Hugging Face's logo

alfredowh
/
dqn-Pacman-v5

Reinforcement Learning
stable-baselines3
ALE/Pacman-v5
deep-reinforcement-learning
Eval Results
Model card Files Community
alfredowh commited on
Commit
f9cd33c
·
1 Parent(s): 76e2f7d

2x timesteps

Browse files
Files changed (10) hide show
  1. README.md +2 -2
  2. args.yml +1 -1
  3. config.yml +1 -1
  4. dqn-ALE-Pacman-v5.zip +2 -2
  5. dqn-ALE-Pacman-v5/data +0 -0
  6. dqn-ALE-Pacman-v5/policy.optimizer.pth +1 -1
  7. dqn-ALE-Pacman-v5/policy.pth +1 -1
  8. replay.mp4 +2 -2
  9. results.json +1 -1
  10. train_eval_metrics.zip +2 -2
README.md CHANGED
@@ -16,7 +16,7 @@ model-index:
16
  type: ALE/Pacman-v5
17
  metrics:
18
  - type: mean_reward
19
- value: 133.70 +/- 45.14
20
  name: mean_reward
21
  verified: false
22
  ---
@@ -72,7 +72,7 @@ OrderedDict([('batch_size', 32),
72
  ('gradient_steps', 1),
73
  ('learning_rate', 0.0001),
74
  ('learning_starts', 100000),
75
- ('n_timesteps', 500000.0),
76
  ('optimize_memory_usage', False),
77
  ('policy', 'CnnPolicy'),
78
  ('target_update_interval', 1000),
 
16
  type: ALE/Pacman-v5
17
  metrics:
18
  - type: mean_reward
19
+ value: 182.20 +/- 52.73
20
  name: mean_reward
21
  verified: false
22
  ---
 
72
  ('gradient_steps', 1),
73
  ('learning_rate', 0.0001),
74
  ('learning_starts', 100000),
75
+ ('n_timesteps', 1000000.0),
76
  ('optimize_memory_usage', False),
77
  ('policy', 'CnnPolicy'),
78
  ('target_update_interval', 1000),
args.yml CHANGED
@@ -54,7 +54,7 @@
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
- - 833607956
58
  - - storage
59
  - null
60
  - - study_name
 
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
+ - 360803983
58
  - - storage
59
  - null
60
  - - study_name
config.yml CHANGED
@@ -18,7 +18,7 @@
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
- - 500000.0
22
  - - optimize_memory_usage
23
  - false
24
  - - policy
 
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
+ - 1000000.0
22
  - - optimize_memory_usage
23
  - false
24
  - - policy
dqn-ALE-Pacman-v5.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f9bae19baaf8d68571e3f128535f123722236a9dfc981d76cba02584590729e0
3
- size 27211924
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:919d94ee7fecdd115450012d1863814d972a7bc7a349445b930387de22800a77
3
+ size 27211927
dqn-ALE-Pacman-v5/data CHANGED
The diff for this file is too large to render. See raw diff
 
dqn-ALE-Pacman-v5/policy.optimizer.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c0e818da462da75ff6936084089377bf25945ebd069bf00cc8b8761240839c5b
3
  size 13502140
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef9577b7f8ee973de03f4a44250c5c168797740d2551479b18813b5af7dfe4dc
3
  size 13502140
dqn-ALE-Pacman-v5/policy.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e9d6b39dc4195cb44e59e1b147c7bb8529fe863fe1ee6ed50a185e452011265b
3
  size 13501274
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c2d993574befb58e2ce054ad61ab8a4e468e4151336235dca080bef36f3a555
3
  size 13501274
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:57ffa84a30a42bc465e79f1dbaf4a73a2d449dda66d8452cce6a111893a94dd8
3
- size 215903
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7dcb5386aabc339858060ac3bad75707743d279f14424cf084f692054b8fcdb
3
+ size 212294
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 133.7, "std_reward": 45.14432411721322, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-11-20T20:00:54.983831"}
 
1
+ {"mean_reward": 182.2, "std_reward": 52.73101554114049, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-11-30T12:45:32.601971"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d7d5a771aa8d8bf077ed81d17a56d1c4eefc1f84e62ddf9e22c95299a2dcc81d
3
- size 74233
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a0e6caacfd7218ce0af9ebbdbd9dec9582effc2c892a7e25d31d6c7229c3007
3
+ size 117030