Hugging Face's logo

ashkid
/
dqn-SpaceInvadersNoFrameskip-v4

Reinforcement Learning
stable-baselines3
SpaceInvadersNoFrameskip-v4
deep-reinforcement-learning
Eval Results (legacy)
Model card Files
xet
 Community
ashkid commited on
Commit
c0affe9
·
verified ·
1 Parent(s): 6d3917e

Initial commit

Browse files
Files changed (10) hide show
  1. README.md +3 -3
  2. args.yml +2 -2
  3. config.yml +2 -2
  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. 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: 159.50 +/- 131.44
20
  name: mean_reward
21
  verified: false
22
  ---
@@ -63,7 +63,7 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
63
 
64
  ## Hyperparameters
65
  ```python
66
- OrderedDict([('batch_size', 512),
67
  ('buffer_size', 1000000),
68
  ('env_wrapper',
69
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
@@ -71,7 +71,7 @@ OrderedDict([('batch_size', 512),
71
  ('exploration_fraction', 0.1),
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: 516.00 +/- 168.59
20
  name: mean_reward
21
  verified: false
22
  ---
 
63
 
64
  ## Hyperparameters
65
  ```python
66
+ OrderedDict([('batch_size', 1024),
67
  ('buffer_size', 1000000),
68
  ('env_wrapper',
69
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
 
71
  ('exploration_fraction', 0.1),
72
  ('frame_stack', 4),
73
  ('gradient_steps', 1),
74
+ ('learning_rate', 0.0001),
75
  ('learning_starts', 100000),
76
  ('n_timesteps', 1000000.0),
77
  ('optimize_memory_usage', False),
args.yml CHANGED
@@ -46,7 +46,7 @@
46
  - - optimize_hyperparameters
47
  - false
48
  - - progress
49
- - false
50
  - - pruner
51
  - median
52
  - - sampler
@@ -56,7 +56,7 @@
56
  - - save_replay_buffer
57
  - false
58
  - - seed
59
- - 2969151151
60
  - - storage
61
  - null
62
  - - study_name
 
46
  - - optimize_hyperparameters
47
  - false
48
  - - progress
49
+ - true
50
  - - pruner
51
  - median
52
  - - sampler
 
56
  - - save_replay_buffer
57
  - false
58
  - - seed
59
+ - 634099584
60
  - - storage
61
  - null
62
  - - study_name
config.yml CHANGED
@@ -1,6 +1,6 @@
1
  !!python/object/apply:collections.OrderedDict
2
  - - - batch_size
3
- - 512
4
  - - buffer_size
5
  - 1000000
6
  - - env_wrapper
@@ -14,7 +14,7 @@
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
- - 0.0003
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
 
1
  !!python/object/apply:collections.OrderedDict
2
  - - - batch_size
3
+ - 1024
4
  - - buffer_size
5
  - 1000000
6
  - - env_wrapper
 
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:e7e5004a7ad6b0045f7ec5d5fa04aac520d975841a987b3321840680a3adc207
3
- size 27220522
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d2a3cfaacf568efccd0e3fb7c638312f19848beb66d157b2bc11880f1f86236
3
+ size 27221034
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:a0c14524923fecd5544c2bbaa61e8514ce9b75ab2d42536a6604028f8305ae4c
3
- size 13505852
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:813386f26ac13747d343b8a49a99b3f2ff154c74e743cbd42b8020b82d893340
3
+ size 13506172
dqn-SpaceInvadersNoFrameskip-v4/policy.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b2fc41c607aff5e7599ff9807ffb7cdb814c9eacfcc6cc4e0ae4978adddfad12
3
- size 13505178
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66d72118acea1ef718f53a1d38bc780c45e6881db93f768562d314252e305887
3
+ size 13505370
dqn-SpaceInvadersNoFrameskip-v4/system_info.txt CHANGED
@@ -2,7 +2,7 @@
2
  - Python: 3.12.3
3
  - Stable-Baselines3: 2.5.0
4
  - PyTorch: 2.3.0+cu121
5
- - GPU Enabled: False
6
  - Numpy: 1.26.4
7
  - Cloudpickle: 3.1.1
8
  - Gymnasium: 1.0.0
 
2
  - Python: 3.12.3
3
  - Stable-Baselines3: 2.5.0
4
  - PyTorch: 2.3.0+cu121
5
+ - GPU Enabled: True
6
  - Numpy: 1.26.4
7
  - Cloudpickle: 3.1.1
8
  - Gymnasium: 1.0.0
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 159.5, "std_reward": 131.44295340565046, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2025-02-17T12:30:17.019987"}
 
1
+ {"mean_reward": 516.0, "std_reward": 168.59418732566078, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2025-02-18T09:06:19.832823"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:31c8e97f9eb7302df6b40e6196d8c46e0adab236d7243e51be32b94180e0f329
3
- size 38926
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:40acfa5007220678e4615dece548d98e6050a3d16660db02751724a729425547
3
+ size 35915