-23 21:09:52,998][12009] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000683_2797568.pth +[2023-02-23 21:09:57,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3208.5, 300 sec: 3415.7). Total num frames: 3637248. Throughput: 0: 797.4. Samples: 907520. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:09:57,821][00363] Avg episode reward: [(0, '25.881')] +[2023-02-23 21:09:59,567][12023] Updated weights for policy 0, policy_version 890 (0.0028) +[2023-02-23 21:10:02,815][00363] Fps is (10 sec: 4096.1, 60 sec: 3345.1, 300 sec: 3443.4). Total num frames: 3657728. Throughput: 0: 796.6. Samples: 913986. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:10:02,819][00363] Avg episode reward: [(0, '25.677')] +[2023-02-23 21:10:07,823][00363] Fps is (10 sec: 3683.5, 60 sec: 3413.0, 300 sec: 3443.4). Total num frames: 3674112. Throughput: 0: 798.5. Samples: 919076. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:10:07,826][00363] Avg episode reward: [(0, '25.527')] +[2023-02-23 21:10:11,857][12023] Updated weights for policy 0, policy_version 900 (0.0012) +[2023-02-23 21:10:12,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3208.5, 300 sec: 3443.4). Total num frames: 3686400. Throughput: 0: 799.6. Samples: 921176. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:10:12,818][00363] Avg episode reward: [(0, '25.982')] +[2023-02-23 21:10:17,815][00363] Fps is (10 sec: 3279.4, 60 sec: 3208.6, 300 sec: 3471.2). Total num frames: 3706880. Throughput: 0: 807.5. Samples: 926354. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:10:17,820][00363] Avg episode reward: [(0, '26.022')] +[2023-02-23 21:10:22,037][12023] Updated weights for policy 0, policy_version 910 (0.0021) +[2023-02-23 21:10:22,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3345.3, 300 sec: 3471.2). Total num frames: 3727360. Throughput: 0: 809.4. Samples: 932936. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:10:22,818][00363] Avg episode reward: [(0, '25.441')] +[2023-02-23 21:10:27,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3345.1, 300 sec: 3443.4). Total num frames: 3743744. Throughput: 0: 830.2. Samples: 935804. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 21:10:27,820][00363] Avg episode reward: [(0, '24.104')] +[2023-02-23 21:10:32,816][00363] Fps is (10 sec: 3276.8, 60 sec: 3345.1, 300 sec: 3457.3). Total num frames: 3760128. Throughput: 0: 853.1. Samples: 940048. Policy #0 lag: (min: 0.0, avg: 0.7, max: 1.0) +[2023-02-23 21:10:32,823][00363] Avg episode reward: [(0, '23.574')] +[2023-02-23 21:10:35,302][12023] Updated weights for policy 0, policy_version 920 (0.0028) +[2023-02-23 21:10:37,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3208.5, 300 sec: 3471.2). Total num frames: 3776512. Throughput: 0: 898.0. Samples: 945252. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:10:37,822][00363] Avg episode reward: [(0, '22.795')] +[2023-02-23 21:10:42,815][00363] Fps is (10 sec: 4096.1, 60 sec: 3413.5, 300 sec: 3471.2). Total num frames: 3801088. Throughput: 0: 914.8. Samples: 948686. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:10:42,820][00363] Avg episode reward: [(0, '22.963')] +[2023-02-23 21:10:44,344][12023] Updated weights for policy 0, policy_version 930 (0.0017) +[2023-02-23 21:10:47,818][00363] Fps is (10 sec: 4094.9, 60 sec: 3549.7, 300 sec: 3457.3). Total num frames: 3817472. Throughput: 0: 907.2. Samples: 954812. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0) +[2023-02-23 21:10:47,823][00363] Avg episode reward: [(0, '23.045')] +[2023-02-23 21:10:52,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3618.2, 300 sec: 3471.2). Total num frames: 3833856. Throughput: 0: 886.7. Samples: 958970. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:10:52,820][00363] Avg episode reward: [(0, '23.190')] +[2023-02-23 21:10:57,454][12023] Updated weights for policy 0, policy_version 940 (0.0021) +[2023-02-23 21:10:57,815][00363] Fps is (10 sec: 3277.7, 60 sec: 3549.9, 300 sec: 3485.1). Total num frames: 3850240. Throughput: 0: 887.5. Samples: 961112. Policy #0 lag: (min: 0.0, avg: 0.6, max: 1.0) +[2023-02-23 21:10:57,822][00363] Avg episode reward: [(0, '24.278')] +[2023-02-23 21:11:02,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3618.1, 300 sec: 3499.0). Total num frames: 3874816. Throughput: 0: 921.3. Samples: 967814. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:11:02,824][00363] Avg episode reward: [(0, '25.470')] +[2023-02-23 21:11:06,717][12023] Updated weights for policy 0, policy_version 950 (0.0021) +[2023-02-23 21:11:07,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3618.6, 300 sec: 3485.1). Total num frames: 3891200. Throughput: 0: 909.2. Samples: 973850. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 21:11:07,818][00363] Avg episode reward: [(0, '25.034')] +[2023-02-23 21:11:12,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3686.4, 300 sec: 3485.1). Total num frames: 3907584. Throughput: 0: 892.8. Samples: 975978. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0) +[2023-02-23 21:11:12,821][00363] Avg episode reward: [(0, '24.888')] +[2023-02-23 21:11:17,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3618.1, 300 sec: 3499.0). Total num frames: 3923968. Throughput: 0: 895.0. Samples: 980322. Policy #0 lag: (min: 0.0, avg: 0.4, max: 2.0) +[2023-02-23 21:11:17,818][00363] Avg episode reward: [(0, '24.159')] +[2023-02-23 21:11:19,493][12023] Updated weights for policy 0, policy_version 960 (0.0031) +[2023-02-23 21:11:22,815][00363] Fps is (10 sec: 3686.4, 60 sec: 3618.1, 300 sec: 3499.0). Total num frames: 3944448. Throughput: 0: 927.5. Samples: 986988. Policy #0 lag: (min: 0.0, avg: 0.5, max: 2.0) +[2023-02-23 21:11:22,818][00363] Avg episode reward: [(0, '23.157')] +[2023-02-23 21:11:27,815][00363] Fps is (10 sec: 4096.0, 60 sec: 3686.4, 300 sec: 3499.0). Total num frames: 3964928. Throughput: 0: 925.2. Samples: 990320. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:11:27,819][00363] Avg episode reward: [(0, '22.139')] +[2023-02-23 21:11:30,113][12023] Updated weights for policy 0, policy_version 970 (0.0019) +[2023-02-23 21:11:32,815][00363] Fps is (10 sec: 3276.8, 60 sec: 3618.1, 300 sec: 3485.1). Total num frames: 3977216. Throughput: 0: 890.8. Samples: 994894. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0) +[2023-02-23 21:11:32,823][00363] Avg episode reward: [(0, '22.260')] +[2023-02-23 21:11:37,815][00363] Fps is (10 sec: 2867.2, 60 sec: 3618.1, 300 sec: 3485.1). Total num frames: 3993600. Throughput: 0: 899.6. Samples: 999454. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0) +[2023-02-23 21:11:37,818][00363] Avg episode reward: [(0, '23.220')] +[2023-02-23 21:11:40,045][12009] Stopping Batcher_0... +[2023-02-23 21:11:40,047][12009] Loop batcher_evt_loop terminating... +[2023-02-23 21:11:40,048][00363] Component Batcher_0 stopped! +[2023-02-23 21:11:40,052][12009] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... +[2023-02-23 21:11:40,090][12023] Weights refcount: 2 0 +[2023-02-23 21:11:40,100][12023] Stopping InferenceWorker_p0-w0... +[2023-02-23 21:11:40,100][00363] Component InferenceWorker_p0-w0 stopped! +[2023-02-23 21:11:40,101][12023] Loop inference_proc0-0_evt_loop terminating... +[2023-02-23 21:11:40,113][00363] Component RolloutWorker_w2 stopped! +[2023-02-23 21:11:40,115][12030] Stopping RolloutWorker_w2... +[2023-02-23 21:11:40,124][12030] Loop rollout_proc2_evt_loop terminating... +[2023-02-23 21:11:40,124][00363] Component RolloutWorker_w7 stopped! +[2023-02-23 21:11:40,126][12035] Stopping RolloutWorker_w7... +[2023-02-23 21:11:40,127][12035] Loop rollout_proc7_evt_loop terminating... +[2023-02-23 21:11:40,129][00363] Component RolloutWorker_w5 stopped! +[2023-02-23 21:11:40,131][12033] Stopping RolloutWorker_w5... +[2023-02-23 21:11:40,132][12033] Loop rollout_proc5_evt_loop terminating... +[2023-02-23 21:11:40,150][00363] Component RolloutWorker_w1 stopped! +[2023-02-23 21:11:40,152][12026] Stopping RolloutWorker_w1... +[2023-02-23 21:11:40,158][12034] Stopping RolloutWorker_w6... +[2023-02-23 21:11:40,158][00363] Component RolloutWorker_w6 stopped! +[2023-02-23 21:11:40,166][00363] Component RolloutWorker_w4 stopped! +[2023-02-23 21:11:40,156][12026] Loop rollout_proc1_evt_loop terminating... +[2023-02-23 21:11:40,166][12032] Stopping RolloutWorker_w4... +[2023-02-23 21:11:40,174][00363] Component RolloutWorker_w0 stopped! +[2023-02-23 21:11:40,159][12034] Loop rollout_proc6_evt_loop terminating... +[2023-02-23 21:11:40,173][12024] Stopping RolloutWorker_w0... +[2023-02-23 21:11:40,172][12032] Loop rollout_proc4_evt_loop terminating... +[2023-02-23 21:11:40,188][00363] Component RolloutWorker_w3 stopped! +[2023-02-23 21:11:40,190][12031] Stopping RolloutWorker_w3... +[2023-02-23 21:11:40,182][12024] Loop rollout_proc0_evt_loop terminating... +[2023-02-23 21:11:40,196][12031] Loop rollout_proc3_evt_loop terminating... +[2023-02-23 21:11:40,215][12009] Removing /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000785_3215360.pth +[2023-02-23 21:11:40,231][12009] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... +[2023-02-23 21:11:40,397][00363] Component LearnerWorker_p0 stopped! +[2023-02-23 21:11:40,407][00363] Waiting for process learner_proc0 to stop... +[2023-02-23 21:11:40,412][12009] Stopping LearnerWorker_p0... +[2023-02-23 21:11:40,412][12009] Loop learner_proc0_evt_loop terminating... +[2023-02-23 21:11:42,218][00363] Waiting for process inference_proc0-0 to join... +[2023-02-23 21:11:42,577][00363] Waiting for process rollout_proc0 to join... +[2023-02-23 21:11:43,068][00363] Waiting for process rollout_proc1 to join... +[2023-02-23 21:11:43,070][00363] Waiting for process rollout_proc2 to join... +[2023-02-23 21:11:43,072][00363] Waiting for process rollout_proc3 to join... +[2023-02-23 21:11:43,074][00363] Waiting for process rollout_proc4 to join... +[2023-02-23 21:11:43,075][00363] Waiting for process rollout_proc5 to join... +[2023-02-23 21:11:43,076][00363] Waiting for process rollout_proc6 to join... +[2023-02-23 21:11:43,077][00363] Waiting for process rollout_proc7 to join... +[2023-02-23 21:11:43,082][00363] Batcher 0 profile tree view: +batching: 26.8718, releasing_batches: 0.0239 +[2023-02-23 21:11:43,083][00363] InferenceWorker_p0-w0 profile tree view: +wait_policy: 0.0091 + wait_policy_total: 557.5544 +update_model: 8.1307 + weight_update: 0.0028 +one_step: 0.0024 + handle_policy_step: 547.3937 + deserialize: 15.2261, stack: 3.1522, obs_to_device_normalize: 117.0745, forward: 270.3107, send_messages: 27.1799 + prepare_outputs: 86.9411 + to_cpu: 54.3151 +[2023-02-23 21:11:43,084][00363] Learner 0 profile tree view: +misc: 0.0051, prepare_batch: 17.0419 +train: 76.5972 + epoch_init: 0.0111, minibatch_init: 0.0106, losses_postprocess: 0.5798, kl_divergence: 0.6407, after_optimizer: 32.8984 + calculate_losses: 27.0207 + losses_init: 0.0035, forward_head: 1.8386, bptt_initial: 17.6977, tail: 1.2137, advantages_returns: 0.3290, losses: 3.5520 + bptt: 2.0859 + bptt_forward_core: 1.9971 + update: 14.8611 + clip: 1.3935 +[2023-02-23 21:11:43,088][00363] RolloutWorker_w0 profile tree view: +wait_for_trajectories: 0.3406, enqueue_policy_requests: 154.3019, env_step: 869.4700, overhead: 22.3603, complete_rollouts: 7.5145 +save_policy_outputs: 21.5735 + split_output_tensors: 10.6250 +[2023-02-23 21:11:43,089][00363] RolloutWorker_w7 profile tree view: +wait_for_trajectories: 0.3045, enqueue_policy_requests: 156.6177, env_step: 866.6779, overhead: 22.6247, complete_rollouts: 7.2682 +save_policy_outputs: 21.5111 + split_output_tensors: 10.6498 +[2023-02-23 21:11:43,090][00363] Loop Runner_EvtLoop terminating... +[2023-02-23 21:11:43,092][00363] Runner profile tree view: +main_loop: 1185.8479 +[2023-02-23 21:11:43,095][00363] Collected {0: 4005888}, FPS: 3378.1 +[2023-02-23 21:11:43,221][00363] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json +[2023-02-23 21:11:43,223][00363] Overriding arg 'num_workers' with value 1 passed from command line +[2023-02-23 21:11:43,226][00363] Adding new argument 'no_render'=True that is not in the saved config file! +[2023-02-23 21:11:43,228][00363] Adding new argument 'save_video'=True that is not in the saved config file! +[2023-02-23 21:11:43,230][00363] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file! +[2023-02-23 21:11:43,232][00363] Adding new argument 'video_name'=None that is not in the saved config file! +[2023-02-23 21:11:43,234][00363] Adding new argument 'max_num_frames'=1000000000.0 that is not in the saved config file! +[2023-02-23 21:11:43,236][00363] Adding new argument 'max_num_episodes'=10 that is not in the saved config file! +[2023-02-23 21:11:43,237][00363] Adding new argument 'push_to_hub'=False that is not in the saved config file! +[2023-02-23 21:11:43,238][00363] Adding new argument 'hf_repository'=None that is not in the saved config file! +[2023-02-23 21:11:43,239][00363] Adding new argument 'policy_index'=0 that is not in the saved config file! +[2023-02-23 21:11:43,240][00363] Adding new argument 'eval_deterministic'=False that is not in the saved config file! +[2023-02-23 21:11:43,241][00363] Adding new argument 'train_script'=None that is not in the saved config file! +[2023-02-23 21:11:43,242][00363] Adding new argument 'enjoy_script'=None that is not in the saved config file! +[2023-02-23 21:11:43,249][00363] Using frameskip 1 and render_action_repeat=4 for evaluation +[2023-02-23 21:11:43,272][00363] Doom resolution: 160x120, resize resolution: (128, 72) +[2023-02-23 21:11:43,274][00363] RunningMeanStd input shape: (3, 72, 128) +[2023-02-23 21:11:43,279][00363] RunningMeanStd input shape: (1,) +[2023-02-23 21:11:43,294][00363] ConvEncoder: input_channels=3 +[2023-02-23 21:11:43,972][00363] Conv encoder output size: 512 +[2023-02-23 21:11:43,974][00363] Policy head output size: 512 +[2023-02-23 21:11:46,331][00363] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth... +[2023-02-23 21:11:47,885][00363] Num frames 100... +[2023-02-23 21:11:48,050][00363] Num frames 200... +[2023-02-23 21:11:48,208][00363] Num frames 300... +[2023-02-23 21:11:48,365][00363] Num frames 400... +[2023-02-23 21:11:48,449][00363] Avg episode rewards: #0: 6.160, true rewards: #0: 4.160 +[2023-02-23 21:11:48,455][00363] Avg episode reward: 6.160, avg true_objective: 4.160 +[2023-02-23 21:11:48,590][00363] Num frames 500... +[2023-02-23 21:11:48,774][00363] Num frames 600... +[2023-02-23 21:11:48,941][00363] Num frames 700... +[2023-02-23 21:11:49,103][00363] Num frames 800... +[2023-02-23 21:11:49,263][00363] Num frames 900... +[2023-02-23 21:11:49,424][00363] Num frames 1000... +[2023-02-23 21:11:49,600][00363] Num frames 1100... +[2023-02-23 21:11:49,762][00363] Num frames 1200... +[2023-02-23 21:11:49,920][00363] Num frames 1300... +[2023-02-23 21:11:50,051][00363] Avg episode rewards: #0: 13.735, true rewards: #0: 6.735 +[2023-02-23 21:11:50,054][00363] Avg episode reward: 13.735, avg true_objective: 6.735 +[2023-02-23 21:11:50,155][00363] Num frames 1400... +[2023-02-23 21:11:50,337][00363] Num frames 1500... +[2023-02-23 21:11:50,485][00363] Num frames 1600... +[2023-02-23 21:11:50,605][00363] Num frames 1700... +[2023-02-23 21:11:50,724][00363] Num frames 1800... +[2023-02-23 21:11:50,845][00363] Num frames 1900... +[2023-02-23 21:11:50,964][00363] Num frames 2000... +[2023-02-23 21:11:51,083][00363] Num frames 2100... +[2023-02-23 21:11:51,199][00363] Num frames 2200... +[2023-02-23 21:11:51,324][00363] Num frames 2300... +[2023-02-23 21:11:51,436][00363] Num frames 2400... +[2023-02-23 21:11:51,553][00363] Num frames 2500... +[2023-02-23 21:11:51,665][00363] Num frames 2600... +[2023-02-23 21:11:51,779][00363] Num frames 2700... +[2023-02-23 21:11:51,909][00363] Num frames 2800... +[2023-02-23 21:11:52,022][00363] Num frames 2900... +[2023-02-23 21:11:52,141][00363] Num frames 3000... +[2023-02-23 21:11:52,255][00363] Num frames 3100... +[2023-02-23 21:11:52,377][00363] Num frames 3200... +[2023-02-23 21:11:52,496][00363] Num frames 3300... +[2023-02-23 21:11:52,622][00363] Num frames 3400... +[2023-02-23 21:11:52,733][00363] Avg episode rewards: #0: 28.156, true rewards: #0: 11.490 +[2023-02-23 21:11:52,736][00363] Avg episode reward: 28.156, avg true_objective: 11.490 +[2023-02-23 21:11:52,797][00363] Num frames 3500... +[2023-02-23 21:11:52,916][00363] Num frames 3600... +[2023-02-23 21:11:53,025][00363] Num frames 3700... +[2023-02-23 21:11:53,139][00363] Num frames 3800... +[2023-02-23 21:11:53,249][00363] Num frames 3900... +[2023-02-23 21:11:53,359][00363] Num frames 4000... +[2023-02-23 21:11:53,474][00363] Num frames 4100... +[2023-02-23 21:11:53,589][00363] Num frames 4200... +[2023-02-23 21:11:53,702][00363] Num frames 4300... +[2023-02-23 21:11:53,819][00363] Num frames 4400... +[2023-02-23 21:11:53,938][00363] Num frames 4500... +[2023-02-23 21:11:54,057][00363] Num frames 4600... +[2023-02-23 21:11:54,172][00363] Num frames 4700... +[2023-02-23 21:11:54,282][00363] Num frames 4800... +[2023-02-23 21:11:54,396][00363] Num frames 4900... +[2023-02-23 21:11:54,506][00363] Num frames 5000... +[2023-02-23 21:11:54,621][00363] Num frames 5100... +[2023-02-23 21:11:54,732][00363] Num frames 5200... +[2023-02-23 21:11:54,843][00363] Num frames 5300... +[2023-02-23 21:11:54,966][00363] Num frames 5400... +[2023-02-23 21:11:55,078][00363] Num frames 5500... +[2023-02-23 21:11:55,186][00363] Avg episode rewards: #0: 35.117, true rewards: #0: 13.868 +[2023-02-23 21:11:55,189][00363] Avg episode reward: 35.117, avg true_objective: 13.868 +[2023-02-23 21:11:55,251][00363] Num frames 5600... +[2023-02-23 21:11:55,359][00363] Num frames 5700... +[2023-02-23 21:11:55,471][00363] Num frames 5800... +[2023-02-23 21:11:55,579][00363] Num frames 5900... +[2023-02-23 21:11:55,694][00363] Num frames 6000... +[2023-02-23 21:11:55,819][00363] Num frames 6100... +[2023-02-23 21:11:55,938][00363] Num frames 6200... +[2023-02-23 21:11:56,055][00363] Num frames 6300... +[2023-02-23 21:11:56,174][00363] Num frames 6400... +[2023-02-23 21:11:56,297][00363] Num frames 6500... +[2023-02-23 21:11:56,409][00363] Num frames 6600... +[2023-02-23 21:11:56,524][00363] Num frames 6700... +[2023-02-23 21:11:56,687][00363] Avg episode rewards: #0: 34.190, true rewards: #0: 13.590 +[2023-02-23 21:11:56,689][00363] Avg episode reward: 34.190, avg true_objective: 13.590 +[2023-02-23 21:11:56,702][00363] Num frames 6800... +[2023-02-23 21:11:56,833][00363] Num frames 6900... +[2023-02-23 21:11:56,965][00363] Num frames 7000... +[2023-02-23 21:11:57,090][00363] Num frames 7100... +[2023-02-23 21:11:57,203][00363] Num frames 7200... +[2023-02-23 21:11:57,318][00363] Num frames 7300... +[2023-02-23 21:11:57,430][00363] Num frames 7400... +[2023-02-23 21:11:57,544][00363] Num frames 7500... +[2023-02-23 21:11:57,661][00363] Num frames 7600... +[2023-02-23 21:11:57,775][00363] Num frames 7700... +[2023-02-23 21:11:57,891][00363] Num frames 7800... +[2023-02-23 21:11:58,014][00363] Num frames 7900... +[2023-02-23 21:11:58,127][00363] Num frames 8000... +[2023-02-23 21:11:58,251][00363] Num frames 8100... +[2023-02-23 21:11:58,367][00363] Num frames 8200... +[2023-02-23 21:11:58,517][00363] Avg episode rewards: #0: 35.306, true rewards: #0: 13.807 +[2023-02-23 21:11:58,520][00363] Avg episode reward: 35.306, avg true_objective: 13.807 +[2023-02-23 21:11:58,544][00363] Num frames 8300... +[2023-02-23 21:11:58,665][00363] Num frames 8400... +[2023-02-23 21:11:58,795][00363] Num frames 8500... +[2023-02-23 21:11:58,917][00363] Num frames 8600... +[2023-02-23 21:11:59,044][00363] Num frames 8700... +[2023-02-23 21:11:59,160][00363] Num frames 8800... +[2023-02-23 21:11:59,273][00363] Num frames 8900... +[2023-02-23 21:11:59,389][00363] Num frames 9000... +[2023-02-23 21:11:59,501][00363] Num frames 9100... +[2023-02-23 21:11:59,618][00363] Num frames 9200... +[2023-02-23 21:11:59,734][00363] Num frames 9300... +[2023-02-23 21:11:59,860][00363] Num frames 9400... +[2023-02-23 21:11:59,988][00363] Num frames 9500... +[2023-02-23 21:12:00,104][00363] Num frames 9600... +[2023-02-23 21:12:00,186][00363] Avg episode rewards: #0: 34.888, true rewards: #0: 13.746 +[2023-02-23 21:12:00,188][00363] Avg episode reward: 34.888, avg true_objective: 13.746 +[2023-02-23 21:12:00,292][00363] Num frames 9700... +[2023-02-23 21:12:00,412][00363] Num frames 9800... +[2023-02-23 21:12:00,576][00363] Num frames 9900... +[2023-02-23 21:12:00,737][00363] Num frames 10000... +[2023-02-23 21:12:00,906][00363] Num frames 10100... +[2023-02-23 21:12:01,054][00363] Avg episode rewards: #0: 31.819, true rewards: #0: 12.694 +[2023-02-23 21:12:01,060][00363] Avg episode reward: 31.819, avg true_objective: 12.694 +[2023-02-23 21:12:01,136][00363] Num frames 10200... +[2023-02-23 21:12:01,295][00363] Num frames 10300... +[2023-02-23 21:12:01,459][00363] Num frames 10400... +[2023-02-23 21:12:01,638][00363] Avg episode rewards: #0: 28.972, true rewards: #0: 11.639 +[2023-02-23 21:12:01,640][00363] Avg episode reward: 28.972, avg true_objective: 11.639 +[2023-02-23 21:12:01,691][00363] Num frames 10500... +[2023-02-23 21:12:01,855][00363] Num frames 10600... +[2023-02-23 21:12:02,020][00363] Num frames 10700... +[2023-02-23 21:12:02,188][00363] Num frames 10800... +[2023-02-23 21:12:02,354][00363] Num frames 10900... +[2023-02-23 21:12:02,526][00363] Num frames 11000... +[2023-02-23 21:12:02,697][00363] Num frames 11100... +[2023-02-23 21:12:02,863][00363] Num frames 11200... +[2023-02-23 21:12:03,045][00363] Num frames 11300... +[2023-02-23 21:12:03,218][00363] Num frames 11400... +[2023-02-23 21:12:03,309][00363] Avg episode rewards: #0: 28.319, true rewards: #0: 11.419 +[2023-02-23 21:12:03,312][00363] Avg episode reward: 28.319, avg true_objective: 11.419 +[2023-02-23 21:13:17,273][00363] Replay video saved to /content/train_dir/default_experiment/replay.mp4!