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.1679842604.debian

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f6ef1512af92c5d2ed23977c5aab37d1d0319ca7892eaf63fe1a67a4c4f2068
+size 40

.summary/0/events.out.tfevents.1679842606.debian

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9066e75831695739280e51dcf4b1d3812e95973932e7164324f974cee36a9c3
+size 40

.summary/0/events.out.tfevents.1679842923.debian

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9483f0cb5f8d398273b504a82631c440dcb02534520e2885ffe9b4473d5f55d8
+size 22503

.summary/0/events.out.tfevents.1679843004.debian

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:503248f8d2fe85b030ef4cbb57c38bd31026c4cd783dade323d07da2d17cc9cd
+size 110

.summary/0/events.out.tfevents.1679843115.debian

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b74bb9649a10bcc2e6df5aa27595f97da062b1153b309f6d253e9c619b6e218
+size 133552

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: doom_health_gathering_supreme
+      type: doom_health_gathering_supreme
+    metrics:
+    - type: mean_reward
+      value: 13.66 +/- 5.39
+      name: mean_reward
+      verified: false
+---
+
+A(n) **APPO** model trained on the **doom_health_gathering_supreme** 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 sergey-antonov/rl_course_vizdoom_health_gathering_supreme
+```
+
+    
+## Using the model
+
+To run the model after download, use the `enjoy` script corresponding to this environment:
+```
+python -m <path.to.enjoy.module> --algo=APPO --env=doom_health_gathering_supreme --train_dir=./train_dir --experiment=rl_course_vizdoom_health_gathering_supreme
+```
+
+
+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 <path.to.train.module> --algo=APPO --env=doom_health_gathering_supreme --train_dir=./train_dir --experiment=rl_course_vizdoom_health_gathering_supreme --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_000000824_3375104_reward_24.707.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b14fb43b0b71abf0d0240184b0c9116d98942fe224819d6bd02128ecc4ccbb1
+size 34928806

checkpoint_p0/checkpoint_000000824_3375104.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10245483978c1dde455e59fc7b2b8c8547b64eb2ed29c3c7b09e20037e82220a
+size 34929220

checkpoint_p0/checkpoint_000000978_4005888.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c920264d5adf2eab8d0eb5c437fdd462ea7393a3804d5aae33d3433eb18f5ac1
+size 34929220

config.json

@@ -0,0 +1,142 @@
+{
+  "help": false,
+  "algo": "APPO",
+  "env": "doom_health_gathering_supreme",
+  "experiment": "default_experiment",
+  "train_dir": "/home/hit/nnet/hf/rl/unit9/train_dir",
+  "restart_behavior": "resume",
+  "device": "gpu",
+  "seed": null,
+  "num_policies": 1,
+  "async_rl": true,
+  "serial_mode": false,
+  "batched_sampling": false,
+  "num_batches_to_accumulate": 2,
+  "worker_num_splits": 2,
+  "policy_workers_per_policy": 1,
+  "max_policy_lag": 1000,
+  "num_workers": 8,
+  "num_envs_per_worker": 4,
+  "batch_size": 1024,
+  "num_batches_per_epoch": 1,
+  "num_epochs": 1,
+  "rollout": 32,
+  "recurrence": 32,
+  "shuffle_minibatches": false,
+  "gamma": 0.99,
+  "reward_scale": 1.0,
+  "reward_clip": 1000.0,
+  "value_bootstrap": false,
+  "normalize_returns": true,
+  "exploration_loss_coeff": 0.001,
+  "value_loss_coeff": 0.5,
+  "kl_loss_coeff": 0.0,
+  "exploration_loss": "symmetric_kl",
+  "gae_lambda": 0.95,
+  "ppo_clip_ratio": 0.1,
+  "ppo_clip_value": 0.2,
+  "with_vtrace": false,
+  "vtrace_rho": 1.0,
+  "vtrace_c": 1.0,
+  "optimizer": "adam",
+  "adam_eps": 1e-06,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.999,
+  "max_grad_norm": 4.0,
+  "learning_rate": 0.0001,
+  "lr_schedule": "constant",
+  "lr_schedule_kl_threshold": 0.008,
+  "lr_adaptive_min": 1e-06,
+  "lr_adaptive_max": 0.01,
+  "obs_subtract_mean": 0.0,
+  "obs_scale": 255.0,
+  "normalize_input": true,
+  "normalize_input_keys": null,
+  "decorrelate_experience_max_seconds": 0,
+  "decorrelate_envs_on_one_worker": true,
+  "actor_worker_gpus": [],
+  "set_workers_cpu_affinity": true,
+  "force_envs_single_thread": false,
+  "default_niceness": 0,
+  "log_to_file": true,
+  "experiment_summaries_interval": 10,
+  "flush_summaries_interval": 30,
+  "stats_avg": 100,
+  "summaries_use_frameskip": true,
+  "heartbeat_interval": 20,
+  "heartbeat_reporting_interval": 600,
+  "train_for_env_steps": 4000000,
+  "train_for_seconds": 10000000000,
+  "save_every_sec": 120,
+  "keep_checkpoints": 2,
+  "load_checkpoint_kind": "latest",
+  "save_milestones_sec": -1,
+  "save_best_every_sec": 5,
+  "save_best_metric": "reward",
+  "save_best_after": 100000,
+  "benchmark": false,
+  "encoder_mlp_layers": [
+    512,
+    512
+  ],
+  "encoder_conv_architecture": "convnet_simple",
+  "encoder_conv_mlp_layers": [
+    512
+  ],
+  "use_rnn": true,
+  "rnn_size": 512,
+  "rnn_type": "gru",
+  "rnn_num_layers": 1,
+  "decoder_mlp_layers": [],
+  "nonlinearity": "elu",
+  "policy_initialization": "orthogonal",
+  "policy_init_gain": 1.0,
+  "actor_critic_share_weights": true,
+  "adaptive_stddev": true,
+  "continuous_tanh_scale": 0.0,
+  "initial_stddev": 1.0,
+  "use_env_info_cache": false,
+  "env_gpu_actions": false,
+  "env_gpu_observations": true,
+  "env_frameskip": 4,
+  "env_framestack": 1,
+  "pixel_format": "CHW",
+  "use_record_episode_statistics": false,
+  "with_wandb": false,
+  "wandb_user": 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,
+  "num_agents": -1,
+  "num_humans": 0,
+  "num_bots": -1,
+  "start_bot_difficulty": null,
+  "timelimit": null,
+  "res_w": 128,
+  "res_h": 72,
+  "wide_aspect_ratio": false,
+  "eval_env_frameskip": 1,
+  "fps": 35,
+  "command_line": "--env=doom_health_gathering_supreme --num_workers=8 --num_envs_per_worker=4 --train_for_env_steps=4000000",
+  "cli_args": {
+    "env": "doom_health_gathering_supreme",
+    "num_workers": 8,
+    "num_envs_per_worker": 4,
+    "train_for_env_steps": 4000000
+  },
+  "git_hash": "unknown",
+  "git_repo_name": "not a git repository"
+}

replay.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:113ed927245d13f8646a38b71b2961005745c4661119af9e7484b725994ba5e5
+size 26803448

sf_log.txt

@@ -0,0 +1,1236 @@
+[2023-03-26 18:02:05,029][28680] Saving configuration to /home/hit/nnet/hf/rl/unit9/train_dir/default_experiment/config.json...
+[2023-03-26 18:02:05,029][28680] Rollout worker 0 uses device cpu
+[2023-03-26 18:02:05,029][28680] Rollout worker 1 uses device cpu
+[2023-03-26 18:02:05,029][28680] Rollout worker 2 uses device cpu
+[2023-03-26 18:02:05,029][28680] Rollout worker 3 uses device cpu
+[2023-03-26 18:02:05,030][28680] Rollout worker 4 uses device cpu
+[2023-03-26 18:02:05,030][28680] Rollout worker 5 uses device cpu
+[2023-03-26 18:02:05,030][28680] Rollout worker 6 uses device cpu
+[2023-03-26 18:02:05,030][28680] Rollout worker 7 uses device cpu
+[2023-03-26 18:02:05,063][28680] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:02:05,063][28680] InferenceWorker_p0-w0: min num requests: 2
+[2023-03-26 18:02:05,081][28680] Starting all processes...
+[2023-03-26 18:02:05,081][28680] Starting process learner_proc0
+[2023-03-26 18:02:05,978][28680] Starting all processes...
+[2023-03-26 18:02:05,980][28695] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:02:05,981][28695] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0
+[2023-03-26 18:02:05,981][28680] Starting process inference_proc0-0
+[2023-03-26 18:02:05,981][28680] Starting process rollout_proc0
+[2023-03-26 18:02:05,981][28680] Starting process rollout_proc1
+[2023-03-26 18:02:05,981][28680] Starting process rollout_proc2
+[2023-03-26 18:02:05,981][28680] Starting process rollout_proc3
+[2023-03-26 18:02:05,983][28680] Starting process rollout_proc4
+[2023-03-26 18:02:05,985][28680] Starting process rollout_proc5
+[2023-03-26 18:02:05,990][28695] Num visible devices: 1
+[2023-03-26 18:02:05,985][28680] Starting process rollout_proc6
+[2023-03-26 18:02:05,986][28680] Starting process rollout_proc7
+[2023-03-26 18:02:06,017][28695] Starting seed is not provided
+[2023-03-26 18:02:06,017][28695] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:02:06,017][28695] Initializing actor-critic model on device cuda:0
+[2023-03-26 18:02:06,018][28695] RunningMeanStd input shape: (3, 72, 128)
+[2023-03-26 18:02:06,018][28695] RunningMeanStd input shape: (1,)
+[2023-03-26 18:02:06,026][28695] ConvEncoder: input_channels=3
+[2023-03-26 18:02:06,137][28695] Conv encoder output size: 512
+[2023-03-26 18:02:06,137][28695] Policy head output size: 512
+[2023-03-26 18:02:06,162][28695] Created Actor Critic model with architecture:
+[2023-03-26 18:02:06,163][28695] 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)
+  )
+)
+[2023-03-26 18:02:07,157][28712] Worker 3 uses CPU cores [6, 7]
+[2023-03-26 18:02:07,190][28730] Worker 6 uses CPU cores [12, 13]
+[2023-03-26 18:02:07,193][28708] Worker 0 uses CPU cores [0, 1]
+[2023-03-26 18:02:07,200][28709] Worker 1 uses CPU cores [2, 3]
+[2023-03-26 18:02:07,256][28710] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:02:07,256][28710] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0
+[2023-03-26 18:02:07,266][28710] Num visible devices: 1
+[2023-03-26 18:02:07,330][28731] Worker 7 uses CPU cores [14, 15]
+[2023-03-26 18:02:07,358][28711] Worker 2 uses CPU cores [4, 5]
+[2023-03-26 18:02:07,465][28713] Worker 4 uses CPU cores [8, 9]
+[2023-03-26 18:02:07,498][28729] Worker 5 uses CPU cores [10, 11]
+[2023-03-26 18:02:08,648][28695] Using optimizer <class 'torch.optim.adam.Adam'>
+[2023-03-26 18:02:08,649][28695] No checkpoints found
+[2023-03-26 18:02:08,649][28695] Did not load from checkpoint, starting from scratch!
+[2023-03-26 18:02:08,649][28695] Initialized policy 0 weights for model version 0
+[2023-03-26 18:02:08,650][28695] LearnerWorker_p0 finished initialization!
+[2023-03-26 18:02:08,651][28695] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:02:08,759][28710] RunningMeanStd input shape: (3, 72, 128)
+[2023-03-26 18:02:08,760][28710] RunningMeanStd input shape: (1,)
+[2023-03-26 18:02:08,767][28710] ConvEncoder: input_channels=3
+[2023-03-26 18:02:08,842][28710] Conv encoder output size: 512
+[2023-03-26 18:02:08,842][28710] Policy head output size: 512
+[2023-03-26 18:02:09,444][28680] Inference worker 0-0 is ready!
+[2023-03-26 18:02:09,444][28680] All inference workers are ready! Signal rollout workers to start!
+[2023-03-26 18:02:09,485][28708] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:02:09,485][28709] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:02:09,485][28729] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:02:09,485][28711] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:02:09,485][28730] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:02:09,491][28731] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:02:09,492][28713] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:02:09,492][28712] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:02:09,634][28708] VizDoom game.init() threw an exception ViZDoomUnexpectedExitException('Controlled ViZDoom instance exited unexpectedly.'). Terminate process...
+[2023-03-26 18:02:09,634][28712] VizDoom game.init() threw an exception ViZDoomUnexpectedExitException('Controlled ViZDoom instance exited unexpectedly.'). Terminate process...
+[2023-03-26 18:02:09,634][28709] VizDoom game.init() threw an exception ViZDoomUnexpectedExitException('Controlled ViZDoom instance exited unexpectedly.'). Terminate process...
+[2023-03-26 18:02:09,634][28729] VizDoom game.init() threw an exception ViZDoomUnexpectedExitException('Controlled ViZDoom instance exited unexpectedly.'). Terminate process...
+[2023-03-26 18:02:09,635][28712] EvtLoop [rollout_proc3_evt_loop, process=rollout_proc3] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 228, in _game_init
+    self.game.init()
+vizdoom.vizdoom.ViZDoomUnexpectedExitException: Controlled ViZDoom instance exited unexpectedly.
+
+During handling of the above exception, another exception occurred:
+
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 379, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 84, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/multiplayer_stats.py", line 51, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 323, in reset
+    self._ensure_initialized()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 274, in _ensure_initialized
+    self.initialize()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 269, in initialize
+    self._game_init()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 244, in _game_init
+    raise EnvCriticalError()
+sample_factory.envs.env_utils.EnvCriticalError
+[2023-03-26 18:02:09,635][28708] EvtLoop [rollout_proc0_evt_loop, process=rollout_proc0] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 228, in _game_init
+    self.game.init()
+vizdoom.vizdoom.ViZDoomUnexpectedExitException: Controlled ViZDoom instance exited unexpectedly.
+
+During handling of the above exception, another exception occurred:
+
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 379, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 84, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/multiplayer_stats.py", line 51, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 323, in reset
+    self._ensure_initialized()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 274, in _ensure_initialized
+    self.initialize()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 269, in initialize
+    self._game_init()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 244, in _game_init
+    raise EnvCriticalError()
+sample_factory.envs.env_utils.EnvCriticalError
+[2023-03-26 18:02:09,635][28709] EvtLoop [rollout_proc1_evt_loop, process=rollout_proc1] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 228, in _game_init
+    self.game.init()
+vizdoom.vizdoom.ViZDoomUnexpectedExitException: Controlled ViZDoom instance exited unexpectedly.
+
+During handling of the above exception, another exception occurred:
+
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 379, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 84, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/multiplayer_stats.py", line 51, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 323, in reset
+    self._ensure_initialized()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 274, in _ensure_initialized
+    self.initialize()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 269, in initialize
+    self._game_init()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 244, in _game_init
+    raise EnvCriticalError()
+sample_factory.envs.env_utils.EnvCriticalError
+[2023-03-26 18:02:09,635][28729] EvtLoop [rollout_proc5_evt_loop, process=rollout_proc5] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 228, in _game_init
+    self.game.init()
+vizdoom.vizdoom.ViZDoomUnexpectedExitException: Controlled ViZDoom instance exited unexpectedly.
+
+During handling of the above exception, another exception occurred:
+
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 379, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 84, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/multiplayer_stats.py", line 51, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 323, in reset
+    self._ensure_initialized()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 274, in _ensure_initialized
+    self.initialize()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 269, in initialize
+    self._game_init()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 244, in _game_init
+    raise EnvCriticalError()
+sample_factory.envs.env_utils.EnvCriticalError
+[2023-03-26 18:02:09,638][28712] Unhandled exception  in evt loop rollout_proc3_evt_loop
+[2023-03-26 18:02:09,638][28708] Unhandled exception  in evt loop rollout_proc0_evt_loop
+[2023-03-26 18:02:09,638][28709] Unhandled exception  in evt loop rollout_proc1_evt_loop
+[2023-03-26 18:02:09,638][28729] Unhandled exception  in evt loop rollout_proc5_evt_loop
+[2023-03-26 18:02:09,838][28713] Decorrelating experience for 0 frames...
+[2023-03-26 18:02:09,839][28731] Decorrelating experience for 0 frames...
+[2023-03-26 18:02:09,847][28730] Decorrelating experience for 0 frames...
+[2023-03-26 18:02:09,847][28711] Decorrelating experience for 0 frames...
+[2023-03-26 18:02:10,023][28711] Decorrelating experience for 32 frames...
+[2023-03-26 18:02:10,023][28730] Decorrelating experience for 32 frames...
+[2023-03-26 18:02:10,026][28713] Decorrelating experience for 32 frames...
+[2023-03-26 18:02:10,222][28731] Decorrelating experience for 32 frames...
+[2023-03-26 18:02:10,236][28713] Decorrelating experience for 64 frames...
+[2023-03-26 18:02:10,237][28730] Decorrelating experience for 64 frames...
+[2023-03-26 18:02:10,238][28711] Decorrelating experience for 64 frames...
+[2023-03-26 18:02:10,412][28731] Decorrelating experience for 64 frames...
+[2023-03-26 18:02:10,428][28713] Decorrelating experience for 96 frames...
+[2023-03-26 18:02:10,430][28730] Decorrelating experience for 96 frames...
+[2023-03-26 18:02:10,608][28731] Decorrelating experience for 96 frames...
+[2023-03-26 18:02:10,631][28711] Decorrelating experience for 96 frames...
+[2023-03-26 18:02:11,738][28695] Signal inference workers to stop experience collection...
+[2023-03-26 18:02:11,741][28710] InferenceWorker_p0-w0: stopping experience collection
+[2023-03-26 18:02:12,665][28695] Signal inference workers to resume experience collection...
+[2023-03-26 18:02:12,666][28710] InferenceWorker_p0-w0: resuming experience collection
+[2023-03-26 18:02:13,122][28680] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 4096. Throughput: 0: nan. Samples: 2232. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0)
+[2023-03-26 18:02:13,122][28680] Avg episode reward: [(0, '3.506')]
+[2023-03-26 18:02:15,674][28710] Updated weights for policy 0, policy_version 10 (0.0258)
+[2023-03-26 18:02:18,122][28680] Fps is (10 sec: 13926.5, 60 sec: 13926.5, 300 sec: 13926.5). Total num frames: 73728. Throughput: 0: 1578.8. Samples: 10126. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-03-26 18:02:18,122][28680] Avg episode reward: [(0, '4.262')]
+[2023-03-26 18:02:18,512][28710] Updated weights for policy 0, policy_version 20 (0.0005)
+[2023-03-26 18:02:21,007][28710] Updated weights for policy 0, policy_version 30 (0.0006)
+[2023-03-26 18:02:23,122][28680] Fps is (10 sec: 14745.7, 60 sec: 14745.7, 300 sec: 14745.7). Total num frames: 151552. Throughput: 0: 3076.4. Samples: 32996. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-03-26 18:02:23,122][28680] Avg episode reward: [(0, '4.314')]
+[2023-03-26 18:02:23,126][28695] Saving new best policy, reward=4.314!
+[2023-03-26 18:02:23,987][28710] Updated weights for policy 0, policy_version 40 (0.0006)
+[2023-03-26 18:02:25,057][28680] Heartbeat connected on Batcher_0
+[2023-03-26 18:02:25,060][28680] Heartbeat connected on LearnerWorker_p0
+[2023-03-26 18:02:25,066][28680] Heartbeat connected on InferenceWorker_p0-w0
+[2023-03-26 18:02:25,071][28680] Heartbeat connected on RolloutWorker_w2
+[2023-03-26 18:02:25,076][28680] Heartbeat connected on RolloutWorker_w4
+[2023-03-26 18:02:25,078][28680] Heartbeat connected on RolloutWorker_w6
+[2023-03-26 18:02:25,081][28680] Heartbeat connected on RolloutWorker_w7
+[2023-03-26 18:02:26,686][28710] Updated weights for policy 0, policy_version 50 (0.0006)
+[2023-03-26 18:02:28,122][28680] Fps is (10 sec: 15155.1, 60 sec: 14745.6, 300 sec: 14745.6). Total num frames: 225280. Throughput: 0: 3488.9. Samples: 54566. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-03-26 18:02:28,122][28680] Avg episode reward: [(0, '4.403')]
+[2023-03-26 18:02:28,123][28695] Saving new best policy, reward=4.403!
+[2023-03-26 18:02:29,415][28710] Updated weights for policy 0, policy_version 60 (0.0006)
+[2023-03-26 18:02:31,832][28710] Updated weights for policy 0, policy_version 70 (0.0006)
+[2023-03-26 18:02:32,422][28711] EvtLoop [rollout_proc2_evt_loop, process=rollout_proc2] unhandled exception in slot='advance_rollouts' connected to emitter=Emitter(object_id='InferenceWorker_p0-w0', signal_name='advance2'), args=(1, 0)
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 241, in advance_rollouts
+    complete_rollouts, episodic_stats = runner.advance_rollouts(policy_id, self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 634, in advance_rollouts
+    new_obs, rewards, terminated, truncated, infos = e.step(actions)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 319, in step
+    return self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 129, in step
+    obs, rew, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 115, in step
+    obs, rew, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 33, in step
+    observation, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 384, in step
+    observation, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 88, in step
+    obs, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 319, in step
+    return self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/multiplayer_stats.py", line 54, in step
+    obs, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 452, in step
+    reward = self.game.make_action(actions_flattened, self.skip_frames)
+vizdoom.vizdoom.SignalException: Signal SIGINT received. ViZDoom instance has been closed.
+[2023-03-26 18:02:32,422][28713] EvtLoop [rollout_proc4_evt_loop, process=rollout_proc4] unhandled exception in slot='advance_rollouts' connected to emitter=Emitter(object_id='InferenceWorker_p0-w0', signal_name='advance4'), args=(1, 0)
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 241, in advance_rollouts
+    complete_rollouts, episodic_stats = runner.advance_rollouts(policy_id, self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 634, in advance_rollouts
+    new_obs, rewards, terminated, truncated, infos = e.step(actions)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 319, in step
+    return self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 129, in step
+    obs, rew, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 115, in step
+    obs, rew, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 33, in step
+    observation, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 384, in step
+    observation, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 88, in step
+    obs, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 319, in step
+    return self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/multiplayer_stats.py", line 54, in step
+    obs, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 452, in step
+    reward = self.game.make_action(actions_flattened, self.skip_frames)
+vizdoom.vizdoom.SignalException: Signal SIGINT received. ViZDoom instance has been closed.
+[2023-03-26 18:02:32,426][28711] Unhandled exception Signal SIGINT received. ViZDoom instance has been closed. in evt loop rollout_proc2_evt_loop
+[2023-03-26 18:02:32,426][28713] Unhandled exception Signal SIGINT received. ViZDoom instance has been closed. in evt loop rollout_proc4_evt_loop
+[2023-03-26 18:02:32,428][28731] EvtLoop [rollout_proc7_evt_loop, process=rollout_proc7] unhandled exception in slot='advance_rollouts' connected to emitter=Emitter(object_id='InferenceWorker_p0-w0', signal_name='advance7'), args=(1, 0)
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 241, in advance_rollouts
+    complete_rollouts, episodic_stats = runner.advance_rollouts(policy_id, self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 634, in advance_rollouts
+    new_obs, rewards, terminated, truncated, infos = e.step(actions)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 319, in step
+    return self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 129, in step
+    obs, rew, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 115, in step
+    obs, rew, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 33, in step
+    observation, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 384, in step
+    observation, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 88, in step
+    obs, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 319, in step
+    return self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/multiplayer_stats.py", line 54, in step
+    obs, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 452, in step
+    reward = self.game.make_action(actions_flattened, self.skip_frames)
+vizdoom.vizdoom.SignalException: Signal SIGINT received. ViZDoom instance has been closed.
+[2023-03-26 18:02:32,429][28731] Unhandled exception Signal SIGINT received. ViZDoom instance has been closed. in evt loop rollout_proc7_evt_loop
+[2023-03-26 18:02:32,428][28730] EvtLoop [rollout_proc6_evt_loop, process=rollout_proc6] unhandled exception in slot='advance_rollouts' connected to emitter=Emitter(object_id='InferenceWorker_p0-w0', signal_name='advance6'), args=(1, 0)
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 241, in advance_rollouts
+    complete_rollouts, episodic_stats = runner.advance_rollouts(policy_id, self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 634, in advance_rollouts
+    new_obs, rewards, terminated, truncated, infos = e.step(actions)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 319, in step
+    return self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 129, in step
+    obs, rew, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 115, in step
+    obs, rew, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 33, in step
+    observation, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 384, in step
+    observation, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 88, in step
+    obs, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 319, in step
+    return self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/multiplayer_stats.py", line 54, in step
+    obs, reward, terminated, truncated, info = self.env.step(action)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 452, in step
+    reward = self.game.make_action(actions_flattened, self.skip_frames)
+vizdoom.vizdoom.SignalException: Signal SIGINT received. ViZDoom instance has been closed.
+[2023-03-26 18:02:32,429][28730] Unhandled exception Signal SIGINT received. ViZDoom instance has been closed. in evt loop rollout_proc6_evt_loop
+[2023-03-26 18:02:32,431][28680] Keyboard interrupt detected in the event loop EvtLoop [Runner_EvtLoop, process=main process 28680], exiting...
+[2023-03-26 18:02:32,431][28680] Runner profile tree view:
+main_loop: 27.3507
+[2023-03-26 18:02:32,431][28680] Collected {0: 294912}, FPS: 10782.6
+[2023-03-26 18:02:32,432][28695] Stopping Batcher_0...
+[2023-03-26 18:02:32,432][28695] Loop batcher_evt_loop terminating...
+[2023-03-26 18:02:32,433][28695] Saving /home/hit/nnet/hf/rl/unit9/train_dir/default_experiment/checkpoint_p0/checkpoint_000000072_294912.pth...
+[2023-03-26 18:02:32,484][28695] Stopping LearnerWorker_p0...
+[2023-03-26 18:02:32,484][28695] Loop learner_proc0_evt_loop terminating...
+[2023-03-26 18:02:32,492][28710] Weights refcount: 2 0
+[2023-03-26 18:02:32,493][28710] Stopping InferenceWorker_p0-w0...
+[2023-03-26 18:02:32,493][28710] Loop inference_proc0-0_evt_loop terminating...
+[2023-03-26 18:03:26,534][29286] Saving configuration to /home/hit/nnet/hf/rl/unit9/train_dir/default_experiment/config.json...
+[2023-03-26 18:03:26,534][29286] Rollout worker 0 uses device cpu
+[2023-03-26 18:03:26,534][29286] Rollout worker 1 uses device cpu
+[2023-03-26 18:03:26,535][29286] Rollout worker 2 uses device cpu
+[2023-03-26 18:03:26,535][29286] Rollout worker 3 uses device cpu
+[2023-03-26 18:03:26,535][29286] Rollout worker 4 uses device cpu
+[2023-03-26 18:03:26,535][29286] Rollout worker 5 uses device cpu
+[2023-03-26 18:03:26,535][29286] Rollout worker 6 uses device cpu
+[2023-03-26 18:03:26,535][29286] Rollout worker 7 uses device cpu
+[2023-03-26 18:03:26,568][29286] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:03:26,568][29286] InferenceWorker_p0-w0: min num requests: 2
+[2023-03-26 18:03:26,585][29286] Starting all processes...
+[2023-03-26 18:03:26,585][29286] Starting process learner_proc0
+[2023-03-26 18:03:27,728][29286] Starting all processes...
+[2023-03-26 18:03:27,731][29300] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:03:27,731][29300] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0
+[2023-03-26 18:03:27,731][29286] Starting process inference_proc0-0
+[2023-03-26 18:03:27,731][29286] Starting process rollout_proc0
+[2023-03-26 18:03:27,731][29286] Starting process rollout_proc1
+[2023-03-26 18:03:27,731][29286] Starting process rollout_proc2
+[2023-03-26 18:03:27,731][29286] Starting process rollout_proc3
+[2023-03-26 18:03:27,735][29286] Starting process rollout_proc4
+[2023-03-26 18:03:27,738][29300] Num visible devices: 1
+[2023-03-26 18:03:27,735][29286] Starting process rollout_proc5
+[2023-03-26 18:03:27,736][29286] Starting process rollout_proc6
+[2023-03-26 18:03:27,738][29286] Starting process rollout_proc7
+[2023-03-26 18:03:27,756][29300] Starting seed is not provided
+[2023-03-26 18:03:27,756][29300] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:03:27,756][29300] Initializing actor-critic model on device cuda:0
+[2023-03-26 18:03:27,757][29300] RunningMeanStd input shape: (3, 72, 128)
+[2023-03-26 18:03:27,757][29300] RunningMeanStd input shape: (1,)
+[2023-03-26 18:03:27,767][29300] ConvEncoder: input_channels=3
+[2023-03-26 18:03:27,871][29300] Conv encoder output size: 512
+[2023-03-26 18:03:27,871][29300] Policy head output size: 512
+[2023-03-26 18:03:27,886][29300] Created Actor Critic model with architecture:
+[2023-03-26 18:03:27,886][29300] 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)
+  )
+)
+[2023-03-26 18:03:29,126][29340] Worker 7 uses CPU cores [14, 15]
+[2023-03-26 18:03:29,132][29342] Worker 5 uses CPU cores [10, 11]
+[2023-03-26 18:03:29,221][29321] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:03:29,221][29321] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0
+[2023-03-26 18:03:29,231][29321] Num visible devices: 1
+[2023-03-26 18:03:29,303][29339] Worker 3 uses CPU cores [6, 7]
+[2023-03-26 18:03:29,342][29322] Worker 2 uses CPU cores [4, 5]
+[2023-03-26 18:03:29,389][29323] Worker 1 uses CPU cores [2, 3]
+[2023-03-26 18:03:29,453][29320] Worker 0 uses CPU cores [0, 1]
+[2023-03-26 18:03:29,521][29344] Worker 6 uses CPU cores [12, 13]
+[2023-03-26 18:03:29,562][29343] Worker 4 uses CPU cores [8, 9]
+[2023-03-26 18:03:29,611][29300] Using optimizer <class 'torch.optim.adam.Adam'>
+[2023-03-26 18:03:29,611][29300] Loading state from checkpoint /home/hit/nnet/hf/rl/unit9/train_dir/default_experiment/checkpoint_p0/checkpoint_000000072_294912.pth...
+[2023-03-26 18:03:29,643][29300] Loading model from checkpoint
+[2023-03-26 18:03:29,645][29300] Loaded experiment state at self.train_step=72, self.env_steps=294912
+[2023-03-26 18:03:29,645][29300] Initialized policy 0 weights for model version 72
+[2023-03-26 18:03:29,647][29300] LearnerWorker_p0 finished initialization!
+[2023-03-26 18:03:29,647][29300] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:03:29,739][29321] RunningMeanStd input shape: (3, 72, 128)
+[2023-03-26 18:03:29,740][29321] RunningMeanStd input shape: (1,)
+[2023-03-26 18:03:29,748][29321] ConvEncoder: input_channels=3
+[2023-03-26 18:03:29,816][29321] Conv encoder output size: 512
+[2023-03-26 18:03:29,817][29321] Policy head output size: 512
+[2023-03-26 18:03:30,414][29286] Inference worker 0-0 is ready!
+[2023-03-26 18:03:30,415][29286] All inference workers are ready! Signal rollout workers to start!
+[2023-03-26 18:03:30,421][29320] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:03:30,421][29342] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:03:30,422][29323] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:03:30,422][29344] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:03:30,422][29339] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:03:30,423][29343] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:03:30,424][29340] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:03:30,424][29322] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:03:30,616][29340] EvtLoop [rollout_proc7_evt_loop, process=rollout_proc7] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 283, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 311, in reset
+    return self.observation(self.env.reset(**kwargs))
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 191, in observation
+    observation = self._transpose(observation)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 182, in _transpose
+    return np.transpose(obs, (2, 0, 1))  # HWC to CHW for PyTorch
+  File "<__array_function__ internals>", line 180, in transpose
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 660, in transpose
+    return _wrapfunc(a, 'transpose', axes)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 54, in _wrapfunc
+    return _wrapit(obj, method, *args, **kwds)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 43, in _wrapit
+    result = getattr(asarray(obj), method)(*args, **kwds)
+ValueError: axes don't match array
+[2023-03-26 18:03:30,617][29340] Unhandled exception axes don't match array in evt loop rollout_proc7_evt_loop
+[2023-03-26 18:03:30,617][29344] EvtLoop [rollout_proc6_evt_loop, process=rollout_proc6] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 283, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 311, in reset
+    return self.observation(self.env.reset(**kwargs))
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 191, in observation
+    observation = self._transpose(observation)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 182, in _transpose
+    return np.transpose(obs, (2, 0, 1))  # HWC to CHW for PyTorch
+  File "<__array_function__ internals>", line 180, in transpose
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 660, in transpose
+    return _wrapfunc(a, 'transpose', axes)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 54, in _wrapfunc
+    return _wrapit(obj, method, *args, **kwds)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 43, in _wrapit
+    result = getattr(asarray(obj), method)(*args, **kwds)
+ValueError: axes don't match array
+[2023-03-26 18:03:30,618][29344] Unhandled exception axes don't match array in evt loop rollout_proc6_evt_loop
+[2023-03-26 18:03:30,620][29320] EvtLoop [rollout_proc0_evt_loop, process=rollout_proc0] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 283, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 311, in reset
+    return self.observation(self.env.reset(**kwargs))
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 191, in observation
+    observation = self._transpose(observation)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 182, in _transpose
+    return np.transpose(obs, (2, 0, 1))  # HWC to CHW for PyTorch
+  File "<__array_function__ internals>", line 180, in transpose
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 660, in transpose
+    return _wrapfunc(a, 'transpose', axes)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 54, in _wrapfunc
+    return _wrapit(obj, method, *args, **kwds)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 43, in _wrapit
+    result = getattr(asarray(obj), method)(*args, **kwds)
+ValueError: axes don't match array
+[2023-03-26 18:03:30,621][29320] Unhandled exception axes don't match array in evt loop rollout_proc0_evt_loop
+[2023-03-26 18:03:30,625][29343] EvtLoop [rollout_proc4_evt_loop, process=rollout_proc4] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 283, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 311, in reset
+    return self.observation(self.env.reset(**kwargs))
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 191, in observation
+    observation = self._transpose(observation)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 182, in _transpose
+    return np.transpose(obs, (2, 0, 1))  # HWC to CHW for PyTorch
+  File "<__array_function__ internals>", line 180, in transpose
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 660, in transpose
+    return _wrapfunc(a, 'transpose', axes)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 54, in _wrapfunc
+    return _wrapit(obj, method, *args, **kwds)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 43, in _wrapit
+    result = getattr(asarray(obj), method)(*args, **kwds)
+ValueError: axes don't match array
+[2023-03-26 18:03:30,626][29343] Unhandled exception axes don't match array in evt loop rollout_proc4_evt_loop
+[2023-03-26 18:03:30,823][29342] EvtLoop [rollout_proc5_evt_loop, process=rollout_proc5] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 283, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 311, in reset
+    return self.observation(self.env.reset(**kwargs))
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 191, in observation
+    observation = self._transpose(observation)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 182, in _transpose
+    return np.transpose(obs, (2, 0, 1))  # HWC to CHW for PyTorch
+  File "<__array_function__ internals>", line 180, in transpose
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 660, in transpose
+    return _wrapfunc(a, 'transpose', axes)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 54, in _wrapfunc
+    return _wrapit(obj, method, *args, **kwds)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 43, in _wrapit
+    result = getattr(asarray(obj), method)(*args, **kwds)
+ValueError: axes don't match array
+[2023-03-26 18:03:30,825][29342] Unhandled exception axes don't match array in evt loop rollout_proc5_evt_loop
+[2023-03-26 18:03:30,901][29339] EvtLoop [rollout_proc3_evt_loop, process=rollout_proc3] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 283, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 311, in reset
+    return self.observation(self.env.reset(**kwargs))
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 191, in observation
+    observation = self._transpose(observation)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 182, in _transpose
+    return np.transpose(obs, (2, 0, 1))  # HWC to CHW for PyTorch
+  File "<__array_function__ internals>", line 180, in transpose
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 660, in transpose
+    return _wrapfunc(a, 'transpose', axes)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 54, in _wrapfunc
+    return _wrapit(obj, method, *args, **kwds)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 43, in _wrapit
+    result = getattr(asarray(obj), method)(*args, **kwds)
+ValueError: axes don't match array
+[2023-03-26 18:03:30,902][29339] Unhandled exception axes don't match array in evt loop rollout_proc3_evt_loop
+[2023-03-26 18:03:31,024][29323] EvtLoop [rollout_proc1_evt_loop, process=rollout_proc1] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 283, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 311, in reset
+    return self.observation(self.env.reset(**kwargs))
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 191, in observation
+    observation = self._transpose(observation)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 182, in _transpose
+    return np.transpose(obs, (2, 0, 1))  # HWC to CHW for PyTorch
+  File "<__array_function__ internals>", line 180, in transpose
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 660, in transpose
+    return _wrapfunc(a, 'transpose', axes)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 54, in _wrapfunc
+    return _wrapit(obj, method, *args, **kwds)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 43, in _wrapit
+    result = getattr(asarray(obj), method)(*args, **kwds)
+ValueError: axes don't match array
+[2023-03-26 18:03:31,025][29323] Unhandled exception axes don't match array in evt loop rollout_proc1_evt_loop
+[2023-03-26 18:03:31,123][29322] EvtLoop [rollout_proc2_evt_loop, process=rollout_proc2] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 283, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/gym/core.py", line 311, in reset
+    return self.observation(self.env.reset(**kwargs))
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 191, in observation
+    observation = self._transpose(observation)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/sample_factory/envs/env_wrappers.py", line 182, in _transpose
+    return np.transpose(obs, (2, 0, 1))  # HWC to CHW for PyTorch
+  File "<__array_function__ internals>", line 180, in transpose
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 660, in transpose
+    return _wrapfunc(a, 'transpose', axes)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 54, in _wrapfunc
+    return _wrapit(obj, method, *args, **kwds)
+  File "/home/hit/app/rl310/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 43, in _wrapit
+    result = getattr(asarray(obj), method)(*args, **kwds)
+ValueError: axes don't match array
+[2023-03-26 18:03:31,124][29322] Unhandled exception axes don't match array in evt loop rollout_proc2_evt_loop
+[2023-03-26 18:03:34,329][29286] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 294912. Throughput: 0: nan. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2023-03-26 18:03:34,985][29286] Keyboard interrupt detected in the event loop EvtLoop [Runner_EvtLoop, process=main process 29286], exiting...
+[2023-03-26 18:03:34,986][29286] Runner profile tree view:
+main_loop: 8.4004
+[2023-03-26 18:03:34,986][29286] Collected {0: 294912}, FPS: 0.0
+[2023-03-26 18:03:34,986][29300] Stopping Batcher_0...
+[2023-03-26 18:03:34,986][29300] Loop batcher_evt_loop terminating...
+[2023-03-26 18:03:34,987][29300] Saving /home/hit/nnet/hf/rl/unit9/train_dir/default_experiment/checkpoint_p0/checkpoint_000000072_294912.pth...
+[2023-03-26 18:03:35,045][29321] Weights refcount: 2 0
+[2023-03-26 18:03:35,046][29321] Stopping InferenceWorker_p0-w0...
+[2023-03-26 18:03:35,046][29321] Loop inference_proc0-0_evt_loop terminating...
+[2023-03-26 18:03:35,138][29300] Stopping LearnerWorker_p0...
+[2023-03-26 18:03:35,138][29300] Loop learner_proc0_evt_loop terminating...
+[2023-03-26 18:05:17,326][29927] Saving configuration to /home/hit/nnet/hf/rl/unit9/train_dir/default_experiment/config.json...
+[2023-03-26 18:05:17,326][29927] Rollout worker 0 uses device cpu
+[2023-03-26 18:05:17,326][29927] Rollout worker 1 uses device cpu
+[2023-03-26 18:05:17,326][29927] Rollout worker 2 uses device cpu
+[2023-03-26 18:05:17,326][29927] Rollout worker 3 uses device cpu
+[2023-03-26 18:05:17,326][29927] Rollout worker 4 uses device cpu
+[2023-03-26 18:05:17,326][29927] Rollout worker 5 uses device cpu
+[2023-03-26 18:05:17,327][29927] Rollout worker 6 uses device cpu
+[2023-03-26 18:05:17,327][29927] Rollout worker 7 uses device cpu
+[2023-03-26 18:05:17,358][29927] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:05:17,358][29927] InferenceWorker_p0-w0: min num requests: 2
+[2023-03-26 18:05:17,373][29927] Starting all processes...
+[2023-03-26 18:05:17,373][29927] Starting process learner_proc0
+[2023-03-26 18:05:18,342][29927] Starting all processes...
+[2023-03-26 18:05:18,345][29977] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:05:18,345][29977] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0
+[2023-03-26 18:05:18,347][29927] Starting process inference_proc0-0
+[2023-03-26 18:05:18,347][29927] Starting process rollout_proc0
+[2023-03-26 18:05:18,347][29927] Starting process rollout_proc1
+[2023-03-26 18:05:18,348][29927] Starting process rollout_proc2
+[2023-03-26 18:05:18,353][29977] Num visible devices: 1
+[2023-03-26 18:05:18,348][29927] Starting process rollout_proc3
+[2023-03-26 18:05:18,350][29927] Starting process rollout_proc4
+[2023-03-26 18:05:18,352][29927] Starting process rollout_proc5
+[2023-03-26 18:05:18,367][29977] Starting seed is not provided
+[2023-03-26 18:05:18,368][29977] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:05:18,368][29977] Initializing actor-critic model on device cuda:0
+[2023-03-26 18:05:18,368][29977] RunningMeanStd input shape: (3, 72, 128)
+[2023-03-26 18:05:18,368][29977] RunningMeanStd input shape: (1,)
+[2023-03-26 18:05:18,364][29927] Starting process rollout_proc6
+[2023-03-26 18:05:18,364][29927] Starting process rollout_proc7
+[2023-03-26 18:05:18,376][29977] ConvEncoder: input_channels=3
+[2023-03-26 18:05:18,460][29977] Conv encoder output size: 512
+[2023-03-26 18:05:18,460][29977] Policy head output size: 512
+[2023-03-26 18:05:18,469][29977] Created Actor Critic model with architecture:
+[2023-03-26 18:05:18,469][29977] 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)
+  )
+)
+[2023-03-26 18:05:19,651][30009] Worker 3 uses CPU cores [6, 7]
+[2023-03-26 18:05:19,661][29990] Worker 0 uses CPU cores [0, 1]
+[2023-03-26 18:05:19,770][30011] Worker 5 uses CPU cores [10, 11]
+[2023-03-26 18:05:19,770][30013] Worker 6 uses CPU cores [12, 13]
+[2023-03-26 18:05:19,805][29992] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:05:19,805][29992] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0
+[2023-03-26 18:05:19,822][29992] Num visible devices: 1
+[2023-03-26 18:05:19,846][29977] Using optimizer <class 'torch.optim.adam.Adam'>
+[2023-03-26 18:05:19,847][29977] Loading state from checkpoint /home/hit/nnet/hf/rl/unit9/train_dir/default_experiment/checkpoint_p0/checkpoint_000000072_294912.pth...
+[2023-03-26 18:05:19,864][30014] Worker 7 uses CPU cores [14, 15]
+[2023-03-26 18:05:19,885][30012] Worker 4 uses CPU cores [8, 9]
+[2023-03-26 18:05:19,889][30008] Worker 2 uses CPU cores [4, 5]
+[2023-03-26 18:05:19,894][29991] Worker 1 uses CPU cores [2, 3]
+[2023-03-26 18:05:20,046][29977] Loading model from checkpoint
+[2023-03-26 18:05:20,056][29977] Loaded experiment state at self.train_step=72, self.env_steps=294912
+[2023-03-26 18:05:20,068][29977] Initialized policy 0 weights for model version 72
+[2023-03-26 18:05:20,135][29977] LearnerWorker_p0 finished initialization!
+[2023-03-26 18:05:20,135][29977] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-03-26 18:05:20,313][29992] RunningMeanStd input shape: (3, 72, 128)
+[2023-03-26 18:05:20,314][29992] RunningMeanStd input shape: (1,)
+[2023-03-26 18:05:20,330][29992] ConvEncoder: input_channels=3
+[2023-03-26 18:05:20,342][29927] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 294912. Throughput: 0: nan. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2023-03-26 18:05:20,630][29992] Conv encoder output size: 512
+[2023-03-26 18:05:20,630][29992] Policy head output size: 512
+[2023-03-26 18:05:21,515][29927] Inference worker 0-0 is ready!
+[2023-03-26 18:05:21,515][29927] All inference workers are ready! Signal rollout workers to start!
+[2023-03-26 18:05:21,548][30008] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:05:21,549][30011] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:05:21,552][30014] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:05:21,552][30013] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:05:21,553][29991] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:05:21,553][30012] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:05:21,553][29990] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:05:21,571][30009] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-03-26 18:05:21,909][30008] Decorrelating experience for 0 frames...
+[2023-03-26 18:05:21,909][29990] Decorrelating experience for 0 frames...
+[2023-03-26 18:05:21,909][30012] Decorrelating experience for 0 frames...
+[2023-03-26 18:05:21,918][29991] Decorrelating experience for 0 frames...
+[2023-03-26 18:05:21,918][30011] Decorrelating experience for 0 frames...
+[2023-03-26 18:05:22,080][30008] Decorrelating experience for 32 frames...
+[2023-03-26 18:05:22,106][30009] Decorrelating experience for 0 frames...
+[2023-03-26 18:05:22,107][29990] Decorrelating experience for 32 frames...
+[2023-03-26 18:05:22,107][30011] Decorrelating experience for 32 frames...
+[2023-03-26 18:05:22,108][30012] Decorrelating experience for 32 frames...
+[2023-03-26 18:05:22,107][30014] Decorrelating experience for 0 frames...
+[2023-03-26 18:05:22,297][30014] Decorrelating experience for 32 frames...
+[2023-03-26 18:05:22,297][30009] Decorrelating experience for 32 frames...
+[2023-03-26 18:05:22,306][30008] Decorrelating experience for 64 frames...
+[2023-03-26 18:05:22,317][30013] Decorrelating experience for 0 frames...
+[2023-03-26 18:05:22,319][30011] Decorrelating experience for 64 frames...
+[2023-03-26 18:05:22,492][30013] Decorrelating experience for 32 frames...
+[2023-03-26 18:05:22,493][30009] Decorrelating experience for 64 frames...
+[2023-03-26 18:05:22,504][30008] Decorrelating experience for 96 frames...
+[2023-03-26 18:05:22,531][29990] Decorrelating experience for 64 frames...
+[2023-03-26 18:05:22,532][30012] Decorrelating experience for 64 frames...
+[2023-03-26 18:05:22,674][30011] Decorrelating experience for 96 frames...
+[2023-03-26 18:05:22,690][30013] Decorrelating experience for 64 frames...
+[2023-03-26 18:05:22,729][29990] Decorrelating experience for 96 frames...
+[2023-03-26 18:05:22,860][30014] Decorrelating experience for 64 frames...
+[2023-03-26 18:05:22,882][30013] Decorrelating experience for 96 frames...
+[2023-03-26 18:05:22,903][30009] Decorrelating experience for 96 frames...
+[2023-03-26 18:05:22,935][30012] Decorrelating experience for 96 frames...
+[2023-03-26 18:05:23,047][29991] Decorrelating experience for 32 frames...
+[2023-03-26 18:05:23,063][30014] Decorrelating experience for 96 frames...
+[2023-03-26 18:05:23,254][29991] Decorrelating experience for 64 frames...
+[2023-03-26 18:05:23,468][29991] Decorrelating experience for 96 frames...
+[2023-03-26 18:05:23,740][29977] Signal inference workers to stop experience collection...
+[2023-03-26 18:05:23,749][29992] InferenceWorker_p0-w0: stopping experience collection
+[2023-03-26 18:05:24,787][29977] Signal inference workers to resume experience collection...
+[2023-03-26 18:05:24,788][29992] InferenceWorker_p0-w0: resuming experience collection
+[2023-03-26 18:05:25,342][29927] Fps is (10 sec: 819.2, 60 sec: 819.2, 300 sec: 819.2). Total num frames: 299008. Throughput: 0: 596.8. Samples: 2984. Policy #0 lag: (min: 0.0, avg: 0.0, max: 0.0)
+[2023-03-26 18:05:25,342][29927] Avg episode reward: [(0, '3.144')]
+[2023-03-26 18:05:26,570][29992] Updated weights for policy 0, policy_version 82 (0.0258)
+[2023-03-26 18:05:28,037][29992] Updated weights for policy 0, policy_version 92 (0.0006)
+[2023-03-26 18:05:29,494][29992] Updated weights for policy 0, policy_version 102 (0.0006)
+[2023-03-26 18:05:30,342][29927] Fps is (10 sec: 14336.2, 60 sec: 14336.2, 300 sec: 14336.2). Total num frames: 438272. Throughput: 0: 1778.6. Samples: 17786. Policy #0 lag: (min: 0.0, avg: 0.9, max: 2.0)
+[2023-03-26 18:05:30,342][29927] Avg episode reward: [(0, '4.552')]
+[2023-03-26 18:05:30,353][29977] Saving new best policy, reward=4.552!
+[2023-03-26 18:05:30,930][29992] Updated weights for policy 0, policy_version 112 (0.0006)
+[2023-03-26 18:05:32,394][29992] Updated weights for policy 0, policy_version 122 (0.0006)
+[2023-03-26 18:05:33,872][29992] Updated weights for policy 0, policy_version 132 (0.0006)
+[2023-03-26 18:05:35,342][29927] Fps is (10 sec: 27852.7, 60 sec: 18841.7, 300 sec: 18841.7). Total num frames: 577536. Throughput: 0: 3982.2. Samples: 59732. Policy #0 lag: (min: 0.0, avg: 0.6, max: 2.0)
+[2023-03-26 18:05:35,342][29927] Avg episode reward: [(0, '4.635')]
+[2023-03-26 18:05:35,345][29992] Updated weights for policy 0, policy_version 142 (0.0006)
+[2023-03-26 18:05:35,345][29977] Saving new best policy, reward=4.635!
+[2023-03-26 18:05:36,830][29992] Updated weights for policy 0, policy_version 152 (0.0006)
+[2023-03-26 18:05:37,354][29927] Heartbeat connected on Batcher_0
+[2023-03-26 18:05:37,356][29927] Heartbeat connected on LearnerWorker_p0
+[2023-03-26 18:05:37,361][29927] Heartbeat connected on RolloutWorker_w0
+[2023-03-26 18:05:37,362][29927] Heartbeat connected on InferenceWorker_p0-w0
+[2023-03-26 18:05:37,365][29927] Heartbeat connected on RolloutWorker_w2
+[2023-03-26 18:05:37,366][29927] Heartbeat connected on RolloutWorker_w3
+[2023-03-26 18:05:37,368][29927] Heartbeat connected on RolloutWorker_w4
+[2023-03-26 18:05:37,368][29927] Heartbeat connected on RolloutWorker_w1
+[2023-03-26 18:05:37,369][29927] Heartbeat connected on RolloutWorker_w5
+[2023-03-26 18:05:37,371][29927] Heartbeat connected on RolloutWorker_w6
+[2023-03-26 18:05:37,373][29927] Heartbeat connected on RolloutWorker_w7
+[2023-03-26 18:05:38,366][29992] Updated weights for policy 0, policy_version 162 (0.0006)
+[2023-03-26 18:05:39,860][29992] Updated weights for policy 0, policy_version 172 (0.0006)
+[2023-03-26 18:05:40,342][29927] Fps is (10 sec: 27852.4, 60 sec: 21094.4, 300 sec: 21094.4). Total num frames: 716800. Throughput: 0: 5056.3. Samples: 101126. Policy #0 lag: (min: 0.0, avg: 0.9, max: 2.0)
+[2023-03-26 18:05:40,342][29927] Avg episode reward: [(0, '4.628')]
+[2023-03-26 18:05:41,310][29992] Updated weights for policy 0, policy_version 182 (0.0006)
+[2023-03-26 18:05:42,797][29992] Updated weights for policy 0, policy_version 192 (0.0006)
+[2023-03-26 18:05:44,410][29992] Updated weights for policy 0, policy_version 202 (0.0007)
+[2023-03-26 18:05:45,342][29927] Fps is (10 sec: 27443.1, 60 sec: 22282.3, 300 sec: 22282.3). Total num frames: 851968. Throughput: 0: 4876.9. Samples: 121922. Policy #0 lag: (min: 0.0, avg: 0.8, max: 1.0)
+[2023-03-26 18:05:45,342][29927] Avg episode reward: [(0, '5.026')]
+[2023-03-26 18:05:45,345][29977] Saving new best policy, reward=5.026!
+[2023-03-26 18:05:45,902][29992] Updated weights for policy 0, policy_version 212 (0.0006)
+[2023-03-26 18:05:47,346][29992] Updated weights for policy 0, policy_version 222 (0.0006)
+[2023-03-26 18:05:48,815][29992] Updated weights for policy 0, policy_version 232 (0.0006)
+[2023-03-26 18:05:50,179][29992] Updated weights for policy 0, policy_version 242 (0.0006)
+[2023-03-26 18:05:50,342][29927] Fps is (10 sec: 27852.9, 60 sec: 23347.2, 300 sec: 23347.2). Total num frames: 995328. Throughput: 0: 5421.9. Samples: 162658. Policy #0 lag: (min: 0.0, avg: 0.9, max: 2.0)
+[2023-03-26 18:05:50,342][29927] Avg episode reward: [(0, '5.257')]
+[2023-03-26 18:05:50,344][29977] Saving new best policy, reward=5.257!
+[2023-03-26 18:05:51,614][29992] Updated weights for policy 0, policy_version 252 (0.0006)
+[2023-03-26 18:05:53,043][29992] Updated weights for policy 0, policy_version 262 (0.0006)
+[2023-03-26 18:05:54,518][29992] Updated weights for policy 0, policy_version 272 (0.0006)
+[2023-03-26 18:05:55,342][29927] Fps is (10 sec: 28262.4, 60 sec: 23990.9, 300 sec: 23990.9). Total num frames: 1134592. Throughput: 0: 5873.7. Samples: 205578. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
+[2023-03-26 18:05:55,342][29927] Avg episode reward: [(0, '6.496')]
+[2023-03-26 18:05:55,391][29977] Saving new best policy, reward=6.496!
+[2023-03-26 18:05:55,972][29992] Updated weights for policy 0, policy_version 282 (0.0006)
+[2023-03-26 18:05:57,434][29992] Updated weights for policy 0, policy_version 292 (0.0006)
+[2023-03-26 18:05:58,854][29992] Updated weights for policy 0, policy_version 302 (0.0006)
+[2023-03-26 18:06:00,284][29992] Updated weights for policy 0, policy_version 312 (0.0006)
+[2023-03-26 18:06:00,342][29927] Fps is (10 sec: 28262.5, 60 sec: 24576.0, 300 sec: 24576.0). Total num frames: 1277952. Throughput: 0: 5669.3. Samples: 226772. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
+[2023-03-26 18:06:00,342][29927] Avg episode reward: [(0, '7.329')]
+[2023-03-26 18:06:00,343][29977] Saving new best policy, reward=7.329!
+[2023-03-26 18:06:01,731][29992] Updated weights for policy 0, policy_version 322 (0.0006)
+[2023-03-26 18:06:03,207][29992] Updated weights for policy 0, policy_version 332 (0.0006)
+[2023-03-26 18:06:04,696][29992] Updated weights for policy 0, policy_version 342 (0.0006)
+[2023-03-26 18:06:05,342][29927] Fps is (10 sec: 28262.2, 60 sec: 24940.1, 300 sec: 24940.1). Total num frames: 1417216. Throughput: 0: 5982.7. Samples: 269220. Policy #0 lag: (min: 0.0, avg: 0.9, max: 2.0)
+[2023-03-26 18:06:05,342][29927] Avg episode reward: [(0, '7.245')]
+[2023-03-26 18:06:06,157][29992] Updated weights for policy 0, policy_version 352 (0.0006)
+[2023-03-26 18:06:07,584][29992] Updated weights for policy 0, policy_version 362 (0.0006)
+[2023-03-26 18:06:09,057][29992] Updated weights for policy 0, policy_version 372 (0.0006)
+[2023-03-26 18:06:10,342][29927] Fps is (10 sec: 27852.9, 60 sec: 25231.4, 300 sec: 25231.4). Total num frames: 1556480. Throughput: 0: 6848.3. Samples: 311156. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
+[2023-03-26 18:06:10,342][29927] Avg episode reward: [(0, '9.407')]
+[2023-03-26 18:06:10,368][29977] Saving new best policy, reward=9.407!
+[2023-03-26 18:06:10,526][29992] Updated weights for policy 0, policy_version 382 (0.0006)
+[2023-03-26 18:06:12,002][29992] Updated weights for policy 0, policy_version 392 (0.0006)
+[2023-03-26 18:06:13,471][29992] Updated weights for policy 0, policy_version 402 (0.0006)
+[2023-03-26 18:06:14,920][29992] Updated weights for policy 0, policy_version 412 (0.0006)
+[2023-03-26 18:06:15,342][29927] Fps is (10 sec: 27852.9, 60 sec: 25469.7, 300 sec: 25469.7). Total num frames: 1695744. Throughput: 0: 6985.5. Samples: 332132. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
+[2023-03-26 18:06:15,342][29927] Avg episode reward: [(0, '10.087')]
+[2023-03-26 18:06:15,350][29977] Saving new best policy, reward=10.087!
+[2023-03-26 18:06:16,436][29992] Updated weights for policy 0, policy_version 422 (0.0006)
+[2023-03-26 18:06:17,888][29992] Updated weights for policy 0, policy_version 432 (0.0006)
+[2023-03-26 18:06:19,344][29992] Updated weights for policy 0, policy_version 442 (0.0006)
+[2023-03-26 18:06:20,342][29927] Fps is (10 sec: 27852.6, 60 sec: 25668.3, 300 sec: 25668.3). Total num frames: 1835008. Throughput: 0: 6981.4. Samples: 373894. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
+[2023-03-26 18:06:20,342][29927] Avg episode reward: [(0, '12.004')]
+[2023-03-26 18:06:20,364][29977] Saving new best policy, reward=12.004!
+[2023-03-26 18:06:20,796][29992] Updated weights for policy 0, policy_version 452 (0.0006)
+[2023-03-26 18:06:22,250][29992] Updated weights for policy 0, policy_version 462 (0.0006)
+[2023-03-26 18:06:23,734][29992] Updated weights for policy 0, policy_version 472 (0.0006)
+[2023-03-26 18:06:25,245][29992] Updated weights for policy 0, policy_version 482 (0.0006)
+[2023-03-26 18:06:25,342][29927] Fps is (10 sec: 27852.9, 60 sec: 27921.1, 300 sec: 25836.3). Total num frames: 1974272. Throughput: 0: 6991.6. Samples: 415748. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
+[2023-03-26 18:06:25,342][29927] Avg episode reward: [(0, '17.983')]
+[2023-03-26 18:06:25,387][29977] Saving new best policy, reward=17.983!
+[2023-03-26 18:06:26,674][29992] Updated weights for policy 0, policy_version 492 (0.0006)
+[2023-03-26 18:06:28,130][29992] Updated weights for policy 0, policy_version 502 (0.0006)
+[2023-03-26 18:06:29,564][29992] Updated weights for policy 0, policy_version 512 (0.0006)
+[2023-03-26 18:06:30,342][29927] Fps is (10 sec: 28262.4, 60 sec: 27989.3, 300 sec: 26038.9). Total num frames: 2117632. Throughput: 0: 6999.7. Samples: 436910. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
+[2023-03-26 18:06:30,342][29927] Avg episode reward: [(0, '17.345')]
+[2023-03-26 18:06:31,026][29992] Updated weights for policy 0, policy_version 522 (0.0006)
+[2023-03-26 18:06:32,483][29992] Updated weights for policy 0, policy_version 532 (0.0006)
+[2023-03-26 18:06:33,928][29992] Updated weights for policy 0, policy_version 542 (0.0006)
+[2023-03-26 18:06:35,342][29927] Fps is (10 sec: 28262.2, 60 sec: 27989.3, 300 sec: 26159.8). Total num frames: 2256896. Throughput: 0: 7034.1. Samples: 479192. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
+[2023-03-26 18:06:35,342][29927] Avg episode reward: [(0, '18.067')]
+[2023-03-26 18:06:35,369][29977] Saving new best policy, reward=18.067!
+[2023-03-26 18:06:35,369][29992] Updated weights for policy 0, policy_version 552 (0.0006)
+[2023-03-26 18:06:36,824][29992] Updated weights for policy 0, policy_version 562 (0.0006)
+[2023-03-26 18:06:38,257][29992] Updated weights for policy 0, policy_version 572 (0.0006)
+[2023-03-26 18:06:39,735][29992] Updated weights for policy 0, policy_version 582 (0.0006)
+[2023-03-26 18:06:40,342][29927] Fps is (10 sec: 28262.5, 60 sec: 28057.6, 300 sec: 26316.8). Total num frames: 2400256. Throughput: 0: 7026.0. Samples: 521750. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
+[2023-03-26 18:06:40,342][29927] Avg episode reward: [(0, '21.111')]
+[2023-03-26 18:06:40,344][29977] Saving new best policy, reward=21.111!
+[2023-03-26 18:06:41,195][29992] Updated weights for policy 0, policy_version 592 (0.0006)
+[2023-03-26 18:06:42,682][29992] Updated weights for policy 0, policy_version 602 (0.0006)
+[2023-03-26 18:06:44,139][29992] Updated weights for policy 0, policy_version 612 (0.0006)
+[2023-03-26 18:06:45,342][29927] Fps is (10 sec: 28262.6, 60 sec: 28125.9, 300 sec: 26407.2). Total num frames: 2539520. Throughput: 0: 7017.7. Samples: 542568. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
+[2023-03-26 18:06:45,342][29927] Avg episode reward: [(0, '22.074')]
+[2023-03-26 18:06:45,344][29977] Saving new best policy, reward=22.074!
+[2023-03-26 18:06:45,578][29992] Updated weights for policy 0, policy_version 622 (0.0006)
+[2023-03-26 18:06:47,011][29992] Updated weights for policy 0, policy_version 632 (0.0006)
+[2023-03-26 18:06:48,491][29992] Updated weights for policy 0, policy_version 642 (0.0006)
+[2023-03-26 18:06:49,961][29992] Updated weights for policy 0, policy_version 652 (0.0006)
+[2023-03-26 18:06:50,342][29927] Fps is (10 sec: 27852.6, 60 sec: 28057.6, 300 sec: 26487.5). Total num frames: 2678784. Throughput: 0: 7011.5. Samples: 584738. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
+[2023-03-26 18:06:50,342][29927] Avg episode reward: [(0, '21.318')]
+[2023-03-26 18:06:51,412][29992] Updated weights for policy 0, policy_version 662 (0.0006)
+[2023-03-26 18:06:52,851][29992] Updated weights for policy 0, policy_version 672 (0.0006)
+[2023-03-26 18:06:54,272][29992] Updated weights for policy 0, policy_version 682 (0.0006)
+[2023-03-26 18:06:55,342][29927] Fps is (10 sec: 28262.3, 60 sec: 28125.9, 300 sec: 26602.5). Total num frames: 2822144. Throughput: 0: 7026.6. Samples: 627352. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
+[2023-03-26 18:06:55,342][29927] Avg episode reward: [(0, '23.056')]
+[2023-03-26 18:06:55,345][29977] Saving new best policy, reward=23.056!
+[2023-03-26 18:06:55,761][29992] Updated weights for policy 0, policy_version 692 (0.0006)
+[2023-03-26 18:06:57,242][29992] Updated weights for policy 0, policy_version 702 (0.0006)
+[2023-03-26 18:06:58,682][29992] Updated weights for policy 0, policy_version 712 (0.0006)
+[2023-03-26 18:07:00,106][29992] Updated weights for policy 0, policy_version 722 (0.0006)
+[2023-03-26 18:07:00,342][29927] Fps is (10 sec: 28262.5, 60 sec: 28057.6, 300 sec: 26665.0). Total num frames: 2961408. Throughput: 0: 7016.6. Samples: 647880. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
+[2023-03-26 18:07:00,342][29927] Avg episode reward: [(0, '22.628')]
+[2023-03-26 18:07:01,599][29992] Updated weights for policy 0, policy_version 732 (0.0006)
+[2023-03-26 18:07:03,173][29992] Updated weights for policy 0, policy_version 742 (0.0007)
+[2023-03-26 18:07:04,669][29992] Updated weights for policy 0, policy_version 752 (0.0007)
+[2023-03-26 18:07:05,342][29927] Fps is (10 sec: 27443.3, 60 sec: 27989.4, 300 sec: 26682.5). Total num frames: 3096576. Throughput: 0: 7011.9. Samples: 689428. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
+[2023-03-26 18:07:05,342][29927] Avg episode reward: [(0, '22.380')]
+[2023-03-26 18:07:06,135][29992] Updated weights for policy 0, policy_version 762 (0.0006)
+[2023-03-26 18:07:07,628][29992] Updated weights for policy 0, policy_version 772 (0.0006)
+[2023-03-26 18:07:09,148][29992] Updated weights for policy 0, policy_version 782 (0.0006)
+[2023-03-26 18:07:10,342][29927] Fps is (10 sec: 27443.4, 60 sec: 27989.3, 300 sec: 26735.7). Total num frames: 3235840. Throughput: 0: 6997.0. Samples: 730612. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
+[2023-03-26 18:07:10,342][29927] Avg episode reward: [(0, '18.703')]
+[2023-03-26 18:07:10,595][29992] Updated weights for policy 0, policy_version 792 (0.0006)
+[2023-03-26 18:07:12,083][29992] Updated weights for policy 0, policy_version 802 (0.0007)
+[2023-03-26 18:07:13,550][29992] Updated weights for policy 0, policy_version 812 (0.0007)
+[2023-03-26 18:07:15,034][29992] Updated weights for policy 0, policy_version 822 (0.0006)
+[2023-03-26 18:07:15,342][29927] Fps is (10 sec: 27852.7, 60 sec: 27989.3, 300 sec: 26784.3). Total num frames: 3375104. Throughput: 0: 6987.0. Samples: 751324. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
+[2023-03-26 18:07:15,342][29927] Avg episode reward: [(0, '24.707')]
+[2023-03-26 18:07:15,345][29977] Saving /home/hit/nnet/hf/rl/unit9/train_dir/default_experiment/checkpoint_p0/checkpoint_000000824_3375104.pth...
+[2023-03-26 18:07:15,395][29977] Saving new best policy, reward=24.707!
+[2023-03-26 18:07:16,510][29992] Updated weights for policy 0, policy_version 832 (0.0006)
+[2023-03-26 18:07:17,986][29992] Updated weights for policy 0, policy_version 842 (0.0006)
+[2023-03-26 18:07:19,428][29992] Updated weights for policy 0, policy_version 852 (0.0006)
+[2023-03-26 18:07:20,342][29927] Fps is (10 sec: 27852.6, 60 sec: 27989.3, 300 sec: 26828.8). Total num frames: 3514368. Throughput: 0: 6976.8. Samples: 793150. Policy #0 lag: (min: 0.0, avg: 0.8, max: 2.0)
+[2023-03-26 18:07:20,342][29927] Avg episode reward: [(0, '22.921')]
+[2023-03-26 18:07:20,925][29992] Updated weights for policy 0, policy_version 862 (0.0006)
+[2023-03-26 18:07:22,390][29992] Updated weights for policy 0, policy_version 872 (0.0006)
+[2023-03-26 18:07:23,920][29992] Updated weights for policy 0, policy_version 882 (0.0007)
+[2023-03-26 18:07:25,342][29927] Fps is (10 sec: 27443.1, 60 sec: 27921.0, 300 sec: 26837.0). Total num frames: 3649536. Throughput: 0: 6945.6. Samples: 834302. Policy #0 lag: (min: 0.0, avg: 0.9, max: 2.0)
+[2023-03-26 18:07:25,342][29927] Avg episode reward: [(0, '23.368')]
+[2023-03-26 18:07:25,405][29992] Updated weights for policy 0, policy_version 892 (0.0006)
+[2023-03-26 18:07:26,893][29992] Updated weights for policy 0, policy_version 902 (0.0006)
+[2023-03-26 18:07:28,321][29992] Updated weights for policy 0, policy_version 912 (0.0006)
+[2023-03-26 18:07:29,734][29992] Updated weights for policy 0, policy_version 922 (0.0006)
+[2023-03-26 18:07:30,342][29927] Fps is (10 sec: 27853.0, 60 sec: 27921.1, 300 sec: 26907.6). Total num frames: 3792896. Throughput: 0: 6946.9. Samples: 855180. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
+[2023-03-26 18:07:30,342][29927] Avg episode reward: [(0, '20.870')]
+[2023-03-26 18:07:31,194][29992] Updated weights for policy 0, policy_version 932 (0.0006)
+[2023-03-26 18:07:32,616][29992] Updated weights for policy 0, policy_version 942 (0.0006)
+[2023-03-26 18:07:34,048][29992] Updated weights for policy 0, policy_version 952 (0.0006)
+[2023-03-26 18:07:35,342][29927] Fps is (10 sec: 28262.5, 60 sec: 27921.1, 300 sec: 26942.6). Total num frames: 3932160. Throughput: 0: 6965.0. Samples: 898162. Policy #0 lag: (min: 0.0, avg: 0.7, max: 2.0)
+[2023-03-26 18:07:35,342][29927] Avg episode reward: [(0, '24.605')]
+[2023-03-26 18:07:35,504][29992] Updated weights for policy 0, policy_version 962 (0.0006)
+[2023-03-26 18:07:36,987][29992] Updated weights for policy 0, policy_version 972 (0.0006)
+[2023-03-26 18:07:37,877][29977] Stopping Batcher_0...
+[2023-03-26 18:07:37,877][29927] Component Batcher_0 stopped!
+[2023-03-26 18:07:37,877][29977] Saving /home/hit/nnet/hf/rl/unit9/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
+[2023-03-26 18:07:37,877][29977] Loop batcher_evt_loop terminating...
+[2023-03-26 18:07:37,898][29992] Weights refcount: 2 0
+[2023-03-26 18:07:37,903][29992] Stopping InferenceWorker_p0-w0...
+[2023-03-26 18:07:37,903][29992] Loop inference_proc0-0_evt_loop terminating...
+[2023-03-26 18:07:37,903][29927] Component InferenceWorker_p0-w0 stopped!
+[2023-03-26 18:07:37,917][29977] Removing /home/hit/nnet/hf/rl/unit9/train_dir/default_experiment/checkpoint_p0/checkpoint_000000072_294912.pth
+[2023-03-26 18:07:37,918][29977] Saving /home/hit/nnet/hf/rl/unit9/train_dir/default_experiment/checkpoint_p0/checkpoint_000000978_4005888.pth...
+[2023-03-26 18:07:37,930][29990] Stopping RolloutWorker_w0...
+[2023-03-26 18:07:37,930][30012] Stopping RolloutWorker_w4...
+[2023-03-26 18:07:37,930][29927] Component RolloutWorker_w0 stopped!
+[2023-03-26 18:07:37,930][29990] Loop rollout_proc0_evt_loop terminating...
+[2023-03-26 18:07:37,930][29927] Component RolloutWorker_w4 stopped!
+[2023-03-26 18:07:37,930][30012] Loop rollout_proc4_evt_loop terminating...
+[2023-03-26 18:07:37,930][30013] Stopping RolloutWorker_w6...
+[2023-03-26 18:07:37,931][29927] Component RolloutWorker_w6 stopped!
+[2023-03-26 18:07:37,931][30013] Loop rollout_proc6_evt_loop terminating...
+[2023-03-26 18:07:37,931][30014] Stopping RolloutWorker_w7...
+[2023-03-26 18:07:37,931][29927] Component RolloutWorker_w7 stopped!
+[2023-03-26 18:07:37,931][30014] Loop rollout_proc7_evt_loop terminating...
+[2023-03-26 18:07:37,931][29927] Component RolloutWorker_w1 stopped!
+[2023-03-26 18:07:37,931][29991] Stopping RolloutWorker_w1...
+[2023-03-26 18:07:37,931][29927] Component RolloutWorker_w3 stopped!
+[2023-03-26 18:07:37,931][29927] Component RolloutWorker_w2 stopped!
+[2023-03-26 18:07:37,931][30009] Stopping RolloutWorker_w3...
+[2023-03-26 18:07:37,931][29991] Loop rollout_proc1_evt_loop terminating...
+[2023-03-26 18:07:37,931][30008] Stopping RolloutWorker_w2...
+[2023-03-26 18:07:37,932][29927] Component RolloutWorker_w5 stopped!
+[2023-03-26 18:07:37,932][30009] Loop rollout_proc3_evt_loop terminating...
+[2023-03-26 18:07:37,932][30008] Loop rollout_proc2_evt_loop terminating...
+[2023-03-26 18:07:37,932][30011] Stopping RolloutWorker_w5...
+[2023-03-26 18:07:37,932][30011] Loop rollout_proc5_evt_loop terminating...
+[2023-03-26 18:07:38,071][29977] Stopping LearnerWorker_p0...
+[2023-03-26 18:07:38,071][29927] Component LearnerWorker_p0 stopped!
+[2023-03-26 18:07:38,071][29977] Loop learner_proc0_evt_loop terminating...
+[2023-03-26 18:07:38,071][29927] Waiting for process learner_proc0 to stop...
+[2023-03-26 18:07:38,782][29927] Waiting for process inference_proc0-0 to join...
+[2023-03-26 18:07:38,782][29927] Waiting for process rollout_proc0 to join...
+[2023-03-26 18:07:38,782][29927] Waiting for process rollout_proc1 to join...
+[2023-03-26 18:07:38,782][29927] Waiting for process rollout_proc2 to join...
+[2023-03-26 18:07:38,782][29927] Waiting for process rollout_proc3 to join...
+[2023-03-26 18:07:38,782][29927] Waiting for process rollout_proc4 to join...
+[2023-03-26 18:07:38,782][29927] Waiting for process rollout_proc5 to join...
+[2023-03-26 18:07:38,782][29927] Waiting for process rollout_proc6 to join...
+[2023-03-26 18:07:38,782][29927] Waiting for process rollout_proc7 to join...
+[2023-03-26 18:07:38,782][29927] Batcher 0 profile tree view:
+batching: 7.1332, releasing_batches: 0.0138
+[2023-03-26 18:07:38,782][29927] InferenceWorker_p0-w0 profile tree view:
+wait_policy: 0.0000
+  wait_policy_total: 2.5493
+update_model: 2.0518
+  weight_update: 0.0006
+one_step: 0.0010
+  handle_policy_step: 124.2579
+    deserialize: 5.1725, stack: 0.6572, obs_to_device_normalize: 34.1223, forward: 47.7467, send_messages: 6.9335
+    prepare_outputs: 25.2262
+      to_cpu: 19.2284
+[2023-03-26 18:07:38,782][29927] Learner 0 profile tree view:
+misc: 0.0033, prepare_batch: 10.1881
+train: 35.8925
+  epoch_init: 0.0030, minibatch_init: 0.0047, losses_postprocess: 0.2331, kl_divergence: 0.2144, after_optimizer: 0.2908
+  calculate_losses: 10.7378
+    losses_init: 0.0019, forward_head: 0.6191, bptt_initial: 7.9405, tail: 0.3217, advantages_returns: 0.0921, losses: 0.9662
+    bptt: 0.7048
+      bptt_forward_core: 0.6812
+  update: 24.1823
+    clip: 0.6307
+[2023-03-26 18:07:38,783][29927] RolloutWorker_w0 profile tree view:
+wait_for_trajectories: 0.0830, enqueue_policy_requests: 4.4344, env_step: 69.2102, overhead: 4.7997, complete_rollouts: 0.1386
+save_policy_outputs: 4.9534
+  split_output_tensors: 2.3552
+[2023-03-26 18:07:38,783][29927] RolloutWorker_w7 profile tree view:
+wait_for_trajectories: 0.0860, enqueue_policy_requests: 4.8443, env_step: 71.0832, overhead: 5.2545, complete_rollouts: 0.1498
+save_policy_outputs: 5.3212
+  split_output_tensors: 2.5308
+[2023-03-26 18:07:38,783][29927] Loop Runner_EvtLoop terminating...
+[2023-03-26 18:07:38,783][29927] Runner profile tree view:
+main_loop: 141.4099
+[2023-03-26 18:07:38,783][29927] Collected {0: 4005888}, FPS: 26242.7