Hugging Face's logo

AhmedTarek
/
dqn-SpaceInvadersNoFrameskip-v4

Reinforcement Learning
stable-baselines3
SpaceInvadersNoFrameskip-v4
deep-reinforcement-learning
Eval Results
Model card Files Community
AhmedTarek commited on
Commit
6a0956c
•
1 Parent(s): c488831

Initial commit

Browse files
Files changed (11) hide show
  1. README.md +4 -4
  2. args.yml +1 -1
  3. config.yml +3 -3
  4. dqn-SpaceInvadersNoFrameskip-v4.zip +2 -2
  5. dqn-SpaceInvadersNoFrameskip-v4/data +0 -0
  6. dqn-SpaceInvadersNoFrameskip-v4/policy.optimizer.pth +2 -2
  7. dqn-SpaceInvadersNoFrameskip-v4/policy.pth +2 -2
  8. dqn-SpaceInvadersNoFrameskip-v4/system_info.txt +1 -1
  9. replay.mp4 +2 -2
  10. results.json +1 -1
  11. 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: 5.00 +/- 7.07
20
  name: mean_reward
21
  verified: false
22
  ---
@@ -62,7 +62,7 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
62
 
63
  ## Hyperparameters
64
  ```python
65
- OrderedDict([('batch_size', 32),
66
  ('buffer_size', 100000),
67
  ('env_wrapper',
68
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
@@ -70,9 +70,9 @@ OrderedDict([('batch_size', 32),
70
  ('exploration_fraction', 0.1),
71
  ('frame_stack', 4),
72
  ('gradient_steps', 1),
73
- ('learning_rate', 0.3),
74
  ('learning_starts', 100000),
75
- ('n_timesteps', 100000),
76
  ('optimize_memory_usage', False),
77
  ('policy', 'CnnPolicy'),
78
  ('target_update_interval', 1000),
 
16
  type: SpaceInvadersNoFrameskip-v4
17
  metrics:
18
  - type: mean_reward
19
+ value: 329.00 +/- 157.97
20
  name: mean_reward
21
  verified: false
22
  ---
 
62
 
63
  ## Hyperparameters
64
  ```python
65
+ OrderedDict([('batch_size', 64),
66
  ('buffer_size', 100000),
67
  ('env_wrapper',
68
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
 
70
  ('exploration_fraction', 0.1),
71
  ('frame_stack', 4),
72
  ('gradient_steps', 1),
73
+ ('learning_rate', 0.003),
74
  ('learning_starts', 100000),
75
+ ('n_timesteps', 200000),
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
- - 853803016
58
  - - storage
59
  - null
60
  - - study_name
 
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
+ - 624506214
58
  - - storage
59
  - null
60
  - - study_name
config.yml CHANGED
@@ -1,6 +1,6 @@
1
  !!python/object/apply:collections.OrderedDict
2
  - - - batch_size
3
- - 32
4
  - - buffer_size
5
  - 100000
6
  - - env_wrapper
@@ -14,11 +14,11 @@
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
- - 0.3
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
- - 100000
22
  - - optimize_memory_usage
23
  - false
24
  - - policy
 
1
  !!python/object/apply:collections.OrderedDict
2
  - - - batch_size
3
+ - 64
4
  - - buffer_size
5
  - 100000
6
  - - env_wrapper
 
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
+ - 0.003
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
+ - 200000
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:db6043915d54b7c57464ffb9d7bee9bcf5b02e335e1115cbfa4fc4ae9e3a3466
3
- size 13715663
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:988c4dd4e111ca597ee2a35ba9ade437af56af4b65353c1c757ef49fae80242d
3
+ size 27220209
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 sha256:601f416f02faa813304477a1ddc15a398d3bea196aa269345846edffe4259c81
3
- size 1120
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aaa5f95af3967656a1ea39ab180e351c18522b9f3f14d9eba492af8e28e32eae
3
+ size 13505852
dqn-SpaceInvadersNoFrameskip-v4/policy.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9d2803d79532a54a441f7d840bb16b71d2cc9793a97eb5174b4849710e3ec2ba
3
- size 13505370
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3bd8415fa75ed4ca1ae0fa36ca46dffa231e3a4a366d95f5fc7165a72021722a
3
+ size 13505178
dqn-SpaceInvadersNoFrameskip-v4/system_info.txt CHANGED
@@ -2,7 +2,7 @@
2
  - Python: 3.10.12
3
  - Stable-Baselines3: 2.3.2
4
  - PyTorch: 2.2.1+cu121
5
- - GPU Enabled: True
6
  - Numpy: 1.25.2
7
  - Cloudpickle: 2.2.1
8
  - Gymnasium: 0.29.1
 
2
  - Python: 3.10.12
3
  - Stable-Baselines3: 2.3.2
4
  - PyTorch: 2.2.1+cu121
5
+ - GPU Enabled: False
6
  - Numpy: 1.25.2
7
  - Cloudpickle: 2.2.1
8
  - Gymnasium: 0.29.1
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3db13127af1afff79a46340c2dcb1019924e473abb0fcb413073b8546932fd7c
3
- size 294075
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1220fd2663a829192a2b709c767eb2debf32246e6f8d3b5838a5840975ff8d0
3
+ size 242623
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 5.0, "std_reward": 7.0710678118654755, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2024-04-30T13:16:13.325933"}
 
1
+ {"mean_reward": 329.0, "std_reward": 157.96835126062436, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2024-05-07T09:27:19.470130"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9968424e683f3ab350336e194b34b1b9736fd336f9332bd512cc2a6b783b7259
3
- size 5940
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:982faaca9889cea78ee7db9b6fe1ca75fc03307bccc39900cafc196670495e61
3
+ size 9716