Upload . with huggingface_hub

.gitattributes

@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+replay.mp4 filter=lfs diff=lfs merge=lfs -text

.summary/0/events.out.tfevents.1673492602.brain2.usc.edu

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fb2bb6e00f09ec189faa9cfd8f9793d154723ce6b6c3cae0c6de99b18f37d9e
+size 137478

README.md

@@ -0,0 +1,56 @@
+---
+library_name: sample-factory
+tags:
+- deep-reinforcement-learning
+- reinforcement-learning
+- sample-factory
+model-index:
+- name: APPO
+  results:
+  - task:
+      type: reinforcement-learning
+      name: reinforcement-learning
+    dataset:
+      name: humanoid
+      type: humanoid
+    metrics:
+    - type: mean_reward
+      value: 33847.53 +/- 6327.36
+      name: mean_reward
+      verified: false
+---
+
+A(n) **APPO** model trained on the **humanoid** environment.
+
+This model was trained using Sample-Factory 2.0: https://github.com/alex-petrenko/sample-factory.
+Documentation for how to use Sample-Factory can be found at https://www.samplefactory.dev/
+
+
+## Downloading the model
+
+After installing Sample-Factory, download the model with:
+```
+python -m sample_factory.huggingface.load_from_hub -r apetrenko/sample_factory_brax_humanoid
+```
+
+    
+## Using the model
+
+To run the model after download, use the `enjoy` script corresponding to this environment:
+```
+python -m sf_examples.brax.enjoy_brax --algo=APPO --env=humanoid --train_dir=./train_dir --experiment=sample_factory_brax_humanoid
+```
+
+
+You can also upload models to the Hugging Face Hub using the same script with the `--push_to_hub` flag.
+See https://www.samplefactory.dev/10-huggingface/huggingface/ for more details
+    
+## Training with this model
+
+To continue training with this model, use the `train` script corresponding to this environment:
+```
+python -m sf_examples.brax.train_brax --algo=APPO --env=humanoid --train_dir=./train_dir --experiment=sample_factory_brax_humanoid --restart_behavior=resume --train_for_env_steps=10000000000
+```
+
+Note, you may have to adjust `--train_for_env_steps` to a suitably high number as the experiment will resume at the number of steps it concluded at.
+    

checkpoint_p0/best_000014914_97779712_reward_35312.879.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d83ea810de8cb1fe3fa77530ef544e2f2603d8c0bbf1145904d0fc132c60ed38
+size 3499383

checkpoint_p0/checkpoint_000011080_72613888.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5ef0093b0501c8500b078298cd42119ac5a7c0c13fdebaf10957d4a58cbd2a4
+size 3499695

checkpoint_p0/checkpoint_000015264_100073472.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9f562dde39eac5d324a2b12b28b6df1d310aa86483746ccc00a7c7b9b98b1d6
+size 3499695

config.json

@@ -0,0 +1,147 @@
+{
+  "help": false,
+  "algo": "APPO",
+  "env": "humanoid",
+  "experiment": "02_v083_brax_basic_benchmark_see_2322090_env_humanoid_u.rnn_False_n.epo_5",
+  "train_dir": "./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm",
+  "restart_behavior": "resume",
+  "device": "gpu",
+  "seed": 2322090,
+  "num_policies": 1,
+  "async_rl": false,
+  "serial_mode": true,
+  "batched_sampling": true,
+  "num_batches_to_accumulate": 2,
+  "worker_num_splits": 1,
+  "policy_workers_per_policy": 1,
+  "max_policy_lag": 1000,
+  "num_workers": 1,
+  "num_envs_per_worker": 1,
+  "batch_size": 32768,
+  "num_batches_per_epoch": 2,
+  "num_epochs": 5,
+  "rollout": 32,
+  "recurrence": 1,
+  "shuffle_minibatches": false,
+  "gamma": 0.99,
+  "reward_scale": 0.01,
+  "reward_clip": 1000.0,
+  "value_bootstrap": true,
+  "normalize_returns": true,
+  "exploration_loss_coeff": 0.0,
+  "value_loss_coeff": 2.0,
+  "kl_loss_coeff": 0.0,
+  "exploration_loss": "entropy",
+  "gae_lambda": 0.95,
+  "ppo_clip_ratio": 0.2,
+  "ppo_clip_value": 1.0,
+  "with_vtrace": false,
+  "vtrace_rho": 1.0,
+  "vtrace_c": 1.0,
+  "optimizer": "adam",
+  "adam_eps": 1e-06,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.999,
+  "max_grad_norm": 1.0,
+  "learning_rate": 0.0003,
+  "lr_schedule": "kl_adaptive_epoch",
+  "lr_schedule_kl_threshold": 0.008,
+  "lr_adaptive_min": 1e-06,
+  "lr_adaptive_max": 0.002,
+  "obs_subtract_mean": 0.0,
+  "obs_scale": 1.0,
+  "normalize_input": true,
+  "normalize_input_keys": null,
+  "decorrelate_experience_max_seconds": 0,
+  "decorrelate_envs_on_one_worker": true,
+  "actor_worker_gpus": [
+    0
+  ],
+  "set_workers_cpu_affinity": true,
+  "force_envs_single_thread": false,
+  "default_niceness": 0,
+  "log_to_file": true,
+  "experiment_summaries_interval": 10,
+  "flush_summaries_interval": 30,
+  "stats_avg": 100,
+  "summaries_use_frameskip": true,
+  "heartbeat_interval": 20,
+  "heartbeat_reporting_interval": 180,
+  "train_for_env_steps": 100000000,
+  "train_for_seconds": 10000000000,
+  "save_every_sec": 120,
+  "keep_checkpoints": 2,
+  "load_checkpoint_kind": "latest",
+  "save_milestones_sec": -1,
+  "save_best_every_sec": 5,
+  "save_best_metric": "reward",
+  "save_best_after": 5000000,
+  "benchmark": false,
+  "encoder_mlp_layers": [
+    512,
+    256,
+    128
+  ],
+  "encoder_conv_architecture": "convnet_simple",
+  "encoder_conv_mlp_layers": [
+    512
+  ],
+  "use_rnn": false,
+  "rnn_size": 512,
+  "rnn_type": "gru",
+  "rnn_num_layers": 1,
+  "decoder_mlp_layers": [],
+  "nonlinearity": "elu",
+  "policy_initialization": "torch_default",
+  "policy_init_gain": 1.0,
+  "actor_critic_share_weights": true,
+  "adaptive_stddev": false,
+  "continuous_tanh_scale": 0.0,
+  "initial_stddev": 1.0,
+  "use_env_info_cache": false,
+  "env_gpu_actions": true,
+  "env_gpu_observations": true,
+  "env_frameskip": 1,
+  "env_framestack": 1,
+  "pixel_format": "CHW",
+  "use_record_episode_statistics": false,
+  "with_wandb": true,
+  "wandb_user": null,
+  "wandb_project": "sample_factory",
+  "wandb_group": null,
+  "wandb_job_type": "SF",
+  "wandb_tags": [],
+  "with_pbt": false,
+  "pbt_mix_policies_in_one_env": true,
+  "pbt_period_env_steps": 5000000,
+  "pbt_start_mutation": 20000000,
+  "pbt_replace_fraction": 0.3,
+  "pbt_mutation_rate": 0.15,
+  "pbt_replace_reward_gap": 0.1,
+  "pbt_replace_reward_gap_absolute": 1e-06,
+  "pbt_optimize_gamma": false,
+  "pbt_target_objective": "true_objective",
+  "pbt_perturb_min": 1.1,
+  "pbt_perturb_max": 1.5,
+  "env_agents": 2048,
+  "clamp_actions": false,
+  "clamp_rew_obs": false,
+  "command_line": "--actor_worker_gpus 0 --wandb_project=sample_factory --with_wandb=True --seed=2322090 --env=humanoid --use_rnn=False --num_epochs=5 --experiment=02_v083_brax_basic_benchmark_see_2322090_env_humanoid_u.rnn_False_n.epo_5 --train_dir=./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm",
+  "cli_args": {
+    "env": "humanoid",
+    "experiment": "02_v083_brax_basic_benchmark_see_2322090_env_humanoid_u.rnn_False_n.epo_5",
+    "train_dir": "./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm",
+    "seed": 2322090,
+    "num_epochs": 5,
+    "actor_worker_gpus": [
+      0
+    ],
+    "use_rnn": false,
+    "with_wandb": true,
+    "wandb_project": "sample_factory"
+  },
+  "git_hash": "6aa87f2d416b9fad874b299d864a522c887c238a",
+  "git_repo_name": "git@github.com:alex-petrenko/sample-factory.git",
+  "train_script": "sf_examples.brax.train_brax",
+  "wandb_unique_id": "02_v083_brax_basic_benchmark_see_2322090_env_humanoid_u.rnn_False_n.epo_5_20230111_190315_359969"
+}

replay.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ed43c64af48c394d505d5c3631c6c859e773f1b34162d7e6e3158ba1b3256ac
+size 2260300

sf_log.txt

@@ -0,0 +1,318 @@
+[2023-01-11 19:03:27,380][454600] Saving configuration to ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/02_v083_brax_basic_benchmark_see_2322090_env_humanoid_u.rnn_False_n.epo_5/config.json...
+[2023-01-11 19:03:27,559][454600] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-01-11 19:03:27,561][454600] Rollout worker 0 uses device cuda:0
+[2023-01-11 19:03:27,562][454600] In synchronous mode, we only accumulate one batch. Setting num_batches_to_accumulate to 1
+[2023-01-11 19:03:27,679][454600] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-01-11 19:03:27,680][454600] InferenceWorker_p0-w0: min num requests: 1
+[2023-01-11 19:03:27,681][454600] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-01-11 19:03:27,682][454600] WARNING! It is generally recommended to enable Fixed KL loss (https://arxiv.org/pdf/1707.06347.pdf) for continuous action tasks to avoid potential numerical issues. I.e. set --kl_loss_coeff=0.1
+[2023-01-11 19:03:27,682][454600] Setting fixed seed 2322090
+[2023-01-11 19:03:27,682][454600] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-01-11 19:03:27,683][454600] Initializing actor-critic model on device cuda:0
+[2023-01-11 19:03:27,683][454600] RunningMeanStd input shape: (240,)
+[2023-01-11 19:03:27,684][454600] RunningMeanStd input shape: (1,)
+[2023-01-11 19:03:27,768][454600] Created Actor Critic model with architecture:
+[2023-01-11 19:03:27,769][454600] ActorCriticSharedWeights(
+  (obs_normalizer): ObservationNormalizer(
+    (running_mean_std): RunningMeanStdDictInPlace(
+      (running_mean_std): ModuleDict(
+        (obs): RunningMeanStdInPlace()
+      )
+    )
+  )
+  (returns_normalizer): RecursiveScriptModule(original_name=RunningMeanStdInPlace)
+  (encoder): MultiInputEncoder(
+    (encoders): ModuleDict(
+      (obs): MlpEncoder(
+        (mlp_head): RecursiveScriptModule(
+          original_name=Sequential
+          (0): RecursiveScriptModule(original_name=Linear)
+          (1): RecursiveScriptModule(original_name=ELU)
+          (2): RecursiveScriptModule(original_name=Linear)
+          (3): RecursiveScriptModule(original_name=ELU)
+          (4): RecursiveScriptModule(original_name=Linear)
+          (5): RecursiveScriptModule(original_name=ELU)
+        )
+      )
+    )
+  )
+  (core): ModelCoreIdentity()
+  (decoder): MlpDecoder(
+    (mlp): Identity()
+  )
+  (critic_linear): Linear(in_features=128, out_features=1, bias=True)
+  (action_parameterization): ActionParameterizationContinuousNonAdaptiveStddev(
+    (distribution_linear): Linear(in_features=128, out_features=17, bias=True)
+  )
+)
+[2023-01-11 19:03:27,771][454600] Using optimizer <class 'torch.optim.adam.Adam'>
+[2023-01-11 19:03:27,775][454600] No checkpoints found
+[2023-01-11 19:03:27,775][454600] Did not load from checkpoint, starting from scratch!
+[2023-01-11 19:03:27,775][454600] Initialized policy 0 weights for model version 0
+[2023-01-11 19:03:27,776][454600] LearnerWorker_p0 finished initialization!
+[2023-01-11 19:03:27,776][454600] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-01-11 19:03:27,811][454600] Inference worker 0-0 is ready!
+[2023-01-11 19:03:27,812][454600] All inference workers are ready! Signal rollout workers to start!
+[2023-01-11 19:03:27,812][454600] EnvRunner 0-0 uses policy 0
+[2023-01-11 19:03:29,365][454600] Resetting env <VectorGymWrapper instance> with 2048 parallel agents...
+[2023-01-11 19:03:36,383][454600] reset() done, obs.shape=torch.Size([2048, 240])!
+[2023-01-11 19:03:36,392][454600] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 0. Throughput: 0: nan. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2023-01-11 19:03:49,705][454600] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 153.8. Samples: 2048. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2023-01-11 19:04:02,759][454600] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 155.3. Samples: 4096. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2023-01-11 19:04:02,763][454600] Heartbeat connected on Batcher_0
+[2023-01-11 19:04:02,763][454600] Heartbeat connected on LearnerWorker_p0
+[2023-01-11 19:04:02,764][454600] Heartbeat connected on InferenceWorker_p0-w0
+[2023-01-11 19:04:02,764][454600] Heartbeat connected on RolloutWorker_w0
+[2023-01-11 19:04:02,764][454600] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 155.3. Samples: 4096. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2023-01-11 19:04:02,772][454600] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 232.9. Samples: 6144. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2023-01-11 19:04:02,776][454600] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 232.9. Samples: 6144. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2023-01-11 19:04:02,782][454600] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 0. Throughput: 0: 310.4. Samples: 8192. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2023-01-11 19:04:07,026][454600] Fps is (10 sec: 154179.0, 60 sec: 21392.9, 300 sec: 21392.9). Total num frames: 655360. Throughput: 0: 20791.2. Samples: 636928. Policy #0 lag: (min: 2.0, avg: 2.0, max: 2.0)
+[2023-01-11 19:04:07,027][454600] Avg episode reward: [(0, '253.034')]
+[2023-01-11 19:04:12,022][454600] Fps is (10 sec: 198596.7, 60 sec: 51502.3, 300 sec: 51502.3). Total num frames: 1835008. Throughput: 0: 37247.2. Samples: 1327104. Policy #0 lag: (min: 7.0, avg: 7.0, max: 7.0)
+[2023-01-11 19:04:12,022][454600] Avg episode reward: [(0, '599.874')]
+[2023-01-11 19:04:17,023][454600] Fps is (10 sec: 236000.6, 60 sec: 74195.2, 300 sec: 74195.2). Total num frames: 3014656. Throughput: 0: 67743.4. Samples: 2752512. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:04:17,024][454600] Avg episode reward: [(0, '1565.579')]
+[2023-01-11 19:04:22,024][454600] Fps is (10 sec: 235864.8, 60 sec: 91915.0, 300 sec: 91915.0). Total num frames: 4194304. Throughput: 0: 128953.6. Samples: 4169728. Policy #0 lag: (min: 2.0, avg: 2.0, max: 2.0)
+[2023-01-11 19:04:22,025][454600] Avg episode reward: [(0, '2174.293')]
+[2023-01-11 19:04:27,022][454600] Fps is (10 sec: 235962.2, 60 sec: 106141.6, 300 sec: 106141.6). Total num frames: 5373952. Throughput: 0: 199881.7. Samples: 4853760. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0)
+[2023-01-11 19:04:27,023][454600] Avg episode reward: [(0, '3782.828')]
+[2023-01-11 19:04:27,029][454600] Saving new best policy, reward=3782.828!
+[2023-01-11 19:04:32,067][454600] Fps is (10 sec: 234929.6, 60 sec: 117711.8, 300 sec: 117711.8). Total num frames: 6553600. Throughput: 0: 213516.5. Samples: 6260736. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0)
+[2023-01-11 19:04:32,068][454600] Avg episode reward: [(0, '4961.963')]
+[2023-01-11 19:04:32,078][454600] Saving new best policy, reward=4961.963!
+[2023-01-11 19:04:37,022][454600] Fps is (10 sec: 229381.5, 60 sec: 162051.1, 300 sec: 126467.7). Total num frames: 7667712. Throughput: 0: 223515.5. Samples: 7661568. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:04:37,022][454600] Avg episode reward: [(0, '6290.788')]
+[2023-01-11 19:04:37,029][454600] Saving new best policy, reward=6290.788!
+[2023-01-11 19:04:42,023][454600] Fps is (10 sec: 230389.8, 60 sec: 225332.4, 300 sec: 134804.7). Total num frames: 8847360. Throughput: 0: 212329.0. Samples: 8339456. Policy #0 lag: (min: 7.0, avg: 7.0, max: 7.0)
+[2023-01-11 19:04:42,023][454600] Avg episode reward: [(0, '10632.291')]
+[2023-01-11 19:04:42,034][454600] Saving new best policy, reward=10632.291!
+[2023-01-11 19:04:47,077][454600] Fps is (10 sec: 234644.0, 60 sec: 226279.6, 300 sec: 141855.6). Total num frames: 10027008. Throughput: 0: 219619.4. Samples: 9736192. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:04:47,077][454600] Avg episode reward: [(0, '12825.875')]
+[2023-01-11 19:04:47,080][454600] Saving new best policy, reward=12825.875!
+[2023-01-11 19:04:52,023][454600] Fps is (10 sec: 229370.2, 60 sec: 226208.8, 300 sec: 147308.5). Total num frames: 11141120. Throughput: 0: 233124.6. Samples: 11126784. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:04:52,024][454600] Avg episode reward: [(0, '15167.174')]
+[2023-01-11 19:04:52,036][454600] Saving new best policy, reward=15167.174!
+[2023-01-11 19:04:57,025][454600] Fps is (10 sec: 230575.5, 60 sec: 227116.0, 300 sec: 152801.4). Total num frames: 12320768. Throughput: 0: 232637.7. Samples: 11796480. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:04:57,026][454600] Avg episode reward: [(0, '16256.997')]
+[2023-01-11 19:04:57,032][454600] Saving new best policy, reward=16256.997!
+[2023-01-11 19:05:02,097][454600] Fps is (10 sec: 234190.3, 60 sec: 227602.7, 300 sec: 157521.0). Total num frames: 13500416. Throughput: 0: 231907.0. Samples: 13205504. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:05:02,098][454600] Avg episode reward: [(0, '19859.633')]
+[2023-01-11 19:05:02,105][454600] Saving new best policy, reward=19859.633!
+[2023-01-11 19:05:07,023][454600] Fps is (10 sec: 229423.3, 60 sec: 232668.0, 300 sec: 161254.0). Total num frames: 14614528. Throughput: 0: 231888.8. Samples: 14604288. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:05:07,023][454600] Avg episode reward: [(0, '20474.635')]
+[2023-01-11 19:05:07,029][454600] Saving new best policy, reward=20474.635!
+[2023-01-11 19:05:12,023][454600] Fps is (10 sec: 231103.2, 60 sec: 232648.6, 300 sec: 165157.9). Total num frames: 15794176. Throughput: 0: 231465.9. Samples: 15269888. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:05:12,023][454600] Avg episode reward: [(0, '22723.672')]
+[2023-01-11 19:05:12,030][454600] Saving new best policy, reward=22723.672!
+[2023-01-11 19:05:17,021][454600] Fps is (10 sec: 229405.8, 60 sec: 231569.1, 300 sec: 168025.6). Total num frames: 16908288. Throughput: 0: 231431.7. Samples: 16664576. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:05:17,022][454600] Avg episode reward: [(0, '23173.908')]
+[2023-01-11 19:05:17,028][454600] Saving new best policy, reward=23173.908!
+[2023-01-11 19:05:22,022][454600] Fps is (10 sec: 229391.6, 60 sec: 231569.5, 300 sec: 171238.5). Total num frames: 18087936. Throughput: 0: 230922.0. Samples: 18053120. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:05:22,023][454600] Avg episode reward: [(0, '23684.512')]
+[2023-01-11 19:05:22,034][454600] Saving ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/02_v083_brax_basic_benchmark_see_2322090_env_humanoid_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000002760_18087936.pth...
+[2023-01-11 19:05:22,076][454600] Saving new best policy, reward=23684.512!
+[2023-01-11 19:05:27,022][454600] Fps is (10 sec: 229354.9, 60 sec: 230467.9, 300 sec: 173569.8). Total num frames: 19202048. Throughput: 0: 231064.2. Samples: 18737152. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:05:27,023][454600] Avg episode reward: [(0, '24190.295')]
+[2023-01-11 19:05:27,029][454600] Saving new best policy, reward=24190.295!
+[2023-01-11 19:05:32,025][454600] Fps is (10 sec: 229313.7, 60 sec: 230630.4, 300 sec: 176262.3). Total num frames: 20381696. Throughput: 0: 231007.3. Samples: 20119552. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:05:32,025][454600] Avg episode reward: [(0, '24498.295')]
+[2023-01-11 19:05:32,038][454600] Saving new best policy, reward=24498.295!
+[2023-01-11 19:05:37,067][454600] Fps is (10 sec: 234886.9, 60 sec: 231388.1, 300 sec: 178673.5). Total num frames: 21561344. Throughput: 0: 230201.2. Samples: 21495808. Policy #0 lag: (min: 2.0, avg: 2.0, max: 2.0)
+[2023-01-11 19:05:37,067][454600] Avg episode reward: [(0, '25790.773')]
+[2023-01-11 19:05:37,069][454600] Saving new best policy, reward=25790.773!
+[2023-01-11 19:05:42,024][454600] Fps is (10 sec: 229405.2, 60 sec: 230466.2, 300 sec: 180491.8). Total num frames: 22675456. Throughput: 0: 231566.0. Samples: 22216704. Policy #0 lag: (min: 2.0, avg: 2.0, max: 2.0)
+[2023-01-11 19:05:42,024][454600] Avg episode reward: [(0, '25062.875')]
+[2023-01-11 19:05:47,024][454600] Fps is (10 sec: 230367.0, 60 sec: 230672.4, 300 sec: 182613.7). Total num frames: 23855104. Throughput: 0: 231212.4. Samples: 23592960. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:05:47,024][454600] Avg episode reward: [(0, '25982.887')]
+[2023-01-11 19:05:47,030][454600] Saving new best policy, reward=25982.887!
+[2023-01-11 19:05:52,021][454600] Fps is (10 sec: 229444.6, 60 sec: 230478.6, 300 sec: 184100.0). Total num frames: 24969216. Throughput: 0: 230342.0. Samples: 24969216. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:05:52,021][454600] Avg episode reward: [(0, '25676.555')]
+[2023-01-11 19:05:57,021][454600] Fps is (10 sec: 229438.2, 60 sec: 230482.8, 300 sec: 185942.5). Total num frames: 26148864. Throughput: 0: 230887.8. Samples: 25659392. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:05:57,022][454600] Avg episode reward: [(0, '24945.438')]
+[2023-01-11 19:06:02,021][454600] Fps is (10 sec: 235920.5, 60 sec: 230762.8, 300 sec: 187658.6). Total num frames: 27328512. Throughput: 0: 231152.5. Samples: 27066368. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:06:02,022][454600] Avg episode reward: [(0, '25933.258')]
+[2023-01-11 19:06:07,021][454600] Fps is (10 sec: 229375.9, 60 sec: 230474.8, 300 sec: 188825.9). Total num frames: 28442624. Throughput: 0: 230884.3. Samples: 28442624. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:06:07,021][454600] Avg episode reward: [(0, '26500.414')]
+[2023-01-11 19:06:07,028][454600] Saving new best policy, reward=26500.414!
+[2023-01-11 19:06:12,054][454600] Fps is (10 sec: 228615.9, 60 sec: 230347.7, 300 sec: 190298.5). Total num frames: 29622272. Throughput: 0: 230213.4. Samples: 29104128. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:06:12,055][454600] Avg episode reward: [(0, '27414.232')]
+[2023-01-11 19:06:12,063][454600] Saving new best policy, reward=27414.232!
+[2023-01-11 19:06:17,022][454600] Fps is (10 sec: 229354.4, 60 sec: 230466.2, 300 sec: 191349.3). Total num frames: 30736384. Throughput: 0: 230119.7. Samples: 30474240. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:06:17,022][454600] Avg episode reward: [(0, '28162.598')]
+[2023-01-11 19:06:17,028][454600] Saving new best policy, reward=28162.598!
+[2023-01-11 19:06:22,022][454600] Fps is (10 sec: 223541.1, 60 sec: 229376.2, 300 sec: 192299.1). Total num frames: 31850496. Throughput: 0: 230332.1. Samples: 31850496. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:06:22,023][454600] Avg episode reward: [(0, '28589.760')]
+[2023-01-11 19:06:22,034][454600] Saving new best policy, reward=28589.760!
+[2023-01-11 19:06:27,032][454600] Fps is (10 sec: 222594.8, 60 sec: 229338.4, 300 sec: 193182.2). Total num frames: 32964608. Throughput: 0: 227239.8. Samples: 32444416. Policy #0 lag: (min: 5.0, avg: 5.0, max: 5.0)
+[2023-01-11 19:06:27,033][454600] Avg episode reward: [(0, '28386.152')]
+[2023-01-11 19:06:32,024][454600] Fps is (10 sec: 222781.7, 60 sec: 228287.3, 300 sec: 194035.0). Total num frames: 34078720. Throughput: 0: 227690.4. Samples: 33839104. Policy #0 lag: (min: 5.0, avg: 5.0, max: 5.0)
+[2023-01-11 19:06:32,024][454600] Avg episode reward: [(0, '29366.443')]
+[2023-01-11 19:06:32,029][454600] Saving new best policy, reward=29366.443!
+[2023-01-11 19:06:37,021][454600] Fps is (10 sec: 229635.4, 60 sec: 228458.3, 300 sec: 195198.1). Total num frames: 35258368. Throughput: 0: 227782.3. Samples: 35219456. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:06:37,021][454600] Avg episode reward: [(0, '29529.770')]
+[2023-01-11 19:06:37,028][454600] Saving new best policy, reward=29529.770!
+[2023-01-11 19:06:42,025][454600] Fps is (10 sec: 229348.6, 60 sec: 228278.0, 300 sec: 195937.5). Total num frames: 36372480. Throughput: 0: 227579.7. Samples: 35901440. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:06:42,026][454600] Avg episode reward: [(0, '28905.598')]
+[2023-01-11 19:06:47,083][454600] Fps is (10 sec: 227954.9, 60 sec: 228057.6, 300 sec: 196926.6). Total num frames: 37552128. Throughput: 0: 226423.9. Samples: 37269504. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:06:47,084][454600] Avg episode reward: [(0, '30232.330')]
+[2023-01-11 19:06:47,086][454600] Saving new best policy, reward=30232.330!
+[2023-01-11 19:06:52,023][454600] Fps is (10 sec: 222876.0, 60 sec: 227183.5, 300 sec: 197314.2). Total num frames: 38600704. Throughput: 0: 225726.0. Samples: 38600704. Policy #0 lag: (min: 2.0, avg: 2.0, max: 2.0)
+[2023-01-11 19:06:52,023][454600] Avg episode reward: [(0, '30448.795')]
+[2023-01-11 19:06:52,123][454600] Saving new best policy, reward=30448.795!
+[2023-01-11 19:06:57,075][454600] Fps is (10 sec: 223000.7, 60 sec: 226986.4, 300 sec: 198224.8). Total num frames: 39780352. Throughput: 0: 225585.0. Samples: 39260160. Policy #0 lag: (min: 2.0, avg: 2.0, max: 2.0)
+[2023-01-11 19:06:57,076][454600] Avg episode reward: [(0, '31469.381')]
+[2023-01-11 19:06:57,078][454600] Saving new best policy, reward=31469.381!
+[2023-01-11 19:07:02,025][454600] Fps is (10 sec: 229324.8, 60 sec: 226084.3, 300 sec: 198871.3). Total num frames: 40894464. Throughput: 0: 225719.5. Samples: 40632320. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:07:02,025][454600] Avg episode reward: [(0, '30854.137')]
+[2023-01-11 19:07:07,024][454600] Fps is (10 sec: 230546.7, 60 sec: 227178.5, 300 sec: 199751.5). Total num frames: 42074112. Throughput: 0: 226315.5. Samples: 42035200. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:07:07,025][454600] Avg episode reward: [(0, '29817.902')]
+[2023-01-11 19:07:12,104][454600] Fps is (10 sec: 234082.2, 60 sec: 227003.7, 300 sec: 200516.5). Total num frames: 43253760. Throughput: 0: 228192.7. Samples: 42729472. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:07:12,104][454600] Avg episode reward: [(0, '31047.578')]
+[2023-01-11 19:07:17,024][454600] Fps is (10 sec: 229380.1, 60 sec: 227182.7, 300 sec: 201094.4). Total num frames: 44367872. Throughput: 0: 228146.1. Samples: 44105728. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
+[2023-01-11 19:07:17,025][454600] Avg episode reward: [(0, '31452.555')]
+[2023-01-11 19:07:22,021][454600] Fps is (10 sec: 224671.6, 60 sec: 227193.5, 300 sec: 201578.2). Total num frames: 45481984. Throughput: 0: 227961.1. Samples: 45477888. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
+[2023-01-11 19:07:22,022][454600] Avg episode reward: [(0, '31025.244')]
+[2023-01-11 19:07:22,033][454600] Saving ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/02_v083_brax_basic_benchmark_see_2322090_env_humanoid_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000006940_45481984.pth...
+[2023-01-11 19:07:27,030][454600] Fps is (10 sec: 222694.8, 60 sec: 227199.7, 300 sec: 202031.5). Total num frames: 46596096. Throughput: 0: 226166.3. Samples: 46080000. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
+[2023-01-11 19:07:27,030][454600] Avg episode reward: [(0, '28707.133')]
+[2023-01-11 19:07:32,024][454600] Fps is (10 sec: 222775.5, 60 sec: 227192.4, 300 sec: 202478.0). Total num frames: 47710208. Throughput: 0: 226671.7. Samples: 47456256. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:07:32,024][454600] Avg episode reward: [(0, '30637.068')]
+[2023-01-11 19:07:37,025][454600] Fps is (10 sec: 229488.7, 60 sec: 227175.5, 300 sec: 203172.0). Total num frames: 48889856. Throughput: 0: 227953.7. Samples: 48859136. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:07:37,026][454600] Avg episode reward: [(0, '31947.164')]
+[2023-01-11 19:07:37,028][454600] Saving new best policy, reward=31947.164!
+[2023-01-11 19:07:42,024][454600] Fps is (10 sec: 222807.4, 60 sec: 226102.1, 300 sec: 203305.7). Total num frames: 49938432. Throughput: 0: 226400.1. Samples: 49436672. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:07:42,025][454600] Avg episode reward: [(0, '30685.115')]
+[2023-01-11 19:07:47,024][454600] Fps is (10 sec: 216277.6, 60 sec: 225226.7, 300 sec: 203694.9). Total num frames: 51052544. Throughput: 0: 226465.2. Samples: 50823168. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:07:47,025][454600] Avg episode reward: [(0, '32077.797')]
+[2023-01-11 19:07:47,030][454600] Saving new best policy, reward=32077.797!
+[2023-01-11 19:07:52,048][454600] Fps is (10 sec: 222283.4, 60 sec: 226001.8, 300 sec: 204049.8). Total num frames: 52166656. Throughput: 0: 223703.1. Samples: 52107264. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:07:52,049][454600] Avg episode reward: [(0, '31880.949')]
+[2023-01-11 19:07:57,023][454600] Fps is (10 sec: 222862.4, 60 sec: 225203.4, 300 sec: 204430.1). Total num frames: 53280768. Throughput: 0: 224501.3. Samples: 52813824. Policy #0 lag: (min: 6.0, avg: 6.0, max: 6.0)
+[2023-01-11 19:07:57,023][454600] Avg episode reward: [(0, '32334.512')]
+[2023-01-11 19:07:57,029][454600] Saving new best policy, reward=32334.512!
+[2023-01-11 19:08:02,020][454600] Fps is (10 sec: 216877.2, 60 sec: 223931.1, 300 sec: 204531.4). Total num frames: 54329344. Throughput: 0: 222112.1. Samples: 54099968. Policy #0 lag: (min: 6.0, avg: 6.0, max: 6.0)
+[2023-01-11 19:08:02,021][454600] Avg episode reward: [(0, '31227.248')]
+[2023-01-11 19:08:07,021][454600] Fps is (10 sec: 222869.4, 60 sec: 223928.3, 300 sec: 205111.4). Total num frames: 55508992. Throughput: 0: 222189.7. Samples: 55476224. Policy #0 lag: (min: 6.0, avg: 6.0, max: 6.0)
+[2023-01-11 19:08:07,021][454600] Avg episode reward: [(0, '32576.342')]
+[2023-01-11 19:08:07,028][454600] Saving new best policy, reward=32576.342!
+[2023-01-11 19:08:12,023][454600] Fps is (10 sec: 229315.5, 60 sec: 223122.5, 300 sec: 205430.8). Total num frames: 56623104. Throughput: 0: 223948.1. Samples: 56156160. Policy #0 lag: (min: 2.0, avg: 2.0, max: 2.0)
+[2023-01-11 19:08:12,024][454600] Avg episode reward: [(0, '32309.672')]
+[2023-01-11 19:08:17,070][454600] Fps is (10 sec: 228250.9, 60 sec: 223743.9, 300 sec: 205939.8). Total num frames: 57802752. Throughput: 0: 223457.3. Samples: 57522176. Policy #0 lag: (min: 2.0, avg: 2.0, max: 2.0)
+[2023-01-11 19:08:17,070][454600] Avg episode reward: [(0, '32164.748')]
+[2023-01-11 19:08:22,022][454600] Fps is (10 sec: 229412.0, 60 sec: 223914.5, 300 sec: 206270.2). Total num frames: 58916864. Throughput: 0: 223294.2. Samples: 58906624. Policy #0 lag: (min: 2.0, avg: 2.0, max: 2.0)
+[2023-01-11 19:08:22,022][454600] Avg episode reward: [(0, '30573.240')]
+[2023-01-11 19:08:27,087][454600] Fps is (10 sec: 228991.7, 60 sec: 224793.9, 300 sec: 206734.2). Total num frames: 60096512. Throughput: 0: 224922.6. Samples: 59572224. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0)
+[2023-01-11 19:08:27,087][454600] Avg episode reward: [(0, '31910.117')]
+[2023-01-11 19:08:32,023][454600] Fps is (10 sec: 229352.0, 60 sec: 225010.7, 300 sec: 216814.9). Total num frames: 61210624. Throughput: 0: 225016.5. Samples: 60948480. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0)
+[2023-01-11 19:08:32,023][454600] Avg episode reward: [(0, '33219.871')]
+[2023-01-11 19:08:32,034][454600] Saving new best policy, reward=33219.871!
+[2023-01-11 19:08:37,100][454600] Fps is (10 sec: 229064.5, 60 sec: 224724.7, 300 sec: 227418.8). Total num frames: 62390272. Throughput: 0: 226794.1. Samples: 62324736. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0)
+[2023-01-11 19:08:37,101][454600] Avg episode reward: [(0, '32330.926')]
+[2023-01-11 19:08:42,021][454600] Fps is (10 sec: 229421.0, 60 sec: 226112.9, 300 sec: 227405.2). Total num frames: 63504384. Throughput: 0: 226792.4. Samples: 63019008. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0)
+[2023-01-11 19:08:42,021][454600] Avg episode reward: [(0, '32682.035')]
+[2023-01-11 19:08:47,072][454600] Fps is (10 sec: 230036.3, 60 sec: 227013.4, 300 sec: 227520.5). Total num frames: 64684032. Throughput: 0: 228706.7. Samples: 64403456. Policy #0 lag: (min: 2.0, avg: 2.0, max: 2.0)
+[2023-01-11 19:08:47,072][454600] Avg episode reward: [(0, '33019.711')]
+[2023-01-11 19:08:52,024][454600] Fps is (10 sec: 229291.9, 60 sec: 227283.2, 300 sec: 227479.7). Total num frames: 65798144. Throughput: 0: 229311.7. Samples: 65796096. Policy #0 lag: (min: 2.0, avg: 2.0, max: 2.0)
+[2023-01-11 19:08:52,025][454600] Avg episode reward: [(0, '33007.113')]
+[2023-01-11 19:08:57,103][454600] Fps is (10 sec: 228646.2, 60 sec: 227976.8, 300 sec: 227566.6). Total num frames: 66977792. Throughput: 0: 228421.9. Samples: 66453504. Policy #0 lag: (min: 2.0, avg: 2.0, max: 2.0)
+[2023-01-11 19:08:57,104][454600] Avg episode reward: [(0, '33615.703')]
+[2023-01-11 19:08:57,107][454600] Saving new best policy, reward=33615.703!
+[2023-01-11 19:09:02,026][454600] Fps is (10 sec: 229327.9, 60 sec: 229353.3, 300 sec: 228598.5). Total num frames: 68091904. Throughput: 0: 229279.0. Samples: 67829760. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
+[2023-01-11 19:09:02,027][454600] Avg episode reward: [(0, '33588.613')]
+[2023-01-11 19:09:07,024][454600] Fps is (10 sec: 224609.2, 60 sec: 228271.1, 300 sec: 228374.6). Total num frames: 69206016. Throughput: 0: 228863.2. Samples: 69206016. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
+[2023-01-11 19:09:07,024][454600] Avg episode reward: [(0, '33701.699')]
+[2023-01-11 19:09:07,031][454600] Saving new best policy, reward=33701.699!
+[2023-01-11 19:09:12,079][454600] Fps is (10 sec: 228179.7, 60 sec: 229163.2, 300 sec: 228333.4). Total num frames: 70385664. Throughput: 0: 228778.8. Samples: 69865472. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
+[2023-01-11 19:09:12,079][454600] Avg episode reward: [(0, '34000.129')]
+[2023-01-11 19:09:12,086][454600] Saving new best policy, reward=34000.129!
+[2023-01-11 19:09:17,023][454600] Fps is (10 sec: 229400.4, 60 sec: 228462.8, 300 sec: 228155.4). Total num frames: 71499776. Throughput: 0: 228646.4. Samples: 71237632. Policy #0 lag: (min: 3.0, avg: 3.0, max: 3.0)
+[2023-01-11 19:09:17,023][454600] Avg episode reward: [(0, '34372.684')]
+[2023-01-11 19:09:17,030][454600] Saving new best policy, reward=34372.684!
+[2023-01-11 19:09:22,024][454600] Fps is (10 sec: 224056.5, 60 sec: 228275.4, 300 sec: 227930.7). Total num frames: 72613888. Throughput: 0: 228855.0. Samples: 72605696. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:09:22,024][454600] Avg episode reward: [(0, '33211.176')]
+[2023-01-11 19:09:22,032][454600] Saving ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/02_v083_brax_basic_benchmark_see_2322090_env_humanoid_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000011080_72613888.pth...
+[2023-01-11 19:09:22,100][454600] Removing ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/02_v083_brax_basic_benchmark_see_2322090_env_humanoid_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000002760_18087936.pth
+[2023-01-11 19:09:27,024][454600] Fps is (10 sec: 222802.5, 60 sec: 227429.9, 300 sec: 227743.2). Total num frames: 73728000. Throughput: 0: 227767.3. Samples: 73269248. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:09:27,024][454600] Avg episode reward: [(0, '34268.336')]
+[2023-01-11 19:09:32,022][454600] Fps is (10 sec: 229424.9, 60 sec: 228287.4, 300 sec: 227932.2). Total num frames: 74907648. Throughput: 0: 227854.0. Samples: 74645504. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:09:32,022][454600] Avg episode reward: [(0, '33945.703')]
+[2023-01-11 19:09:37,023][454600] Fps is (10 sec: 229387.8, 60 sec: 227483.5, 300 sec: 227709.6). Total num frames: 76021760. Throughput: 0: 227242.3. Samples: 76021760. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:09:37,024][454600] Avg episode reward: [(0, '32851.797')]
+[2023-01-11 19:09:42,042][454600] Fps is (10 sec: 228909.9, 60 sec: 228202.5, 300 sec: 227736.6). Total num frames: 77201408. Throughput: 0: 227775.9. Samples: 76689408. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:09:42,043][454600] Avg episode reward: [(0, '33755.750')]
+[2023-01-11 19:09:47,023][454600] Fps is (10 sec: 229373.2, 60 sec: 227374.1, 300 sec: 227709.7). Total num frames: 78315520. Throughput: 0: 227434.3. Samples: 78063616. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:09:47,024][454600] Avg episode reward: [(0, '34199.566')]
+[2023-01-11 19:09:52,048][454600] Fps is (10 sec: 229235.7, 60 sec: 228193.2, 300 sec: 227691.7). Total num frames: 79495168. Throughput: 0: 227296.9. Samples: 79439872. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:09:52,049][454600] Avg episode reward: [(0, '34146.453')]
+[2023-01-11 19:09:57,024][454600] Fps is (10 sec: 229366.0, 60 sec: 227493.6, 300 sec: 227544.5). Total num frames: 80609280. Throughput: 0: 228289.8. Samples: 80125952. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:09:57,024][454600] Avg episode reward: [(0, '33985.309')]
+[2023-01-11 19:10:02,023][454600] Fps is (10 sec: 223382.2, 60 sec: 227204.2, 300 sec: 227487.3). Total num frames: 81723392. Throughput: 0: 227645.7. Samples: 81481728. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:10:02,024][454600] Avg episode reward: [(0, '34004.363')]
+[2023-01-11 19:10:07,025][454600] Fps is (10 sec: 229351.9, 60 sec: 228280.2, 300 sec: 227486.1). Total num frames: 82903040. Throughput: 0: 227914.0. Samples: 82862080. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:10:07,026][454600] Avg episode reward: [(0, '34364.102')]
+[2023-01-11 19:10:12,025][454600] Fps is (10 sec: 229339.5, 60 sec: 227396.9, 300 sec: 227485.1). Total num frames: 84017152. Throughput: 0: 228643.4. Samples: 83558400. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:10:12,025][454600] Avg episode reward: [(0, '34041.414')]
+[2023-01-11 19:10:17,024][454600] Fps is (10 sec: 229386.7, 60 sec: 228277.9, 300 sec: 227485.9). Total num frames: 85196800. Throughput: 0: 228633.7. Samples: 84934656. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:10:17,025][454600] Avg episode reward: [(0, '33549.828')]
+[2023-01-11 19:10:18,676][454600] Early stopping after 2 epochs (4 sgd steps), loss delta 0.0000007
+[2023-01-11 19:10:22,109][454600] Fps is (10 sec: 233959.5, 60 sec: 229051.2, 300 sec: 227642.9). Total num frames: 86376448. Throughput: 0: 228213.7. Samples: 86310912. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:10:22,109][454600] Avg episode reward: [(0, '35236.949')]
+[2023-01-11 19:10:22,116][454600] Saving new best policy, reward=35236.949!
+[2023-01-11 19:10:27,025][454600] Fps is (10 sec: 229361.7, 60 sec: 229371.2, 300 sec: 227487.5). Total num frames: 87490560. Throughput: 0: 228734.1. Samples: 86978560. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:10:27,026][454600] Avg episode reward: [(0, '32445.162')]
+[2023-01-11 19:10:32,023][454600] Fps is (10 sec: 224758.3, 60 sec: 228279.6, 300 sec: 227299.3). Total num frames: 88604672. Throughput: 0: 228787.7. Samples: 88358912. Policy #0 lag: (min: 9.0, avg: 9.0, max: 9.0)
+[2023-01-11 19:10:32,023][454600] Avg episode reward: [(0, '34647.660')]
+[2023-01-11 19:10:37,031][454600] Fps is (10 sec: 229240.0, 60 sec: 229346.6, 300 sec: 227482.0). Total num frames: 89784320. Throughput: 0: 228735.0. Samples: 89729024. Policy #0 lag: (min: 8.0, avg: 8.0, max: 8.0)
+[2023-01-11 19:10:37,032][454600] Avg episode reward: [(0, '33732.051')]
+[2023-01-11 19:10:42,022][454600] Fps is (10 sec: 229395.1, 60 sec: 228360.3, 300 sec: 227266.8). Total num frames: 90898432. Throughput: 0: 229067.3. Samples: 90433536. Policy #0 lag: (min: 8.0, avg: 8.0, max: 8.0)
+[2023-01-11 19:10:42,022][454600] Avg episode reward: [(0, '34734.863')]
+[2023-01-11 19:10:47,051][454600] Fps is (10 sec: 228906.3, 60 sec: 229268.6, 300 sec: 227463.8). Total num frames: 92078080. Throughput: 0: 229094.7. Samples: 91797504. Policy #0 lag: (min: 8.0, avg: 8.0, max: 8.0)
+[2023-01-11 19:10:47,052][454600] Avg episode reward: [(0, '35127.297')]
+[2023-01-11 19:10:52,104][454600] Fps is (10 sec: 234005.2, 60 sec: 229162.1, 300 sec: 227423.5). Total num frames: 93257728. Throughput: 0: 229154.5. Samples: 93192192. Policy #0 lag: (min: 8.0, avg: 8.0, max: 8.0)
+[2023-01-11 19:10:52,105][454600] Avg episode reward: [(0, '34548.809')]
+[2023-01-11 19:10:57,024][454600] Fps is (10 sec: 230008.8, 60 sec: 229375.4, 300 sec: 227263.2). Total num frames: 94371840. Throughput: 0: 229334.0. Samples: 93878272. Policy #0 lag: (min: 8.0, avg: 8.0, max: 8.0)
+[2023-01-11 19:10:57,024][454600] Avg episode reward: [(0, '34352.727')]
+[2023-01-11 19:11:02,022][454600] Fps is (10 sec: 224670.1, 60 sec: 229380.5, 300 sec: 227264.7). Total num frames: 95485952. Throughput: 0: 229070.3. Samples: 95242240. Policy #0 lag: (min: 8.0, avg: 8.0, max: 8.0)
+[2023-01-11 19:11:02,022][454600] Avg episode reward: [(0, '33813.215')]
+[2023-01-11 19:11:07,023][454600] Fps is (10 sec: 229401.0, 60 sec: 229383.6, 300 sec: 227289.7). Total num frames: 96665600. Throughput: 0: 229587.1. Samples: 96622592. Policy #0 lag: (min: 8.0, avg: 8.0, max: 8.0)
+[2023-01-11 19:11:07,023][454600] Avg episode reward: [(0, '34654.340')]
+[2023-01-11 19:11:12,023][454600] Fps is (10 sec: 229344.1, 60 sec: 229381.2, 300 sec: 227264.4). Total num frames: 97779712. Throughput: 0: 229840.1. Samples: 97320960. Policy #0 lag: (min: 8.0, avg: 8.0, max: 8.0)
+[2023-01-11 19:11:12,024][454600] Avg episode reward: [(0, '35312.879')]
+[2023-01-11 19:11:12,035][454600] Saving new best policy, reward=35312.879!
+[2023-01-11 19:11:17,024][454600] Fps is (10 sec: 229354.7, 60 sec: 229378.3, 300 sec: 227486.3). Total num frames: 98959360. Throughput: 0: 229734.5. Samples: 98697216. Policy #0 lag: (min: 8.0, avg: 8.0, max: 8.0)
+[2023-01-11 19:11:17,024][454600] Avg episode reward: [(0, '34817.219')]
+[2023-01-11 19:11:21,772][454600] Saving ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/02_v083_brax_basic_benchmark_see_2322090_env_humanoid_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000015264_100073472.pth...
+[2023-01-11 19:11:21,814][454600] Removing ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/02_v083_brax_basic_benchmark_see_2322090_env_humanoid_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000006940_45481984.pth
+[2023-01-11 19:11:21,818][454600] Stopping InferenceWorker_p0-w0...
+[2023-01-11 19:11:21,818][454600] Stopping Batcher_0...
+[2023-01-11 19:11:21,819][454600] Stopping RolloutWorker_w0...
+[2023-01-11 19:11:21,819][454600] Component InferenceWorker_p0-w0 stopped!
+[2023-01-11 19:11:21,819][454600] Saving ./train_dir/v083_brax_basic_benchmark/v083_brax_basic_benchmark_slurm/02_v083_brax_basic_benchmark_see_2322090_env_humanoid_u.rnn_False_n.epo_5/checkpoint_p0/checkpoint_000015264_100073472.pth...
+[2023-01-11 19:11:21,863][454600] Stopping LearnerWorker_p0...
+[2023-01-11 19:11:21,864][454600] Component Batcher_0 stopped!
+[2023-01-11 19:11:21,864][454600] Component RolloutWorker_w0 stopped!
+[2023-01-11 19:11:21,864][454600] Component LearnerWorker_p0 stopped!
+[2023-01-11 19:11:21,864][454600] Batcher 0 profile tree view:
+batching: 0.3849, releasing_batches: 0.0669
+[2023-01-11 19:11:21,865][454600] InferenceWorker_p0-w0 profile tree view:
+update_model: 0.4926
+one_step: 0.0013
+  handle_policy_step: 65.3049
+    deserialize: 0.5337, stack: 0.0729, obs_to_device_normalize: 11.5510, forward: 41.6508, prepare_outputs: 7.0596, send_messages: 0.9048
+[2023-01-11 19:11:21,865][454600] Learner 0 profile tree view:
+misc: 0.0056, prepare_batch: 5.9925
+train: 138.8240
+  epoch_init: 0.0650, minibatch_init: 1.0265, losses_postprocess: 48.8531, kl_divergence: 5.8442, after_optimizer: 0.4433
+  calculate_losses: 18.3675
+    losses_init: 0.0348, forward_head: 3.0278, bptt_initial: 0.1284, bptt: 0.1504, tail: 8.9931, advantages_returns: 1.1807, losses: 3.5237
+  update: 62.2542
+    clip: 9.0029
+[2023-01-11 19:11:21,865][454600] RolloutWorker_w0 profile tree view:
+wait_for_trajectories: 0.0862, enqueue_policy_requests: 6.0452, process_policy_outputs: 3.8395, env_step: 203.9272, finalize_trajectories: 0.1686, complete_rollouts: 0.0726
+post_env_step: 21.0621
+  process_env_step: 8.4224
+[2023-01-11 19:11:21,865][454600] Loop Runner_EvtLoop terminating...
+[2023-01-11 19:11:21,866][454600] Runner profile tree view:
+main_loop: 474.1841
+[2023-01-11 19:11:21,866][454600] Collected {0: 100073472}, FPS: 211043.5