Hugging Face's logo

joosma
/
ppo-v3

Reinforcement Learning
TensorBoard
LunarLander-v2
ppo
deep-reinforcement-learning
custom-implementation
deep-rl-course
Eval Results
Model card Files Metrics Community
joosma commited on
Commit
23be8a7
•
1 Parent(s): 54fdcdd

Push agent to the Hub

Browse files
Files changed (5) hide show
  1. README.md +6 -6
  2. logs/events.out.tfevents.1716287720.d0dfc23ae7b3.9391.0 +3 -0
  3. model.pt +1 -1
  4. replay.mp4 +0 -0
  5. results.json +1 -1
README.md CHANGED
@@ -17,7 +17,7 @@ model-index:
17
  type: LunarLander-v2
18
  metrics:
19
  - type: mean_reward
20
- value: -189.02 +/- 94.90
21
  name: mean_reward
22
  verified: false
23
  ---
@@ -37,9 +37,9 @@ model-index:
37
  'wandb_entity': None
38
  'capture_video': False
39
  'env_id': 'LunarLander-v2'
40
- 'total_timesteps': 100000
41
- 'learning_rate': 0.00025
42
- 'num_envs': 2048
43
  'num_steps': 2048
44
  'anneal_lr': True
45
  'gae': True
@@ -55,7 +55,7 @@ model-index:
55
  'max_grad_norm': 0.5
56
  'target_kl': None
57
  'repo_id': 'joosma/ppo-v3'
58
- 'batch_size': 4194304
59
- 'minibatch_size': 419430}
60
  ```
61
 
 
17
  type: LunarLander-v2
18
  metrics:
19
  - type: mean_reward
20
+ value: -151.06 +/- 77.67
21
  name: mean_reward
22
  verified: false
23
  ---
 
37
  'wandb_entity': None
38
  'capture_video': False
39
  'env_id': 'LunarLander-v2'
40
+ 'total_timesteps': 1000000
41
+ 'learning_rate': 0.0002
42
+ 'num_envs': 20
43
  'num_steps': 2048
44
  'anneal_lr': True
45
  'gae': True
 
55
  'max_grad_norm': 0.5
56
  'target_kl': None
57
  'repo_id': 'joosma/ppo-v3'
58
+ 'batch_size': 40960
59
+ 'minibatch_size': 4096}
60
  ```
61
 
logs/events.out.tfevents.1716287720.d0dfc23ae7b3.9391.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0bd1907b67cdb569f7319c02e7bd8a42336495e2505d3b94a4363841312faa7
3
+ size 970556
model.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ea4629b9510ade23c006536ad792ecfb1b63f646f7a1aca7e04da8ed725ada69
3
  size 43026
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45f2ef9a8faa2ad6a055de06f9e35808170a3ecd032b7e4d2feef3736a03452a
3
  size 43026
replay.mp4 CHANGED
Binary files a/replay.mp4 and b/replay.mp4 differ
 
results.json CHANGED
@@ -1 +1 @@
1
- {"env_id": "LunarLander-v2", "mean_reward": -189.0192891408004, "std_reward": 94.90027590606195, "n_evaluation_episodes": 10, "eval_datetime": "2024-05-21T10:32:01.843519"}
 
1
+ {"env_id": "LunarLander-v2", "mean_reward": -151.06307082532751, "std_reward": 77.67048090574565, "n_evaluation_episodes": 10, "eval_datetime": "2024-05-21T10:40:37.266777"}