Upload folder using huggingface_hub

Files changed (10) hide show

.summary/0/events.out.tfevents.1756487512.Pac ADDED Viewed

+version https://git-lfs.github.com/spec/v1
+oid sha256:1ce2a1118f92d8787511b29bc3293bc3923c6f3be26b19fd9a64573a83ddebbc
+size 40

.summary/0/events.out.tfevents.1756487556.Pac ADDED Viewed

+version https://git-lfs.github.com/spec/v1
+oid sha256:01cc3a02253c72350b7c23d9c28f8b72293f5c943d39054cd349ba00b4c19fdb
+size 40

.summary/0/events.out.tfevents.1756487755.Pac ADDED Viewed

+version https://git-lfs.github.com/spec/v1
+oid sha256:5e323a977f06005638c5a746ccc1adb91ff2756c339d7d9434aaa2786ec31607
+size 119794

.summary/0/events.out.tfevents.1756489979.Pac ADDED Viewed

File without changes

.summary/0/events.out.tfevents.1756490729.Pac ADDED Viewed

+version https://git-lfs.github.com/spec/v1
+oid sha256:4711b8e85b5053f2d3ec95793131588ea0b6081ca0070afd89df574edc2a37ce
+size 1465481

README.md CHANGED Viewed

@@ -15,7 +15,7 @@ model-index:
       type: doom_health_gathering_supreme
     metrics:
     - type: mean_reward
-      value: 4.06 +/- 0.29
       name: mean_reward
       verified: false
 ---

       type: doom_health_gathering_supreme
     metrics:
     - type: mean_reward
+      value: 4.08 +/- 0.63
       name: mean_reward
       verified: false
 ---

config.json CHANGED Viewed

@@ -16,8 +16,8 @@
   "policy_workers_per_policy": 1,
   "max_policy_lag": 1000,
   "num_workers": 10,
-  "num_envs_per_worker": 4,
-  "batch_size": 1024,
   "num_batches_per_epoch": 1,
   "num_epochs": 1,
   "rollout": 64,
@@ -33,7 +33,7 @@
   "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,
@@ -43,7 +43,7 @@
   "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,
@@ -65,7 +65,7 @@
   "summaries_use_frameskip": true,
   "heartbeat_interval": 20,
   "heartbeat_reporting_interval": 600,
-  "train_for_env_steps": 20000000,
   "train_for_seconds": 10000000000,
   "save_every_sec": 120,
   "keep_checkpoints": 2,

   "policy_workers_per_policy": 1,
   "max_policy_lag": 1000,
   "num_workers": 10,
+  "num_envs_per_worker": 8,
+  "batch_size": 16384,
   "num_batches_per_epoch": 1,
   "num_epochs": 1,
   "rollout": 64,
   "kl_loss_coeff": 0.0,
   "exploration_loss": "symmetric_kl",
   "gae_lambda": 0.95,
+  "ppo_clip_ratio": 0.2,
   "ppo_clip_value": 0.2,
   "with_vtrace": false,
   "vtrace_rho": 1.0,
   "adam_beta1": 0.9,
   "adam_beta2": 0.999,
   "max_grad_norm": 4.0,
+  "learning_rate": 0.0002,
   "lr_schedule": "constant",
   "lr_schedule_kl_threshold": 0.008,
   "lr_adaptive_min": 1e-06,
   "summaries_use_frameskip": true,
   "heartbeat_interval": 20,
   "heartbeat_reporting_interval": 600,
+  "train_for_env_steps": 30000000,
   "train_for_seconds": 10000000000,
   "save_every_sec": 120,
   "keep_checkpoints": 2,

git.diff ADDED Viewed

The diff for this file is too large to render. See raw diff

replay.mp4 CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d68220307c311d960a42551365fa1cf0216627cfc8ff7b2308e884c156a3ce8
-size 6191141

 version https://git-lfs.github.com/spec/v1
+oid sha256:326b2aa7398ea99309b188513a36416adb9ad337a3a44b1bee2642f7bab21038
+size 5657570

sf_log.txt CHANGED Viewed

The diff for this file is too large to render. See raw diff