Upload folder using huggingface_hub

README.md

@@ -1,10 +1,10 @@
 ---
 tags:
 - Pixelcopter-PLE-v0
-- reinforce
+- actor-critic
 - reinforcement-learning
-- custom-implementation
-- deep-rl-class
+- deep-rl-course
+- unit4
 model-index:
 - name: Reinforce-PixelCopter
   results:
@@ -15,9 +15,9 @@ model-index:
       name: Pixelcopter-PLE-v0
       type: Pixelcopter-PLE-v0
     metrics:
-    - type: mean_reward
-      value: 27.28 +/- 26.56
-      name: mean_reward
+    - type: shaped_return
+      value: 129.24 +/- 95.79
+      name: shaped_return
       verified: false
 ---
 # **REINFORCE** Agent playing **Pixelcopter-PLE-v0**

hyperparameters.json

@@ -1,17 +1,7 @@
 {
-  "env_id": "Pixelcopter-PLE-v0",
-  "h_size": 64,
-  "n_training_episodes": 40000,
-  "max_t": 10000,
+  "h_size": 128,
+  "lr": 0.0003,
   "gamma": 0.99,
-  "lr": 0.0001,
-  "print_every": 500,
-  "obs_scale": 1.0,
-  "clip_obs": null,
-  "use_ema_baseline": true,
-  "baseline_beta": 0.98,
-  "normalize_advantages": true,
-  "use_entropy_bonus": true,
-  "entropy_coef": 0.005,
-  "n_evaluation_episodes": 50
+  "train_max_t": 1000,
+  "n_training_episodes": 20000
 }

model_state_dict.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e22871fd7a25e348514dad16edac8e4c846b58ecb7c99624efb10ce94849660
+size 75339

results.json

@@ -1,9 +1,8 @@
 {
   "env_id": "Pixelcopter-PLE-v0",
-  "mean_reward": 27.28,
-  "std_reward": 26.560903599087137,
-  "mean_episode_length": 194.28,
-  "std_episode_length": 151.0614497481075,
-  "n_evaluation_episodes": 50,
-  "eval_datetime": "2026-02-17T12:21:52.963761"
+  "shaped_mean_return": 129.24,
+  "shaped_std_return": 95.78685922400838,
+  "mean_episode_length": 131.22,
+  "std_episode_length": 95.70962125094842,
+  "eval_datetime": "2026-02-17T12:42:30.534318"
 }