Push agent to the Hub

README.md

@@ -17,7 +17,7 @@ model-index:
       type: LunarLander-v2
     metrics:
     - type: mean_reward
-      value: -52.45 +/- 45.54
+      value: -174.84 +/- 112.02
       name: mean_reward
       verified: false
 ---
@@ -33,20 +33,20 @@ model-index:
 'env_id': 'LunarLander-v2'
 'learning_rate': 0.00025
 'seed': 1
-'total_timesteps': 3000000
+'total_timesteps': 300000
 'torch_deterministic': True
 'cuda': True
 'track': False
 'wandb_project_name': 'ppo-implementation-details'
 'wandb_entity': None
 'capture_video': False
-'num_envs': 1024
-'num_steps': 512
+'num_envs': 2048
+'num_steps': 1024
 'anneal_lr': True
 'gae': True
 'gamma': 0.99
 'gae_lambda': 0.95
-'num_minibatches': 64
+'num_minibatches': 256
 'update_epochs': 4
 'norm_adv': True
 'clip_coef': 0.2
@@ -56,7 +56,7 @@ model-index:
 'max_grad_norm': 0.5
 'target_kl': None
 'repo_id': 'SuperDan/ppo-lunar-sharedMLP'
-'batch_size': 524288
+'batch_size': 2097152
 'minibatch_size': 8192}
   ```
   

logs/events.out.tfevents.1702175722.79641db95cbe.31743.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c69191c1f834d2c4d2e30b2f92c8acdb634bf4bcc4060c8c137538705299de69
+size 787

model.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a91ec2b7f2bb9ff785b5b25f6fcfdb70cfbe7afd80eecfed0e7b10a62af9622e
+oid sha256:88bcc5467c57ead0a38a64f0e057789b675891fc492cc2e68c6ecef2ffda2ce3
 size 1082216

results.json

@@ -1 +1 @@
-{"env_id": "LunarLander-v2", "mean_reward": -52.44801921893865, "std_reward": 45.53821134379092, "n_evaluation_episodes": 10, "eval_datetime": "2023-12-10T02:25:26.041794"}
+{"env_id": "LunarLander-v2", "mean_reward": -174.83829995558182, "std_reward": 112.01766212394249, "n_evaluation_episodes": 10, "eval_datetime": "2023-12-10T02:35:29.158855"}