Hugging Face's logo

eLarry
/
dqn-SpaceInvadersNoFrameskip-v4

Reinforcement Learning
stable-baselines3
SpaceInvadersNoFrameskip-v4
deep-reinforcement-learning
Eval Results
Model card Files Community
eLarry commited on
Commit
2438fea
•
1 Parent(s): 2958b85

Initial commit

Browse files
Files changed (9) hide show
  1. README.md +2 -3
  2. args.yml +3 -4
  3. config.yml +1 -3
  4. dqn-SpaceInvadersNoFrameskip-v4.zip +2 -2
  5. dqn-SpaceInvadersNoFrameskip-v4/data +0 -0
  6. dqn-SpaceInvadersNoFrameskip-v4/policy.pth +1 -1
  7. replay.mp4 +2 -2
  8. results.json +1 -1
  9. 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: 269.50 +/- 76.53
20
  name: mean_reward
21
  verified: false
22
  ---
@@ -68,10 +68,9 @@ OrderedDict([('batch_size', 32),
68
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
69
  ('exploration_final_eps', 0.01),
70
  ('exploration_fraction', 0.1),
71
- ('exploration_initial_eps', 0.02),
72
  ('frame_stack', 4),
73
  ('gradient_steps', 1),
74
- ('learning_rate', 0.0003),
75
  ('learning_starts', 100000),
76
  ('n_timesteps', 1000000.0),
77
  ('optimize_memory_usage', False),
 
16
  type: SpaceInvadersNoFrameskip-v4
17
  metrics:
18
  - type: mean_reward
19
+ value: 374.00 +/- 214.89
20
  name: mean_reward
21
  verified: false
22
  ---
 
68
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
69
  ('exploration_final_eps', 0.01),
70
  ('exploration_fraction', 0.1),
 
71
  ('frame_stack', 4),
72
  ('gradient_steps', 1),
73
+ ('learning_rate', 0.0001),
74
  ('learning_starts', 100000),
75
  ('n_timesteps', 1000000.0),
76
  ('optimize_memory_usage', False),
args.yml CHANGED
@@ -16,8 +16,7 @@
16
  - - gym_packages
17
  - []
18
  - - hyperparams
19
- - exploration_initial_eps: 0.02
20
- learning_rate: 0.0003
21
  - - log_folder
22
  - logs/
23
  - - log_interval
@@ -55,7 +54,7 @@
55
  - - save_replay_buffer
56
  - false
57
  - - seed
58
- - 3585225183
59
  - - storage
60
  - null
61
  - - study_name
@@ -65,7 +64,7 @@
65
  - - track
66
  - false
67
  - - trained_agent
68
- - logs/dqn/SpaceInvadersNoFrameskip-v4_6/best_model.zip
69
  - - truncate_last_trajectory
70
  - true
71
  - - uuid
 
16
  - - gym_packages
17
  - []
18
  - - hyperparams
19
+ - null
 
20
  - - log_folder
21
  - logs/
22
  - - log_interval
 
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
+ - 682855770
58
  - - storage
59
  - null
60
  - - study_name
 
64
  - - track
65
  - false
66
  - - trained_agent
67
+ - ''
68
  - - truncate_last_trajectory
69
  - true
70
  - - uuid
config.yml CHANGED
@@ -9,14 +9,12 @@
9
  - 0.01
10
  - - exploration_fraction
11
  - 0.1
12
- - - exploration_initial_eps
13
- - 0.02
14
  - - frame_stack
15
  - 4
16
  - - gradient_steps
17
  - 1
18
  - - learning_rate
19
- - 0.0003
20
  - - learning_starts
21
  - 100000
22
  - - n_timesteps
 
9
  - 0.01
10
  - - exploration_fraction
11
  - 0.1
 
 
12
  - - frame_stack
13
  - 4
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
+ - 0.0001
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
dqn-SpaceInvadersNoFrameskip-v4.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ad3397e8e9424f74db2496d461a179cb4338fe496c02e960dace1c38549d0a78
3
- size 13717563
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de3f8e78936e95fe66aabb761622ff2eb0c546ee022b51b59f882cd084b35f1c
3
+ size 13719681
dqn-SpaceInvadersNoFrameskip-v4/data CHANGED
The diff for this file is too large to render. See raw diff
 
dqn-SpaceInvadersNoFrameskip-v4/policy.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:07e2b50ea8ced85ecdd4c42e7a6bc8064ad83284f36f9cc548e677d33ee2b78a
3
  size 13504745
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58ab51134641614d8419713c9c198b6758c9ce9cc1ebbb60244514b3cd4e393e
3
  size 13504745
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fa8df3f94821154b6815796bb0076d90bcfc8528f464d4601146a5ba9a0fe4b1
3
- size 266589
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:322cf0cd61bf5e0031f1dcdc699552efb08db7bd19365cc56f504f76f4a212eb
3
+ size 234682
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 269.5, "std_reward": 76.53267276137689, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-02-19T22:55:39.813698"}
 
1
+ {"mean_reward": 374.0, "std_reward": 214.89299662855464, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-02-20T03:47:57.282577"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cc0eb1796333c77349e8bd99bf9a19acbb036cf5ccf02263d62566183f3b8a91
3
- size 6363
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eea199f4702bc78439f76abfe3ab53224930438b63c7d7c9223761d773d2084d
3
+ size 6036