Upload folder using huggingface_hub

.summary/0/events.out.tfevents.1697167753.rhmmedcatt-proliant-ml350-gen10

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c9566c4d95cde3fbb3b20bc7f6e6b264a959a3e94b9aeb94f1a1ed7335cfa02
+size 7920903

.summary/1/events.out.tfevents.1697167753.rhmmedcatt-proliant-ml350-gen10

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34387d77b1f713cb441a32b1dd092fda8c8252c3a3294d379fa5e9a49f854777
+size 4577307

README.md

@@ -15,7 +15,7 @@ model-index:
       type: atari_montezuma
     metrics:
     - type: mean_reward
-      value: 0.00 +/- 0.00
+      value: 260.00 +/- 174.36
       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_montezuma
+python -m sample_factory.huggingface.load_from_hub -r MattStammers/APPO-atari_montezuma
 ```
 
     
+## 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_montezuma --train_dir=./train_dir --experiment=appo-atari_montezuma
+python -m sf_examples.atari.enjoy_atari --algo=APPO --env=atari_montezuma --train_dir=./train_dir --experiment=APPO-atari_montezuma
 ```
 
 
@@ -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_montezuma --train_dir=./train_dir --experiment=appo-atari_montezuma --restart_behavior=resume --train_for_env_steps=10000000000
+python -m sf_examples.atari.train_atari --algo=APPO --env=atari_montezuma --train_dir=./train_dir --experiment=APPO-atari_montezuma --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_000059024_61472768_reward_0.580.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a44b78f1edab4db8152f8c04fb02b4c3ed7240e32ec451b0ba622e6816cb890
+size 20795763

checkpoint_p0/checkpoint_000096560_99909632.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e9889d6b3026e233fb400e07f3ab51b3d638d2d956d2cce83a6cfddf227c296
+size 20796099

checkpoint_p0/checkpoint_000096656_100007936.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc5cc3c0dd33a6693bd057cea5e7d5d9796f6ab4caa7d72ab0c7e4857d71cd01
+size 20796099

checkpoint_p0/milestones/checkpoint_000007904_8093696.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:183b3ac11673a991a9dc25b8cfdce85da1315f711d560b0721c56fa0024494f4
+size 20796955

checkpoint_p0/milestones/checkpoint_000015968_16351232.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bca6ac45112433031a60706d47082fd198ce11b612f94d1295b7bcf76f12c032
+size 20797011

checkpoint_p0/milestones/checkpoint_000023848_24772608.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16529c92598297ec38758d32ad3eb3b79765ffe786e4a16d1a06703a7814d7a3
+size 20797011

checkpoint_p0/milestones/checkpoint_000031408_33193984.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:766760020dbf2f994e59bbc390a59eb8e3e05fe32e6dff5b9803e6d1058912fb
+size 20797011

checkpoint_p0/milestones/checkpoint_000039440_41418752.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30705d8a65c360cb7703a4ac7585a726c1405497bc643137b14b1864efadeb23
+size 20797011

checkpoint_p0/milestones/checkpoint_000047536_49709056.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f258239d58f88bcee3df34dc06dac9cdc2301364c412ca60d8bf5a5e4a1af803
+size 20797011

checkpoint_p0/milestones/checkpoint_000055600_57966592.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3d9046e1f545a9bef226aeeef6cbeedc62564952419ca3575f88e209ac604b5
+size 20797011

checkpoint_p0/milestones/checkpoint_000063664_66224128.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:762d335be95a706e51a18d3240e994463c47adad1d86027cee0d297e96e67fdd
+size 20797011

checkpoint_p0/milestones/checkpoint_000071696_74448896.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdebed22d7ea5cebdd44e533e37388de694f7d3406ec17ef79f7cc5a306ebbf6
+size 20797011

checkpoint_p0/milestones/checkpoint_000079696_82640896.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:601b6f128220af5c48b1eab696ef658e8a56e7d938d834b87b0450bf9c1ff4b0
+size 20797011

checkpoint_p0/milestones/checkpoint_000087728_90865664.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c669335e381f784229ebe6292f7bafe589a998b6636dd8248474cfdf94922d76
+size 20797011

checkpoint_p0/milestones/checkpoint_000095760_99090432.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7319aac0195c6ed5a41ad6231f02c5088adb20c94244eb361063d8939613e4a
+size 20797011

checkpoint_p1/best_000075248_77299712_reward_0.770.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e5bc40be372b6548079777cb059aa423193a4abea71bf7847ad35bf6e1c9cf8
+size 20795763

checkpoint_p1/checkpoint_000097376_100368384.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:644c1c3117f08bde1a2dc515c92c044c882627a02999fd91aaa18a5412f1919a
+size 20796099

checkpoint_p1/checkpoint_000097472_100466688.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19fa547c71aa1e64e9c5591b3003d131c179faa86e793250b2a89604a6274351
+size 20796099

checkpoint_p1/milestones/checkpoint_000007968_8159232.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97dd7345730d79ca02e94211ceda732f3f49e856e1c0841457bd548f81beadf5
+size 20796955

checkpoint_p1/milestones/checkpoint_000015936_16318464.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1babdf2373a31b95f892c7c059cb065e610013e31f252ff0ce8aabb040df97bf
+size 20797011

checkpoint_p1/milestones/checkpoint_000023936_24510464.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44fac38abf0c295b2fdc5fe9885923487a63ddb01a9ca7471a10fe808e27d76a
+size 20797011

checkpoint_p1/milestones/checkpoint_000032000_32768000.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c204fbee2aab460a7b538b9cbee9d3f9dec7db94efb1780ac0deb3ef709965ab
+size 20797011

checkpoint_p1/milestones/checkpoint_000040032_40992768.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95dd44ee8737103491289e4c34f0ee592eefb7107d8808181e3c94ea2795066d
+size 20797011

checkpoint_p1/milestones/checkpoint_000047840_48988160.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2642f74c721850e0cf8a04ae39fa009ad1ffba22c4e17e814a7f77945f47ef12
+size 20797011

checkpoint_p1/milestones/checkpoint_000055632_57212928.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5267ccf9089766ae06b7e220630d11128abd4262368b7c2af652f44a4b1d6f3
+size 20797011

checkpoint_p1/milestones/checkpoint_000063792_65568768.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:692b91b28b14e4168febf9e771cc878553a8e2ef676ef9ca0c1a6e295b78bfb5
+size 20797011

checkpoint_p1/milestones/checkpoint_000072016_73990144.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d5b0f958d251ff8a2a0f48088959e85716b415597cf01fd43e608238e305599
+size 20797011

checkpoint_p1/milestones/checkpoint_000080000_82575360.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fd45cb9635f1c5fd4ce6d83c22ec8e040f3446114e1072d8ff308dfc93776ea
+size 20797011

checkpoint_p1/milestones/checkpoint_000088352_91127808.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:452582b0f256d401e180e714332b1ed0d98c37d1c0c4230a2139613d139681ed
+size 20797011

checkpoint_p1/milestones/checkpoint_000096576_99549184.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89b15356ad088282d710c4a99ab45637fa799a1a5c9f8bbae112749b271f007c
+size 20797011

config.json

@@ -2,23 +2,23 @@
   "help": false,
   "algo": "APPO",
   "env": "atari_montezuma",
-  "experiment": "atari_montezuma",
+  "experiment": "atari_montezuma_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_montezuma",
   "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_montezuma --experiment=atari_montezuma --num_policies=2 --train_dir=./train_atari --with_wandb=true --wandb_user=matt-stammers --wandb_project=atari --wandb_group=atari_montezuma --wandb_job_type=SF --wandb_tags=atari",
+  "command_line": "--algo=APPO --env=atari_montezuma --experiment=atari_montezuma_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_montezuma --wandb_job_type=SF --wandb_tags=atari",
   "cli_args": {
     "algo": "APPO",
     "env": "atari_montezuma",
-    "experiment": "atari_montezuma",
+    "experiment": "atari_montezuma_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_montezuma",
     "wandb_job_type": "SF",
     "wandb_tags": [
@@ -140,5 +158,5 @@
   },
   "git_hash": "5fff97c2f535da5987d358cdbe6927cccd43621e",
   "git_repo_name": "not a git repository",
-  "wandb_unique_id": "atari_montezuma_20230926_161401_289480"
+  "wandb_unique_id": "atari_montezuma_APPO_20231013_042910_885856"
 }

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 sha256:094639578a14b890cca0e16a74d962c3662994895a372d47c792a0ac0e29f408
-size 2398270
+oid sha256:b708d4b5470b0cbbbaf0510c192707597646ec4eb2bebfc3fbe9a9c65e77c441
+size 15658526