Hugging Face's logo

dficenec
/
dqn-SpaceInvadersNoFrameskip-v4

Reinforcement Learning
stable-baselines3
SpaceInvadersNoFrameskip-v4
deep-reinforcement-learning
Eval Results
Model card Files Community
dficenec commited on
Commit
0d8bffe
•
1 Parent(s): 4a51c4a

Initial commit

Browse files
Files changed (10) hide show
  1. README.md +5 -5
  2. args.yml +1 -1
  3. config.yml +4 -4
  4. dqn-SpaceInvadersNoFrameskip-v4.zip +2 -2
  5. dqn-SpaceInvadersNoFrameskip-v4/data +0 -0
  6. dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth +1 -1
  7. dqn-SpaceInvadersNoFrameskip-v4/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: SpaceInvadersNoFrameskip-v4
17
  metrics:
18
  - type: mean_reward
19
- value: 668.50 +/- 301.71
20
  name: mean_reward
21
  verified: false
22
  ---
@@ -57,17 +57,17 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
57
 
58
  ## Hyperparameters
59
  ```python
60
- OrderedDict([('batch_size', 64),
61
- ('buffer_size', 500000),
62
  ('env_wrapper',
63
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
64
  ('exploration_final_eps', 0.01),
65
  ('exploration_fraction', 0.1),
66
  ('frame_stack', 4),
67
  ('gradient_steps', 1),
68
- ('learning_rate', 0.0002),
69
  ('learning_starts', 100000),
70
- ('n_timesteps', 2000000.0),
71
  ('optimize_memory_usage', False),
72
  ('policy', 'CnnPolicy'),
73
  ('target_update_interval', 1000),
 
16
  type: SpaceInvadersNoFrameskip-v4
17
  metrics:
18
  - type: mean_reward
19
+ value: 673.50 +/- 187.70
20
  name: mean_reward
21
  verified: false
22
  ---
 
57
 
58
  ## Hyperparameters
59
  ```python
60
+ OrderedDict([('batch_size', 128),
61
+ ('buffer_size', 800000),
62
  ('env_wrapper',
63
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
64
  ('exploration_final_eps', 0.01),
65
  ('exploration_fraction', 0.1),
66
  ('frame_stack', 4),
67
  ('gradient_steps', 1),
68
+ ('learning_rate', 0.00027),
69
  ('learning_starts', 100000),
70
+ ('n_timesteps', 2500000.0),
71
  ('optimize_memory_usage', False),
72
  ('policy', 'CnnPolicy'),
73
  ('target_update_interval', 1000),
args.yml CHANGED
@@ -54,7 +54,7 @@
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
- - 2340552822
58
  - - storage
59
  - null
60
  - - study_name
 
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
+ - 4250593871
58
  - - storage
59
  - null
60
  - - study_name
config.yml CHANGED
@@ -1,8 +1,8 @@
1
  !!python/object/apply:collections.OrderedDict
2
  - - - batch_size
3
- - 64
4
  - - buffer_size
5
- - 500000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
@@ -14,11 +14,11 @@
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
- - 0.0002
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
- - 2000000.0
22
  - - optimize_memory_usage
23
  - false
24
  - - policy
 
1
  !!python/object/apply:collections.OrderedDict
2
  - - - batch_size
3
+ - 128
4
  - - buffer_size
5
+ - 800000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
 
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
+ - 0.00027
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
+ - 2500000.0
22
  - - optimize_memory_usage
23
  - false
24
  - - policy
dqn-SpaceInvadersNoFrameskip-v4.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:08ac423b28df8db07adc044e2dfcf5ff460691eca1fbea6cd0b01bdf44fbafb1
3
- size 27224788
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac37685a87eb019ae577a82442c59e897c6626583be607b77c1482e88f809181
3
+ size 27224789
dqn-SpaceInvadersNoFrameskip-v4/data CHANGED
The diff for this file is too large to render. See raw diff
 
dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:082f9f4330d6b0106670dbe1ca4b53ae51001185a07fabe83b3171bf1d4eec50
3
  size 13505739
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d34df9b87e43ba89bb25f84cc72330453f65f1323f6c8cad5b9e36fb0d73e278
3
  size 13505739
dqn-SpaceInvadersNoFrameskip-v4/policy.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bbc0e52379a753bdf6d79d23d60e7fc1c445982923fe38008431e6e45a2056ee
3
  size 13504937
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cbd360565bf70b30fca50305577c7e1c9ea90e0b50b277d12975ec0dbaaa3841
3
  size 13504937
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cc0508a5949428bea347ecacbde062b8c2d329c2d82962bf90fe503aca607718
3
- size 165592
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9b31c65770c6ce442476d4cf3401c13d968e56e4a2b80668a59bd305042ae6f
3
+ size 164059
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 668.5, "std_reward": 301.71219730067264, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-20T21:35:56.072311"}
 
1
+ {"mean_reward": 673.5, "std_reward": 187.69722960129167, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-21T01:59:43.276135"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:15d845076f7cd742ae8b42ad5b34a4d59b4da40944885515b0f7aa8357821d64
3
- size 58669
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7571c6938f8f5fb8b630aea997dd5a37767ed3bf1de9193facc248a697f1e44
3
+ size 73982