Upload folder using huggingface_hub

.summary/0/events.out.tfevents.1756806802.a1e0233c2656

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0df08ea8be895cd332bddead4c58dfda6ed0c895f60433ac1828c55c6a298f4d
+size 2497

README.md

@@ -15,7 +15,7 @@ model-index:
       type: doom_health_gathering_supreme
     metrics:
     - type: mean_reward
-      value: 4.06 +/- 0.70
+      value: 3.95 +/- 0.57
       name: mean_reward
       verified: false
 ---

checkpoint_p0/checkpoint_000000005_20480.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3578342954be56affd16ea702feccdba842a0bc2b16223716e02d6786fe9b7f
+size 34929457

config.json

@@ -65,7 +65,7 @@
   "summaries_use_frameskip": true,
   "heartbeat_interval": 20,
   "heartbeat_reporting_interval": 600,
-  "train_for_env_steps": 4000,
+  "train_for_env_steps": 10000,
   "train_for_seconds": 10000000000,
   "save_every_sec": 120,
   "keep_checkpoints": 2,

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a04e9665fb99c18667cadeb0eff2e27a518aac8d67611b0b9cfadca13422fd1c
-size 6063304
+oid sha256:b60dd4467cfe59ca184996dba397af9bb352f8426644f69ad1955c5c195afe5a
+size 5808317

sf_log.txt

@@ -708,3 +708,474 @@ main_loop: 55.3351
 [2025-09-02 09:52:13,559][02807] Avg episode rewards: #0: 4.564, true rewards: #0: 4.064
 [2025-09-02 09:52:13,560][02807] Avg episode reward: 4.564, avg true_objective: 4.064
 [2025-09-02 09:52:40,090][02807] Replay video saved to /content/train_dir/default_experiment/replay.mp4!
+[2025-09-02 09:52:44,244][02807] The model has been pushed to https://huggingface.co/WangChongan/rl_course_vizdoom_health_gathering_supreme
+[2025-09-02 09:53:22,722][02807] Environment doom_basic already registered, overwriting...
+[2025-09-02 09:53:22,724][02807] Environment doom_two_colors_easy already registered, overwriting...
+[2025-09-02 09:53:22,725][02807] Environment doom_two_colors_hard already registered, overwriting...
+[2025-09-02 09:53:22,726][02807] Environment doom_dm already registered, overwriting...
+[2025-09-02 09:53:22,727][02807] Environment doom_dwango5 already registered, overwriting...
+[2025-09-02 09:53:22,728][02807] Environment doom_my_way_home_flat_actions already registered, overwriting...
+[2025-09-02 09:53:22,729][02807] Environment doom_defend_the_center_flat_actions already registered, overwriting...
+[2025-09-02 09:53:22,730][02807] Environment doom_my_way_home already registered, overwriting...
+[2025-09-02 09:53:22,731][02807] Environment doom_deadly_corridor already registered, overwriting...
+[2025-09-02 09:53:22,735][02807] Environment doom_defend_the_center already registered, overwriting...
+[2025-09-02 09:53:22,736][02807] Environment doom_defend_the_line already registered, overwriting...
+[2025-09-02 09:53:22,737][02807] Environment doom_health_gathering already registered, overwriting...
+[2025-09-02 09:53:22,738][02807] Environment doom_health_gathering_supreme already registered, overwriting...
+[2025-09-02 09:53:22,739][02807] Environment doom_battle already registered, overwriting...
+[2025-09-02 09:53:22,740][02807] Environment doom_battle2 already registered, overwriting...
+[2025-09-02 09:53:22,743][02807] Environment doom_duel_bots already registered, overwriting...
+[2025-09-02 09:53:22,744][02807] Environment doom_deathmatch_bots already registered, overwriting...
+[2025-09-02 09:53:22,745][02807] Environment doom_duel already registered, overwriting...
+[2025-09-02 09:53:22,746][02807] Environment doom_deathmatch_full already registered, overwriting...
+[2025-09-02 09:53:22,746][02807] Environment doom_benchmark already registered, overwriting...
+[2025-09-02 09:53:22,747][02807] register_encoder_factory: <function make_vizdoom_encoder at 0x7f1b7c8de020>
+[2025-09-02 09:53:22,771][02807] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
+[2025-09-02 09:53:22,773][02807] Overriding arg 'train_for_env_steps' with value 10000 passed from command line
+[2025-09-02 09:53:22,776][02807] Experiment dir /content/train_dir/default_experiment already exists!
+[2025-09-02 09:53:22,779][02807] Resuming existing experiment from /content/train_dir/default_experiment...
+[2025-09-02 09:53:22,780][02807] Weights and Biases integration disabled
+[2025-09-02 09:53:22,783][02807] Environment var CUDA_VISIBLE_DEVICES is 
+
+[2025-09-02 09:53:25,606][02807] Starting experiment with the following configuration:
+help=False
+algo=APPO
+env=doom_health_gathering_supreme
+experiment=default_experiment
+train_dir=/content/train_dir
+restart_behavior=resume
+device=cpu
+seed=None
+num_policies=1
+async_rl=True
+serial_mode=False
+batched_sampling=False
+num_batches_to_accumulate=2
+worker_num_splits=2
+policy_workers_per_policy=1
+max_policy_lag=1000
+num_workers=8
+num_envs_per_worker=4
+batch_size=1024
+num_batches_per_epoch=1
+num_epochs=1
+rollout=32
+recurrence=32
+shuffle_minibatches=False
+gamma=0.99
+reward_scale=1.0
+reward_clip=1000.0
+value_bootstrap=False
+normalize_returns=True
+exploration_loss_coeff=0.001
+value_loss_coeff=0.5
+kl_loss_coeff=0.0
+exploration_loss=symmetric_kl
+gae_lambda=0.95
+ppo_clip_ratio=0.1
+ppo_clip_value=0.2
+with_vtrace=False
+vtrace_rho=1.0
+vtrace_c=1.0
+optimizer=adam
+adam_eps=1e-06
+adam_beta1=0.9
+adam_beta2=0.999
+max_grad_norm=4.0
+learning_rate=0.0001
+lr_schedule=constant
+lr_schedule_kl_threshold=0.008
+lr_adaptive_min=1e-06
+lr_adaptive_max=0.01
+obs_subtract_mean=0.0
+obs_scale=255.0
+normalize_input=True
+normalize_input_keys=None
+decorrelate_experience_max_seconds=0
+decorrelate_envs_on_one_worker=True
+actor_worker_gpus=[]
+set_workers_cpu_affinity=True
+force_envs_single_thread=False
+default_niceness=0
+log_to_file=True
+experiment_summaries_interval=10
+flush_summaries_interval=30
+stats_avg=100
+summaries_use_frameskip=True
+heartbeat_interval=20
+heartbeat_reporting_interval=600
+train_for_env_steps=10000
+train_for_seconds=10000000000
+save_every_sec=120
+keep_checkpoints=2
+load_checkpoint_kind=latest
+save_milestones_sec=-1
+save_best_every_sec=5
+save_best_metric=reward
+save_best_after=100000
+benchmark=False
+encoder_mlp_layers=[512, 512]
+encoder_conv_architecture=convnet_simple
+encoder_conv_mlp_layers=[512]
+use_rnn=True
+rnn_size=512
+rnn_type=gru
+rnn_num_layers=1
+decoder_mlp_layers=[]
+nonlinearity=elu
+policy_initialization=orthogonal
+policy_init_gain=1.0
+actor_critic_share_weights=True
+adaptive_stddev=True
+continuous_tanh_scale=0.0
+initial_stddev=1.0
+use_env_info_cache=False
+env_gpu_actions=False
+env_gpu_observations=True
+env_frameskip=4
+env_framestack=1
+pixel_format=CHW
+use_record_episode_statistics=False
+with_wandb=False
+wandb_user=None
+wandb_project=sample_factory
+wandb_group=None
+wandb_job_type=SF
+wandb_tags=[]
+with_pbt=False
+pbt_mix_policies_in_one_env=True
+pbt_period_env_steps=5000000
+pbt_start_mutation=20000000
+pbt_replace_fraction=0.3
+pbt_mutation_rate=0.15
+pbt_replace_reward_gap=0.1
+pbt_replace_reward_gap_absolute=1e-06
+pbt_optimize_gamma=False
+pbt_target_objective=true_objective
+pbt_perturb_min=1.1
+pbt_perturb_max=1.5
+num_agents=-1
+num_humans=0
+num_bots=-1
+start_bot_difficulty=None
+timelimit=None
+res_w=128
+res_h=72
+wide_aspect_ratio=False
+eval_env_frameskip=1
+fps=35
+command_line=--env=doom_health_gathering_supreme --num_workers=8 --num_envs_per_worker=4 --train_for_env_steps=4000000
+cli_args={'env': 'doom_health_gathering_supreme', 'num_workers': 8, 'num_envs_per_worker': 4, 'train_for_env_steps': 4000000}
+git_hash=unknown
+git_repo_name=not a git repository
+[2025-09-02 09:53:25,608][02807] Saving configuration to /content/train_dir/default_experiment/config.json...
+[2025-09-02 09:53:25,610][02807] Rollout worker 0 uses device cpu
+[2025-09-02 09:53:25,611][02807] Rollout worker 1 uses device cpu
+[2025-09-02 09:53:25,612][02807] Rollout worker 2 uses device cpu
+[2025-09-02 09:53:25,613][02807] Rollout worker 3 uses device cpu
+[2025-09-02 09:53:25,614][02807] Rollout worker 4 uses device cpu
+[2025-09-02 09:53:25,615][02807] Rollout worker 5 uses device cpu
+[2025-09-02 09:53:25,616][02807] Rollout worker 6 uses device cpu
+[2025-09-02 09:53:25,617][02807] Rollout worker 7 uses device cpu
+[2025-09-02 09:53:25,690][02807] InferenceWorker_p0-w0: min num requests: 2
+[2025-09-02 09:53:25,722][02807] Starting all processes...
+[2025-09-02 09:53:25,723][02807] Starting process learner_proc0
+[2025-09-02 09:53:25,798][02807] Starting all processes...
+[2025-09-02 09:53:25,804][02807] Starting process inference_proc0-0
+[2025-09-02 09:53:25,804][02807] Starting process rollout_proc0
+[2025-09-02 09:53:25,805][02807] Starting process rollout_proc1
+[2025-09-02 09:53:25,805][02807] Starting process rollout_proc2
+[2025-09-02 09:53:25,805][02807] Starting process rollout_proc3
+[2025-09-02 09:53:25,805][02807] Starting process rollout_proc4
+[2025-09-02 09:53:25,805][02807] Starting process rollout_proc5
+[2025-09-02 09:53:25,805][02807] Starting process rollout_proc6
+[2025-09-02 09:53:25,805][02807] Starting process rollout_proc7
+[2025-09-02 09:53:45,617][06913] Starting seed is not provided
+[2025-09-02 09:53:45,618][06913] Initializing actor-critic model on device cpu
+[2025-09-02 09:53:45,619][06913] RunningMeanStd input shape: (3, 72, 128)
+[2025-09-02 09:53:45,621][06913] RunningMeanStd input shape: (1,)
+[2025-09-02 09:53:45,683][02807] Heartbeat connected on Batcher_0
+[2025-09-02 09:53:45,719][06913] ConvEncoder: input_channels=3
+[2025-09-02 09:53:45,846][06933] Worker 7 uses CPU cores [1]
+[2025-09-02 09:53:45,881][02807] Heartbeat connected on RolloutWorker_w7
+[2025-09-02 09:53:45,899][06929] Worker 3 uses CPU cores [1]
+[2025-09-02 09:53:45,923][02807] Heartbeat connected on RolloutWorker_w3
+[2025-09-02 09:53:45,957][06932] Worker 6 uses CPU cores [0]
+[2025-09-02 09:53:45,965][02807] Heartbeat connected on RolloutWorker_w6
+[2025-09-02 09:53:46,157][06934] Worker 5 uses CPU cores [1]
+[2025-09-02 09:53:46,167][06927] Worker 0 uses CPU cores [0]
+[2025-09-02 09:53:46,171][02807] Heartbeat connected on RolloutWorker_w5
+[2025-09-02 09:53:46,184][06926] Worker 1 uses CPU cores [1]
+[2025-09-02 09:53:46,196][02807] Heartbeat connected on RolloutWorker_w1
+[2025-09-02 09:53:46,197][02807] Heartbeat connected on RolloutWorker_w0
+[2025-09-02 09:53:46,218][06928] Worker 2 uses CPU cores [0]
+[2025-09-02 09:53:46,224][02807] Heartbeat connected on RolloutWorker_w2
+[2025-09-02 09:53:46,269][02807] Heartbeat connected on InferenceWorker_p0-w0
+[2025-09-02 09:53:46,297][06931] Worker 4 uses CPU cores [0]
+[2025-09-02 09:53:46,299][02807] Heartbeat connected on RolloutWorker_w4
+[2025-09-02 09:53:46,325][06913] Conv encoder output size: 512
+[2025-09-02 09:53:46,325][06913] Policy head output size: 512
+[2025-09-02 09:53:46,342][06913] Created Actor Critic model with architecture:
+[2025-09-02 09:53:46,342][06913] ActorCriticSharedWeights(
+  (obs_normalizer): ObservationNormalizer(
+    (running_mean_std): RunningMeanStdDictInPlace(
+      (running_mean_std): ModuleDict(
+        (obs): RunningMeanStdInPlace()
+      )
+    )
+  )
+  (returns_normalizer): RecursiveScriptModule(original_name=RunningMeanStdInPlace)
+  (encoder): VizdoomEncoder(
+    (basic_encoder): ConvEncoder(
+      (enc): RecursiveScriptModule(
+        original_name=ConvEncoderImpl
+        (conv_head): RecursiveScriptModule(
+          original_name=Sequential
+          (0): RecursiveScriptModule(original_name=Conv2d)
+          (1): RecursiveScriptModule(original_name=ELU)
+          (2): RecursiveScriptModule(original_name=Conv2d)
+          (3): RecursiveScriptModule(original_name=ELU)
+          (4): RecursiveScriptModule(original_name=Conv2d)
+          (5): RecursiveScriptModule(original_name=ELU)
+        )
+        (mlp_layers): RecursiveScriptModule(
+          original_name=Sequential
+          (0): RecursiveScriptModule(original_name=Linear)
+          (1): RecursiveScriptModule(original_name=ELU)
+        )
+      )
+    )
+  )
+  (core): ModelCoreRNN(
+    (core): GRU(512, 512)
+  )
+  (decoder): MlpDecoder(
+    (mlp): Identity()
+  )
+  (critic_linear): Linear(in_features=512, out_features=1, bias=True)
+  (action_parameterization): ActionParameterizationDefault(
+    (distribution_linear): Linear(in_features=512, out_features=5, bias=True)
+  )
+)
+[2025-09-02 09:53:46,691][06913] Using optimizer <class 'torch.optim.adam.Adam'>
+[2025-09-02 09:53:48,476][06913] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000003_12288.pth...
+[2025-09-02 09:53:48,535][06913] Loading model from checkpoint
+[2025-09-02 09:53:48,541][06913] Loaded experiment state at self.train_step=3, self.env_steps=12288
+[2025-09-02 09:53:48,541][06913] Initialized policy 0 weights for model version 3
+[2025-09-02 09:53:48,547][06930] RunningMeanStd input shape: (3, 72, 128)
+[2025-09-02 09:53:48,548][06930] RunningMeanStd input shape: (1,)
+[2025-09-02 09:53:48,556][06913] LearnerWorker_p0 finished initialization!
+[2025-09-02 09:53:48,561][02807] Heartbeat connected on LearnerWorker_p0
+[2025-09-02 09:53:48,599][06930] ConvEncoder: input_channels=3
+[2025-09-02 09:53:48,812][06930] Conv encoder output size: 512
+[2025-09-02 09:53:48,815][06930] Policy head output size: 512
+[2025-09-02 09:53:48,852][02807] Inference worker 0-0 is ready!
+[2025-09-02 09:53:48,853][02807] All inference workers are ready! Signal rollout workers to start!
+[2025-09-02 09:53:49,154][06929] Doom resolution: 160x120, resize resolution: (128, 72)
+[2025-09-02 09:53:49,156][06933] Doom resolution: 160x120, resize resolution: (128, 72)
+[2025-09-02 09:53:49,162][06926] Doom resolution: 160x120, resize resolution: (128, 72)
+[2025-09-02 09:53:49,163][06934] Doom resolution: 160x120, resize resolution: (128, 72)
+[2025-09-02 09:53:49,264][06931] Doom resolution: 160x120, resize resolution: (128, 72)
+[2025-09-02 09:53:49,270][06932] Doom resolution: 160x120, resize resolution: (128, 72)
+[2025-09-02 09:53:49,275][06927] Doom resolution: 160x120, resize resolution: (128, 72)
+[2025-09-02 09:53:49,270][06928] Doom resolution: 160x120, resize resolution: (128, 72)
+[2025-09-02 09:53:50,355][06933] Decorrelating experience for 0 frames...
+[2025-09-02 09:53:50,840][06928] Decorrelating experience for 0 frames...
+[2025-09-02 09:53:50,848][06927] Decorrelating experience for 0 frames...
+[2025-09-02 09:53:50,855][06931] Decorrelating experience for 0 frames...
+[2025-09-02 09:53:51,241][06933] Decorrelating experience for 32 frames...
+[2025-09-02 09:53:51,324][06929] Decorrelating experience for 0 frames...
+[2025-09-02 09:53:51,738][06931] Decorrelating experience for 32 frames...
+[2025-09-02 09:53:51,741][06927] Decorrelating experience for 32 frames...
+[2025-09-02 09:53:52,340][06929] Decorrelating experience for 32 frames...
+[2025-09-02 09:53:52,374][06926] Decorrelating experience for 0 frames...
+[2025-09-02 09:53:52,783][02807] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 12288. Throughput: 0: nan. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2025-09-02 09:53:53,520][06932] Decorrelating experience for 0 frames...
+[2025-09-02 09:53:53,539][06929] Decorrelating experience for 64 frames...
+[2025-09-02 09:53:54,094][06927] Decorrelating experience for 64 frames...
+[2025-09-02 09:53:54,096][06931] Decorrelating experience for 64 frames...
+[2025-09-02 09:53:54,589][06928] Decorrelating experience for 32 frames...
+[2025-09-02 09:53:55,140][06933] Decorrelating experience for 64 frames...
+[2025-09-02 09:53:55,319][06932] Decorrelating experience for 32 frames...
+[2025-09-02 09:53:55,346][06929] Decorrelating experience for 96 frames...
+[2025-09-02 09:53:55,920][06931] Decorrelating experience for 96 frames...
+[2025-09-02 09:53:56,155][06926] Decorrelating experience for 32 frames...
+[2025-09-02 09:53:56,743][06928] Decorrelating experience for 64 frames...
+[2025-09-02 09:53:57,345][06934] Decorrelating experience for 0 frames...
+[2025-09-02 09:53:57,380][06933] Decorrelating experience for 96 frames...
+[2025-09-02 09:53:57,783][02807] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 12288. Throughput: 0: 0.0. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2025-09-02 09:53:57,785][02807] Avg episode reward: [(0, '1.440')]
+[2025-09-02 09:53:57,906][06927] Decorrelating experience for 96 frames...
+[2025-09-02 09:53:58,494][06932] Decorrelating experience for 64 frames...
+[2025-09-02 09:53:58,797][06926] Decorrelating experience for 64 frames...
+[2025-09-02 09:53:59,833][06926] Decorrelating experience for 96 frames...
+[2025-09-02 09:53:59,863][06928] Decorrelating experience for 96 frames...
+[2025-09-02 09:54:02,783][02807] Fps is (10 sec: 0.0, 60 sec: 0.0, 300 sec: 0.0). Total num frames: 12288. Throughput: 0: 133.0. Samples: 1330. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2025-09-02 09:54:02,785][02807] Avg episode reward: [(0, '2.453')]
+[2025-09-02 09:54:03,868][06932] Decorrelating experience for 96 frames...
+[2025-09-02 09:54:04,999][06934] Decorrelating experience for 32 frames...
+[2025-09-02 09:54:06,481][06913] Signal inference workers to stop experience collection...
+[2025-09-02 09:54:06,526][06930] InferenceWorker_p0-w0: stopping experience collection
+[2025-09-02 09:54:07,030][06934] Decorrelating experience for 64 frames...
+[2025-09-02 09:54:07,600][06934] Decorrelating experience for 96 frames...
+[2025-09-02 09:54:07,754][06913] Signal inference workers to resume experience collection...
+[2025-09-02 09:54:07,761][06913] Stopping Batcher_0...
+[2025-09-02 09:54:07,762][02807] Component Batcher_0 stopped!
+[2025-09-02 09:54:07,768][06913] Loop batcher_evt_loop terminating...
+[2025-09-02 09:54:07,804][06930] Weights refcount: 2 0
+[2025-09-02 09:54:07,810][02807] Component InferenceWorker_p0-w0 stopped!
+[2025-09-02 09:54:07,818][06930] Stopping InferenceWorker_p0-w0...
+[2025-09-02 09:54:07,819][06930] Loop inference_proc0-0_evt_loop terminating...
+[2025-09-02 09:54:08,117][06926] Stopping RolloutWorker_w1...
+[2025-09-02 09:54:08,122][06926] Loop rollout_proc1_evt_loop terminating...
+[2025-09-02 09:54:08,117][02807] Component RolloutWorker_w1 stopped!
+[2025-09-02 09:54:08,144][06933] Stopping RolloutWorker_w7...
+[2025-09-02 09:54:08,147][06933] Loop rollout_proc7_evt_loop terminating...
+[2025-09-02 09:54:08,144][02807] Component RolloutWorker_w7 stopped!
+[2025-09-02 09:54:08,168][06934] Stopping RolloutWorker_w5...
+[2025-09-02 09:54:08,168][02807] Component RolloutWorker_w5 stopped!
+[2025-09-02 09:54:08,169][06934] Loop rollout_proc5_evt_loop terminating...
+[2025-09-02 09:54:08,180][06929] Stopping RolloutWorker_w3...
+[2025-09-02 09:54:08,190][06929] Loop rollout_proc3_evt_loop terminating...
+[2025-09-02 09:54:08,180][02807] Component RolloutWorker_w3 stopped!
+[2025-09-02 09:54:08,303][02807] Component RolloutWorker_w2 stopped!
+[2025-09-02 09:54:08,308][06928] Stopping RolloutWorker_w2...
+[2025-09-02 09:54:08,317][02807] Component RolloutWorker_w4 stopped!
+[2025-09-02 09:54:08,321][06931] Stopping RolloutWorker_w4...
+[2025-09-02 09:54:08,322][06931] Loop rollout_proc4_evt_loop terminating...
+[2025-09-02 09:54:08,326][02807] Component RolloutWorker_w6 stopped!
+[2025-09-02 09:54:08,329][06932] Stopping RolloutWorker_w6...
+[2025-09-02 09:54:08,330][06932] Loop rollout_proc6_evt_loop terminating...
+[2025-09-02 09:54:08,314][06928] Loop rollout_proc2_evt_loop terminating...
+[2025-09-02 09:54:08,347][02807] Component RolloutWorker_w0 stopped!
+[2025-09-02 09:54:08,351][06927] Stopping RolloutWorker_w0...
+[2025-09-02 09:54:08,352][06927] Loop rollout_proc0_evt_loop terminating...
+[2025-09-02 09:54:13,196][06913] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000005_20480.pth...
+[2025-09-02 09:54:13,259][06913] Saving /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000005_20480.pth...
+[2025-09-02 09:54:13,366][06913] Stopping LearnerWorker_p0...
+[2025-09-02 09:54:13,367][06913] Loop learner_proc0_evt_loop terminating...
+[2025-09-02 09:54:13,366][02807] Component LearnerWorker_p0 stopped!
+[2025-09-02 09:54:13,371][02807] Waiting for process learner_proc0 to stop...
+[2025-09-02 09:54:14,291][02807] Waiting for process inference_proc0-0 to join...
+[2025-09-02 09:54:14,293][02807] Waiting for process rollout_proc0 to join...
+[2025-09-02 09:54:14,296][02807] Waiting for process rollout_proc1 to join...
+[2025-09-02 09:54:14,297][02807] Waiting for process rollout_proc2 to join...
+[2025-09-02 09:54:14,299][02807] Waiting for process rollout_proc3 to join...
+[2025-09-02 09:54:14,300][02807] Waiting for process rollout_proc4 to join...
+[2025-09-02 09:54:14,301][02807] Waiting for process rollout_proc5 to join...
+[2025-09-02 09:54:14,302][02807] Waiting for process rollout_proc6 to join...
+[2025-09-02 09:54:14,305][02807] Waiting for process rollout_proc7 to join...
+[2025-09-02 09:54:14,306][02807] Batcher 0 profile tree view:
+batching: 0.0391, releasing_batches: 0.0067
+[2025-09-02 09:54:14,308][02807] InferenceWorker_p0-w0 profile tree view:
+update_model: 0.0657
+wait_policy: 0.0001
+  wait_policy_total: 7.4236
+one_step: 0.0267
+  handle_policy_step: 9.7252
+    deserialize: 0.3177, stack: 0.0320, obs_to_device_normalize: 1.0536, forward: 7.7469, send_messages: 0.2031
+    prepare_outputs: 0.1186
+      to_cpu: 0.0155
+[2025-09-02 09:54:14,309][02807] Learner 0 profile tree view:
+misc: 0.0000, prepare_batch: 2.8606
+train: 9.2517
+  epoch_init: 0.0000, minibatch_init: 0.0000, losses_postprocess: 0.0027, kl_divergence: 0.0009, after_optimizer: 0.0044
+  calculate_losses: 4.0182
+    losses_init: 0.0000, forward_head: 3.5911, bptt_initial: 0.0130, tail: 0.0037, advantages_returns: 0.0010, losses: 0.0029
+    bptt: 0.4059
+      bptt_forward_core: 0.4034
+  update: 5.2216
+    clip: 0.0072
+[2025-09-02 09:54:14,310][02807] RolloutWorker_w0 profile tree view:
+wait_for_trajectories: 0.0010, enqueue_policy_requests: 0.1876, env_step: 3.5443, overhead: 0.0776, complete_rollouts: 0.0482
+save_policy_outputs: 0.0791
+  split_output_tensors: 0.0251
+[2025-09-02 09:54:14,311][02807] RolloutWorker_w7 profile tree view:
+wait_for_trajectories: 0.0010, enqueue_policy_requests: 0.2246, env_step: 4.1101, overhead: 0.0910, complete_rollouts: 0.0875
+save_policy_outputs: 0.0999
+  split_output_tensors: 0.0628
+[2025-09-02 09:54:14,312][02807] Loop Runner_EvtLoop terminating...
+[2025-09-02 09:54:14,313][02807] Runner profile tree view:
+main_loop: 48.5912
+[2025-09-02 09:54:14,314][02807] Collected {0: 20480}, FPS: 168.6
+[2025-09-02 10:01:30,977][02807] Loading existing experiment configuration from /content/train_dir/default_experiment/config.json
+[2025-09-02 10:01:30,979][02807] Overriding arg 'num_workers' with value 1 passed from command line
+[2025-09-02 10:01:30,981][02807] Adding new argument 'no_render'=True that is not in the saved config file!
+[2025-09-02 10:01:30,982][02807] Adding new argument 'save_video'=True that is not in the saved config file!
+[2025-09-02 10:01:30,983][02807] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
+[2025-09-02 10:01:30,985][02807] Adding new argument 'video_name'=None that is not in the saved config file!
+[2025-09-02 10:01:30,986][02807] Adding new argument 'max_num_frames'=100000 that is not in the saved config file!
+[2025-09-02 10:01:30,987][02807] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
+[2025-09-02 10:01:30,988][02807] Adding new argument 'push_to_hub'=True that is not in the saved config file!
+[2025-09-02 10:01:30,989][02807] Adding new argument 'hf_repository'='WangChongan/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file!
+[2025-09-02 10:01:30,990][02807] Adding new argument 'policy_index'=0 that is not in the saved config file!
+[2025-09-02 10:01:30,991][02807] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
+[2025-09-02 10:01:30,992][02807] Adding new argument 'train_script'=None that is not in the saved config file!
+[2025-09-02 10:01:30,993][02807] Adding new argument 'enjoy_script'=None that is not in the saved config file!
+[2025-09-02 10:01:30,994][02807] Using frameskip 1 and render_action_repeat=4 for evaluation
+[2025-09-02 10:01:31,023][02807] RunningMeanStd input shape: (3, 72, 128)
+[2025-09-02 10:01:31,025][02807] RunningMeanStd input shape: (1,)
+[2025-09-02 10:01:31,036][02807] ConvEncoder: input_channels=3
+[2025-09-02 10:01:31,074][02807] Conv encoder output size: 512
+[2025-09-02 10:01:31,076][02807] Policy head output size: 512
+[2025-09-02 10:01:31,092][02807] Loading state from checkpoint /content/train_dir/default_experiment/checkpoint_p0/checkpoint_000000005_20480.pth...
+[2025-09-02 10:01:31,614][02807] Num frames 100...
+[2025-09-02 10:01:31,834][02807] Num frames 200...
+[2025-09-02 10:01:32,036][02807] Num frames 300...
+[2025-09-02 10:01:32,261][02807] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2025-09-02 10:01:32,263][02807] Avg episode reward: 3.840, avg true_objective: 3.840
+[2025-09-02 10:01:32,297][02807] Num frames 400...
+[2025-09-02 10:01:32,497][02807] Num frames 500...
+[2025-09-02 10:01:32,706][02807] Num frames 600...
+[2025-09-02 10:01:32,905][02807] Num frames 700...
+[2025-09-02 10:01:33,096][02807] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2025-09-02 10:01:33,098][02807] Avg episode reward: 3.840, avg true_objective: 3.840
+[2025-09-02 10:01:33,167][02807] Num frames 800...
+[2025-09-02 10:01:33,364][02807] Num frames 900...
+[2025-09-02 10:01:33,564][02807] Num frames 1000...
+[2025-09-02 10:01:33,770][02807] Num frames 1100...
+[2025-09-02 10:01:33,928][02807] Avg episode rewards: #0: 3.840, true rewards: #0: 3.840
+[2025-09-02 10:01:33,929][02807] Avg episode reward: 3.840, avg true_objective: 3.840
+[2025-09-02 10:01:34,029][02807] Num frames 1200...
+[2025-09-02 10:01:34,236][02807] Num frames 1300...
+[2025-09-02 10:01:34,435][02807] Num frames 1400...
+[2025-09-02 10:01:34,633][02807] Num frames 1500...
+[2025-09-02 10:01:34,839][02807] Avg episode rewards: #0: 4.170, true rewards: #0: 3.920
+[2025-09-02 10:01:34,840][02807] Avg episode reward: 4.170, avg true_objective: 3.920
+[2025-09-02 10:01:34,907][02807] Num frames 1600...
+[2025-09-02 10:01:35,105][02807] Num frames 1700...
+[2025-09-02 10:01:35,307][02807] Num frames 1800...
+[2025-09-02 10:01:35,504][02807] Avg episode rewards: #0: 3.942, true rewards: #0: 3.742
+[2025-09-02 10:01:35,505][02807] Avg episode reward: 3.942, avg true_objective: 3.742
+[2025-09-02 10:01:35,567][02807] Num frames 1900...
+[2025-09-02 10:01:35,768][02807] Num frames 2000...
+[2025-09-02 10:01:35,985][02807] Num frames 2100...
+[2025-09-02 10:01:36,193][02807] Num frames 2200...
+[2025-09-02 10:01:36,359][02807] Avg episode rewards: #0: 3.925, true rewards: #0: 3.758
+[2025-09-02 10:01:36,360][02807] Avg episode reward: 3.925, avg true_objective: 3.758
+[2025-09-02 10:01:36,448][02807] Num frames 2300...
+[2025-09-02 10:01:36,645][02807] Num frames 2400...
+[2025-09-02 10:01:36,862][02807] Num frames 2500...
+[2025-09-02 10:01:37,148][02807] Num frames 2600...
+[2025-09-02 10:01:37,319][02807] Avg episode rewards: #0: 3.913, true rewards: #0: 3.770
+[2025-09-02 10:01:37,322][02807] Avg episode reward: 3.913, avg true_objective: 3.770
+[2025-09-02 10:01:37,489][02807] Num frames 2700...
+[2025-09-02 10:01:37,755][02807] Num frames 2800...
+[2025-09-02 10:01:38,039][02807] Num frames 2900...
+[2025-09-02 10:01:38,320][02807] Num frames 3000...
+[2025-09-02 10:01:38,442][02807] Avg episode rewards: #0: 3.904, true rewards: #0: 3.779
+[2025-09-02 10:01:38,449][02807] Avg episode reward: 3.904, avg true_objective: 3.779
+[2025-09-02 10:01:38,673][02807] Num frames 3100...
+[2025-09-02 10:01:38,964][02807] Num frames 3200...
+[2025-09-02 10:01:39,274][02807] Num frames 3300...
+[2025-09-02 10:01:39,555][02807] Num frames 3400...
+[2025-09-02 10:01:39,753][02807] Num frames 3500...
+[2025-09-02 10:01:39,943][02807] Avg episode rewards: #0: 4.297, true rewards: #0: 3.963
+[2025-09-02 10:01:39,944][02807] Avg episode reward: 4.297, avg true_objective: 3.963
+[2025-09-02 10:01:40,012][02807] Num frames 3600...
+[2025-09-02 10:01:40,230][02807] Num frames 3700...
+[2025-09-02 10:01:40,434][02807] Num frames 3800...
+[2025-09-02 10:01:40,638][02807] Num frames 3900...
+[2025-09-02 10:01:40,798][02807] Avg episode rewards: #0: 4.251, true rewards: #0: 3.951
+[2025-09-02 10:01:40,799][02807] Avg episode reward: 4.251, avg true_objective: 3.951
+[2025-09-02 10:02:06,456][02807] Replay video saved to /content/train_dir/default_experiment/replay.mp4!