Upload folder using huggingface_hub

.summary/0/events.out.tfevents.1696863903.rhmmedcatt-proliant-ml350-gen10

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c4a74cc34c773abfc6dfbf77062939cd809b3ca0ee5a9564d23f341b5dc189a
+size 8560456

.summary/1/events.out.tfevents.1696863903.rhmmedcatt-proliant-ml350-gen10

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:277ca59b1cb8e08678d84a7d9d7a4fd317e00bd21f6f571553749ebf590d4494
+size 4927650

README.md

@@ -15,7 +15,7 @@ model-index:
       type: atari_boxing
     metrics:
     - type: mean_reward
-      value: 48.30 +/- 11.05
+      value: 96.80 +/- 4.75
       name: mean_reward
       verified: false
 ---
@@ -30,15 +30,128 @@ Documentation for how to use Sample-Factory can be found at https://www.samplefa
 
 After installing Sample-Factory, download the model with:
 ```
-python -m sample_factory.huggingface.load_from_hub -r MattStammers/appo-atari_boxing
+python -m sample_factory.huggingface.load_from_hub -r MattStammers/APPO-atari_boxing
 ```
 
     
+## About the Model
+
+This model as with all the others in the benchmarks was trained initially asynchronously un-seeded to 10 million steps for the purposes of setting a sample factory async baseline for this model on this environment but only 3/57 made it. 
+
+The aim is to reach state-of-the-art (SOTA) performance on each atari environment. I will flag the models with SOTA when they reach at or near these levels. 
+
+The hyperparameters used in the model are the ones I have pushed to my fork of sample-factory: https://github.com/MattStammers/sample-factory. Given that https://huggingface.co/edbeeching has kindly shared his.
+I saved time and energy by using many of his tuned hyperparameters to maximise performance. However, he used 2 billion training steps. I have started as explained above at 10 million then moved to 100m to see how performance goes:
+```
+hyperparameters =  {
+  "device": "gpu",
+  "seed": 1234,
+  "num_policies": 2,
+  "async_rl": true,
+  "serial_mode": false,
+  "batched_sampling": true,
+  "num_batches_to_accumulate": 2,
+  "worker_num_splits": 1,
+  "policy_workers_per_policy": 1,
+  "max_policy_lag": 1000,
+  "num_workers": 16,
+  "num_envs_per_worker": 2,
+  "batch_size": 1024,
+  "num_batches_per_epoch": 8,
+  "num_epochs": 4,
+  "rollout": 128,
+  "recurrence": 1,
+  "shuffle_minibatches": false,
+  "gamma": 0.99,
+  "reward_scale": 1.0,
+  "reward_clip": 1000.0,
+  "value_bootstrap": false,
+  "normalize_returns": true,
+  "exploration_loss_coeff": 0.0004677351413,
+  "value_loss_coeff": 0.5,
+  "kl_loss_coeff": 0.0,
+  "exploration_loss": "entropy",
+  "gae_lambda": 0.95,
+  "ppo_clip_ratio": 0.1,
+  "ppo_clip_value": 1.0,
+  "with_vtrace": false,
+  "vtrace_rho": 1.0,
+  "vtrace_c": 1.0,
+  "optimizer": "adam",
+  "adam_eps": 1e-05,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.999,
+  "max_grad_norm": 0.0,
+  "learning_rate": 0.0003033891184,
+  "lr_schedule": "linear_decay",
+  "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": [
+    "obs"
+  ],
+  "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": 3,
+  "flush_summaries_interval": 30,
+  "stats_avg": 100,
+  "summaries_use_frameskip": true,
+  "heartbeat_interval": 10,
+  "heartbeat_reporting_interval": 60,
+  "train_for_env_steps": 100000000,
+  "train_for_seconds": 10000000000,
+  "save_every_sec": 120,
+  "keep_checkpoints": 2,
+  "load_checkpoint_kind": "latest",
+  "save_milestones_sec": 1200,
+  "save_best_every_sec": 5,
+  "save_best_metric": "reward",
+  "save_best_after": 100000,
+  "benchmark": false,
+  "encoder_mlp_layers": [
+    512,
+    512
+  ],
+  "encoder_conv_architecture": "convnet_atari",
+  "encoder_conv_mlp_layers": [
+    512
+  ],
+  "use_rnn": false,
+  "rnn_size": 512,
+  "rnn_type": "gru",
+  "rnn_num_layers": 1,
+  "decoder_mlp_layers": [],
+  "nonlinearity": "relu",
+  "policy_initialization": "orthogonal",
+  "policy_init_gain": 1.0,
+  "actor_critic_share_weights": true,
+  "adaptive_stddev": false,
+  "continuous_tanh_scale": 0.0,
+  "initial_stddev": 1.0,
+  "use_env_info_cache": false,
+  "env_gpu_actions": false,
+  "env_gpu_observations": true,
+  "env_frameskip": 4,
+  "env_framestack": 4,
+  }
+
+  ```
+
+
+    
 ## Using the model
 
 To run the model after download, use the `enjoy` script corresponding to this environment:
 ```
-python -m sf_examples.atari.enjoy_atari --algo=APPO --env=atari_boxing --train_dir=./train_dir --experiment=appo-atari_boxing
+python -m sf_examples.atari.enjoy_atari --algo=APPO --env=atari_boxing --train_dir=./train_dir --experiment=APPO-atari_boxing
 ```
 
 
@@ -49,7 +162,7 @@ See https://www.samplefactory.dev/10-huggingface/huggingface/ for more details
 
 To continue training with this model, use the `train` script corresponding to this environment:
 ```
-python -m sf_examples.atari.train_atari --algo=APPO --env=atari_boxing --train_dir=./train_dir --experiment=appo-atari_boxing --restart_behavior=resume --train_for_env_steps=10000000000
+python -m sf_examples.atari.train_atari --algo=APPO --env=atari_boxing --train_dir=./train_dir --experiment=APPO-atari_boxing --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_000092928_95158272_reward_95.170.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34ddc84ec2a904e8d31688ca4a74f7ffc3cefda1592908a6847a416d43c5a136
+size 20795763

checkpoint_p0/checkpoint_000097632_99975168.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:761242c18dbd237a72ef261ee778a0bca017b45786945c47bfddfee929247b8f
+size 20796099

checkpoint_p0/checkpoint_000097664_100007936.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f15d5f7157a377656e50e453c1de879c4b1db825d8460acdc0ceefded6357151
+size 20796099

checkpoint_p0/milestones/checkpoint_000007392_7569408.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:327fec4ffdff9d9968bb6fb755b19459d9392344362841d41081a9a8f38246a2
+size 20796955

checkpoint_p0/milestones/checkpoint_000014848_15204352.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d10515f1b10b74be981196f3240045943143b12034dafc9beb5532c0e2e1568
+size 20797011

checkpoint_p0/milestones/checkpoint_000022336_22872064.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:317c981de988b275585b9507de8de0ca3e6e968865a2c6b5e5b8a7fc552fc6ac
+size 20797011

checkpoint_p0/milestones/checkpoint_000029792_30507008.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db9eb58d74d1ee57d04433730da2eebb400e47bfa54dc6887e0cfcf120b99442
+size 20797011

checkpoint_p0/milestones/checkpoint_000037248_38141952.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa00708e25adc4b5b2cec8c41baca8854e338b5badd7a695f7a7cd662c1aaf6f
+size 20797011

checkpoint_p0/milestones/checkpoint_000044704_45776896.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8885bf8319b48e95e1e543bb24cefa590c7de054b31d4bbd8f86fa7da114847a
+size 20797011

checkpoint_p0/milestones/checkpoint_000052160_53411840.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c32188826791503ab19d2cf76efc9b19e217d5bb5e9ad12562297c59d002a841
+size 20797011

checkpoint_p0/milestones/checkpoint_000059616_61046784.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:362a636b30775a7427bb25e926819878435f525e87fe207d1839d5d9fcac1045
+size 20797011

checkpoint_p0/milestones/checkpoint_000067072_68681728.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4494fb512637c936c30abfe3cf3e79753dcba9d1c0bc348a191383f29b553ebe
+size 20797011

checkpoint_p0/milestones/checkpoint_000074528_76316672.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d178f6a8be246fc835fb6363fd74cbe46b435c0518ec1ccffe75a1d32213d6f6
+size 20797011

checkpoint_p0/milestones/checkpoint_000081984_83951616.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ab3823bfdeaa1b9397479a8e28f3171ce73f51469c9ad43f5aa81bd0ab31177
+size 20797011

checkpoint_p0/milestones/checkpoint_000089440_91586560.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8813dea6c88ffc030e6c59dab8a277d0e8a194c0bda74cde51210f9ff1f6c8c1
+size 20797011

checkpoint_p0/milestones/checkpoint_000096896_99221504.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38d2ce7923ebf11e7f3b0549522fda8b4934478ac0f61035ac9e577f9d15c82d
+size 20797011

checkpoint_p1/best_000095104_97386496_reward_95.970.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a28f02a4258d46415f094a13f241ffebd441987f6f3c5fd0cea6fef57d85a2d
+size 20795763

checkpoint_p1/checkpoint_000098464_100827136.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b6def9ab7434a0c4120b231f9f9f1ed433e8630a6682c50f70e27ae89f20094
+size 20796099

checkpoint_p1/checkpoint_000098496_100859904.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d67d1f46a3800b57cba9ec679a7ed0082614ad09f56eb0b9f0cd38a6136ca4a7
+size 20796099

checkpoint_p1/milestones/checkpoint_000007488_7667712.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d11b2f8c6493c22ae5d5acbdcd2af043fc75fe129396e484392a2d3b16276781
+size 20796955

checkpoint_p1/milestones/checkpoint_000015040_15400960.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe12917be03c5a722f587713347a9c278c530a8efa5193761778f1bf63ddfef0
+size 20797011

checkpoint_p1/milestones/checkpoint_000022592_23134208.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:551ca53149597862890142278882e937ae5e3855c6cce779e5c7797781113491
+size 20797011

checkpoint_p1/milestones/checkpoint_000030112_30834688.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9a4f54585245e13404409e3e1f0999a5f96a4b64478189c5ae0b9d0025df48b
+size 20797011

checkpoint_p1/milestones/checkpoint_000037632_38535168.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0ce65436dae63ec17e57791cbbefa4964a4c7a8947705ffceabfcda1647b66e
+size 20797011

checkpoint_p1/milestones/checkpoint_000045152_46235648.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e2b21b00e9ad410761b4140f0be6af777c860dc4f4a731fb9f44d89149773fe
+size 20797011

checkpoint_p1/milestones/checkpoint_000052672_53936128.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3df89c74e3d579a558deaa43da534b25f9f172e6b636e762b4162b105ad8e2f0
+size 20797011

checkpoint_p1/milestones/checkpoint_000060160_61603840.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5954b3a2569520f314bee588d2a10e5325bc7156ac5bf6a0b5fe2e82ee8a2b6f
+size 20797011

checkpoint_p1/milestones/checkpoint_000067680_69304320.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b82869653aa1e29e634bf2024e2b48a638c43b6e4850ad4984a96cacc3fcca2e
+size 20797011

checkpoint_p1/milestones/checkpoint_000075200_77004800.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af426e6892bdabdb51f62eefda554eeff6cab5bef36dfca63b10b4f4bdda269f
+size 20797011

checkpoint_p1/milestones/checkpoint_000082688_84672512.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec40b4575bedc0da4e0b2f201ac1c9264bad65b64f6624513d7b787c6af655b5
+size 20797011

checkpoint_p1/milestones/checkpoint_000090208_92372992.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67909043475f80456f74b9c651971f5b99af82c0271bf7360ef01a6a513262c2
+size 20797011

checkpoint_p1/milestones/checkpoint_000097696_100040704.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e917fdfcd08ae2e10ab91ca9a04814a22717d530b5db77fa6749d13b55c0bfe6
+size 20797067

config.json

@@ -2,23 +2,23 @@
   "help": false,
   "algo": "APPO",
   "env": "atari_boxing",
-  "experiment": "atari_boxing",
+  "experiment": "atari_boxing_APPO",
   "train_dir": "./train_atari",
-  "restart_behavior": "resume",
+  "restart_behavior": "restart",
   "device": "gpu",
-  "seed": null,
+  "seed": 1234,
   "num_policies": 2,
-  "async_rl": false,
+  "async_rl": true,
   "serial_mode": false,
-  "batched_sampling": false,
+  "batched_sampling": true,
   "num_batches_to_accumulate": 2,
   "worker_num_splits": 1,
   "policy_workers_per_policy": 1,
   "max_policy_lag": 1000,
-  "num_workers": 8,
-  "num_envs_per_worker": 1,
-  "batch_size": 256,
-  "num_batches_per_epoch": 4,
+  "num_workers": 16,
+  "num_envs_per_worker": 2,
+  "batch_size": 1024,
+  "num_batches_per_epoch": 8,
   "num_epochs": 4,
   "rollout": 128,
   "recurrence": 1,
@@ -28,7 +28,7 @@
   "reward_clip": 1000.0,
   "value_bootstrap": false,
   "normalize_returns": true,
-  "exploration_loss_coeff": 0.01,
+  "exploration_loss_coeff": 0.0004677351413,
   "value_loss_coeff": 0.5,
   "kl_loss_coeff": 0.0,
   "exploration_loss": "entropy",
@@ -42,8 +42,8 @@
   "adam_eps": 1e-05,
   "adam_beta1": 0.9,
   "adam_beta2": 0.999,
-  "max_grad_norm": 0.5,
-  "learning_rate": 0.00025,
+  "max_grad_norm": 0.0,
+  "learning_rate": 0.0003033891184,
   "lr_schedule": "linear_decay",
   "lr_schedule_kl_threshold": 0.008,
   "lr_adaptive_min": 1e-06,
@@ -51,7 +51,9 @@
   "obs_subtract_mean": 0.0,
   "obs_scale": 255.0,
   "normalize_input": true,
-  "normalize_input_keys": null,
+  "normalize_input_keys": [
+    "obs"
+  ],
   "decorrelate_experience_max_seconds": 0,
   "decorrelate_envs_on_one_worker": true,
   "actor_worker_gpus": [],
@@ -63,14 +65,14 @@
   "flush_summaries_interval": 30,
   "stats_avg": 100,
   "summaries_use_frameskip": true,
-  "heartbeat_interval": 20,
-  "heartbeat_reporting_interval": 180,
-  "train_for_env_steps": 10000000,
+  "heartbeat_interval": 10,
+  "heartbeat_reporting_interval": 60,
+  "train_for_env_steps": 100000000,
   "train_for_seconds": 10000000000,
   "save_every_sec": 120,
   "keep_checkpoints": 2,
   "load_checkpoint_kind": "latest",
-  "save_milestones_sec": -1,
+  "save_milestones_sec": 1200,
   "save_best_every_sec": 5,
   "save_best_metric": "reward",
   "save_best_after": 100000,
@@ -104,7 +106,7 @@
   "use_record_episode_statistics": true,
   "with_wandb": true,
   "wandb_user": "matt-stammers",
-  "wandb_project": "atari",
+  "wandb_project": "atari_APPO",
   "wandb_group": "atari_boxing",
   "wandb_job_type": "SF",
   "wandb_tags": [
@@ -122,16 +124,32 @@
   "pbt_target_objective": "true_objective",
   "pbt_perturb_min": 1.1,
   "pbt_perturb_max": 1.5,
-  "command_line": "--algo=APPO --env=atari_boxing --experiment=atari_boxing --num_policies=2 --train_dir=./train_atari --with_wandb=true --wandb_user=matt-stammers --wandb_project=atari --wandb_group=atari_boxing --wandb_job_type=SF --wandb_tags=atari",
+  "command_line": "--algo=APPO --env=atari_boxing --experiment=atari_boxing_APPO --num_policies=2 --restart_behavior=restart --train_dir=./train_atari --train_for_env_steps=100000000 --seed=1234 --num_workers=16 --num_envs_per_worker=2 --num_batches_per_epoch=8 --async_rl=true --batched_sampling=true --batch_size=1024 --max_grad_norm=0 --learning_rate=0.0003033891184 --heartbeat_interval=10 --heartbeat_reporting_interval=60 --save_milestones_sec=1200 --num_epochs=4 --exploration_loss_coeff=0.0004677351413 --with_wandb=true --wandb_user=matt-stammers --wandb_project=atari_APPO --wandb_group=atari_boxing --wandb_job_type=SF --wandb_tags=atari",
   "cli_args": {
     "algo": "APPO",
     "env": "atari_boxing",
-    "experiment": "atari_boxing",
+    "experiment": "atari_boxing_APPO",
     "train_dir": "./train_atari",
+    "restart_behavior": "restart",
+    "seed": 1234,
     "num_policies": 2,
+    "async_rl": true,
+    "batched_sampling": true,
+    "num_workers": 16,
+    "num_envs_per_worker": 2,
+    "batch_size": 1024,
+    "num_batches_per_epoch": 8,
+    "num_epochs": 4,
+    "exploration_loss_coeff": 0.0004677351413,
+    "max_grad_norm": 0.0,
+    "learning_rate": 0.0003033891184,
+    "heartbeat_interval": 10,
+    "heartbeat_reporting_interval": 60,
+    "train_for_env_steps": 100000000,
+    "save_milestones_sec": 1200,
     "with_wandb": true,
     "wandb_user": "matt-stammers",
-    "wandb_project": "atari",
+    "wandb_project": "atari_APPO",
     "wandb_group": "atari_boxing",
     "wandb_job_type": "SF",
     "wandb_tags": [
@@ -140,5 +158,5 @@
   },
   "git_hash": "5fff97c2f535da5987d358cdbe6927cccd43621e",
   "git_repo_name": "not a git repository",
-  "wandb_unique_id": "atari_boxing_20230925_211117_092055"
+  "wandb_unique_id": "atari_boxing_APPO_20231009_160500_703478"
 }

git.diff

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:28f3fe545f35e5db1e419bf1e198c0bc5ef89459d55f59e2d06d3c108907e732
-size 14403396
+oid sha256:3357904f421d3f4924836316b1741bf64d5dd0e807d5e80ac07059b4c52a7008
+size 14426734

replay.mp4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24efef86bc75ed3df07e30962e2cf35de72bbe5165f6986d0184b78a17da8bbd
-size 3038483
+oid sha256:28ea44a343699018ca701ae1f7b0c242417f88d4207fa39cf282ee22bf2946ff
+size 1841726