Upload folder using huggingface_hub

.summary/0/events.out.tfevents.1696983031.rhmmedcatt-proliant-ml350-gen10

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d37ac961d2840ba02a5ac448a94d96e10541d910179d30f268791676dad5ba4b
+size 8314870

.summary/1/events.out.tfevents.1696983031.rhmmedcatt-proliant-ml350-gen10

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e3868b97e308e4fe9d2a3c1c17a9e79a05c4a583ab05b70220eaa80c0880aae
+size 4784004

README.md

@@ -15,7 +15,7 @@ model-index:
       type: atari_enduro
     metrics:
     - type: mean_reward
-      value: 489.30 +/- 121.86
+      value: 2182.80 +/- 148.82
       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_enduro
+python -m sample_factory.huggingface.load_from_hub -r MattStammers/APPO-atari_enduro
 ```
 
     
+## 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_enduro --train_dir=./train_dir --experiment=appo-atari_enduro
+python -m sf_examples.atari.enjoy_atari --algo=APPO --env=atari_enduro --train_dir=./train_dir --experiment=APPO-atari_enduro
 ```
 
 
@@ -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_enduro --train_dir=./train_dir --experiment=appo-atari_enduro --restart_behavior=resume --train_for_env_steps=10000000000
+python -m sf_examples.atari.train_atari --algo=APPO --env=atari_enduro --train_dir=./train_dir --experiment=APPO-atari_enduro --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_000091424_93618176_reward_2115.710.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6da4f23278f4b1253d5b8d3937fe68eb1ba377dcc7a1089771706fd05ccec30
+size 20740275

checkpoint_p0/checkpoint_000097664_100007936.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25703ab635a4c356a27052bfe9c4288f3d294ba0af0261ed5d9c68f67d38aece
+size 20740611

checkpoint_p0/checkpoint_000097696_100040704.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6de08de2a03611aee9f915dd03eda646200a3e5400de9c800ceb1aa03d24644e
+size 20740611

checkpoint_p0/milestones/checkpoint_000007552_7733248.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2f1c7992661d8182a4a6a4b125709ad3a99542bd717cb9e16bf93e0444e6d47
+size 20741467

checkpoint_p0/milestones/checkpoint_000015232_15597568.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16bb6f2f1b77b0a0971c1350b6d79ebef964cff6e384241c0726d987aa90c632
+size 20741523

checkpoint_p0/milestones/checkpoint_000022912_23461888.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3fb9225f135d6565f33ac60268ca2af73298939dcee469f4bf5ac399721af5a
+size 20741523

checkpoint_p0/milestones/checkpoint_000030592_31326208.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fdb463bb768b98fca3884f7c40ac859fe7aad9fc6b66043a8861d8932ad7d943
+size 20741523

checkpoint_p0/milestones/checkpoint_000038240_39157760.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1776fda7ca0a10115ccb91b3b63df387adfa91349351d56ab0eb8d65dbab937
+size 20741523

checkpoint_p0/milestones/checkpoint_000045920_47022080.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:751f33ad5d505f8feaf9a6827f3d519eba6ed4d32515eaddcbdea83a9294787a
+size 20741523

checkpoint_p0/milestones/checkpoint_000053600_54886400.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6adfe709b49fa91fa84e222860d95d58d6c0ee0c096103bb2405912b00700dd
+size 20741523

checkpoint_p0/milestones/checkpoint_000061280_62750720.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01dfbaf8f7b893be9cfe1c67edaa3d283b20065c843fddce88e261c570c84ee2
+size 20741523

checkpoint_p0/milestones/checkpoint_000068960_70615040.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e366a8cb65a418e23f2f580d6006ae92163bb4428939911b0b71db0969adc50
+size 20741523

checkpoint_p0/milestones/checkpoint_000076608_78446592.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8162b7fa1ee4082c6c42a6567dcf0efaf19cdfa109cf37d4c9d0a0a09968824
+size 20741523

checkpoint_p0/milestones/checkpoint_000084288_86310912.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a56db1e48663f2508603527eab935706f7512b46a3cf12089cd78bf734c55b6
+size 20741523

checkpoint_p0/milestones/checkpoint_000091968_94175232.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c2f48035ac787fcc7f706600507728be852aa6c9f8a323511d8c4b0eaf96cbb
+size 20741523

checkpoint_p1/best_000086816_88899584_reward_1855.350.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:726c1b457bc52497b3316022e5d05056614d6b83cbe0b97ba8f66ad3c9770470
+size 20740275

checkpoint_p1/checkpoint_000097632_99975168.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd28b65a69892d6fd1100a4f61f9b2a6d5c396d38cd50f6304897090baf7811b
+size 20740611

checkpoint_p1/checkpoint_000097984_100335616.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a812e0036046aaaccbb54e1c7d6ac4bd1b09f6253e1ad7d37c07938c9ece06a3
+size 20740611

checkpoint_p1/milestones/checkpoint_000007584_7766016.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64dd5be719bc672a711877ea255cb7dfef66d7b8f41de5a97e68355adba3b43e
+size 20741467

checkpoint_p1/milestones/checkpoint_000015296_15663104.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8ec6a96b5190597238482f6713dc9bc338902f5356a2937d3fbd2000327707d
+size 20741523

checkpoint_p1/milestones/checkpoint_000022976_23527424.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff6bbb55c867208bec0f7fe71880fd27ae3903e61be0e1a676ee7e97ef2ab695
+size 20741523

checkpoint_p1/milestones/checkpoint_000030688_31424512.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b46630704429ff9cb2bbb2d4e29cbbcd6583d7f622a8de9e30892a2297a71da2
+size 20741523

checkpoint_p1/milestones/checkpoint_000038368_39288832.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:638c6780bda92de83a3d7851ef6bacabcff9c12451f33bc415b1ea99c06a59ee
+size 20741523

checkpoint_p1/milestones/checkpoint_000046080_47185920.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17957e008973a395a125885559e877c1dd877ff483ddb82c7dbbfd8e596fea95
+size 20741523

checkpoint_p1/milestones/checkpoint_000053792_55083008.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0e185da6de9031e407b559ba36c5cb65c56b97c5290591a66679850c1947ec2
+size 20741523

checkpoint_p1/milestones/checkpoint_000061472_62947328.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9f5060990c76caab366b26f86ab3f65c85d3e02712f0f3c1d66c1005be8d273a
+size 20741523

checkpoint_p1/milestones/checkpoint_000069152_70811648.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ad0a1c81b9611b679c599101bd8edf80806b2c6db5d4385fcaffc210849b070
+size 20741523

checkpoint_p1/milestones/checkpoint_000076864_78708736.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2f90cc366aa7e6040ca02c7f917755413ef687131327d112a254f3a25256404
+size 20741523

checkpoint_p1/milestones/checkpoint_000084544_86573056.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93b0bcbe0c918399a94c23c592313ab8abc38a00e5b7b7df4c9db1b456c1a791
+size 20741523

checkpoint_p1/milestones/checkpoint_000092256_94470144.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3332a78f98b197160881783ce421a05a1d896b77070c400b825584ec1f87ea32
+size 20741523

config.json

@@ -2,23 +2,23 @@
   "help": false,
   "algo": "APPO",
   "env": "atari_enduro",
-  "experiment": "atari_enduro",
+  "experiment": "atari_enduro_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_enduro",
   "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_enduro --experiment=atari_enduro --num_policies=2 --train_dir=./train_atari --with_wandb=true --wandb_user=matt-stammers --wandb_project=atari --wandb_group=atari_enduro --wandb_job_type=SF --wandb_tags=atari",
+  "command_line": "--algo=APPO --env=atari_enduro --experiment=atari_enduro_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_enduro --wandb_job_type=SF --wandb_tags=atari",
   "cli_args": {
     "algo": "APPO",
     "env": "atari_enduro",
-    "experiment": "atari_enduro",
+    "experiment": "atari_enduro_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_enduro",
     "wandb_job_type": "SF",
     "wandb_tags": [
@@ -140,5 +158,5 @@
   },
   "git_hash": "5fff97c2f535da5987d358cdbe6927cccd43621e",
   "git_repo_name": "not a git repository",
-  "wandb_unique_id": "atari_enduro_20230926_044835_177035"
+  "wandb_unique_id": "atari_enduro_APPO_20231011_011028_415653"
 }

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 sha256:6ec1e82dbf54d9118b7e2993a8f9fca652894c96719faf7163721ae3a6f3a609
-size 24674022
+oid sha256:bb5d78255e78273452ab1f2994ba7e98bf3397bde2d8cc8c44c5b591c2b7e6d1
+size 88728837