vwxyzjn commited on
Commit
86408bc
1 Parent(s): 4702cd6

pushing model

Browse files
README.md ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - CartPole-v1
4
+ - deep-reinforcement-learning
5
+ - reinforcement-learning
6
+ - custom-implementation
7
+ ---
8
+
9
+ # (CleanRL) **PPO** Agent Playing **CartPole-v1**
10
+
11
+ This is a trained model of a PPO agent playing CartPole-v1.
12
+ The model was trained by using [CleanRL](https://github.com/vwxyzjn/cleanrl) and the training code can be
13
+ found [here](https://github.com/vwxyzjn/cleanrl/blob/master/cleanrl/ppo.py).
14
+
15
+
16
+ # Hyperparameters
17
+ ```python
18
+ {'anneal_lr': True,
19
+ 'batch_size': 512,
20
+ 'capture_video': True,
21
+ 'clip_coef': 0.2,
22
+ 'clip_vloss': True,
23
+ 'cuda': False,
24
+ 'ent_coef': 0.01,
25
+ 'env_id': 'CartPole-v1',
26
+ 'exp_name': 'ppo',
27
+ 'gae_lambda': 0.95,
28
+ 'gamma': 0.99,
29
+ 'hf_repo_id': 'vwxyzjn/ppo2',
30
+ 'learning_rate': 0.00025,
31
+ 'max_grad_norm': 0.5,
32
+ 'minibatch_size': 128,
33
+ 'norm_adv': True,
34
+ 'num_envs': 4,
35
+ 'num_minibatches': 4,
36
+ 'num_steps': 128,
37
+ 'save_model': True,
38
+ 'seed': 1,
39
+ 'target_kl': None,
40
+ 'torch_deterministic': True,
41
+ 'total_timesteps': 500000,
42
+ 'track': False,
43
+ 'update_epochs': 4,
44
+ 'vf_coef': 0.5,
45
+ 'wandb_entity': None,
46
+ 'wandb_project_name': 'cleanRL'}
47
+ ```
48
+
agent.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b78bfad86d1fa939203501f3b6e31a70f8db7a7a33bd6f86c98aeb98c420676a
3
+ size 40129
events.out.tfevents.1665691845.pop-os.2085552.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:367eb007bb2963dca13b51b8d815c361a445b6ac51381c0433314d8d676a9d95
3
+ size 3486