Hugging Face's logo

SolomonBlue
/
dqn-SpaceInvadersNoFrameskip-v4

Reinforcement Learning
stable-baselines3
SpaceInvadersNoFrameskip-v4
deep-reinforcement-learning
Eval Results
Model card Files Community
SolomonBlue commited on
Commit
3b052ab
·
verified ·
1 Parent(s): 774c2e6

Initial commit

Browse files
Files changed (9) 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 +1 -1
  7. dqn-SpaceInvadersNoFrameskip-v4/policy.pth +1 -1
  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: 361.00 +/- 162.29
20
  name: mean_reward
21
  verified: false
22
  ---
@@ -64,7 +64,7 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
64
  ## Hyperparameters
65
  ```python
66
  OrderedDict([('batch_size', 128),
67
- ('buffer_size', 300000),
68
  ('env_wrapper',
69
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
70
  ('exploration_final_eps', 0.01),
@@ -72,12 +72,12 @@ OrderedDict([('batch_size', 128),
72
  ('frame_stack', 4),
73
  ('gradient_steps', 1),
74
  ('learning_rate', 0.00025),
75
- ('learning_starts', 50000),
76
  ('n_timesteps', 1000000.0),
77
  ('optimize_memory_usage', False),
78
  ('policy', 'CnnPolicy'),
79
  ('target_update_interval', 1000),
80
- ('train_freq', 4),
81
  ('normalize', False)])
82
  ```
83
 
 
16
  type: SpaceInvadersNoFrameskip-v4
17
  metrics:
18
  - type: mean_reward
19
+ value: 666.00 +/- 267.61
20
  name: mean_reward
21
  verified: false
22
  ---
 
64
  ## Hyperparameters
65
  ```python
66
  OrderedDict([('batch_size', 128),
67
+ ('buffer_size', 500000),
68
  ('env_wrapper',
69
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
70
  ('exploration_final_eps', 0.01),
 
72
  ('frame_stack', 4),
73
  ('gradient_steps', 1),
74
  ('learning_rate', 0.00025),
75
+ ('learning_starts', 100000),
76
  ('n_timesteps', 1000000.0),
77
  ('optimize_memory_usage', False),
78
  ('policy', 'CnnPolicy'),
79
  ('target_update_interval', 1000),
80
+ ('train_freq', 8),
81
  ('normalize', False)])
82
  ```
83
 
args.yml CHANGED
@@ -56,7 +56,7 @@
56
  - - save_replay_buffer
57
  - false
58
  - - seed
59
- - 4132879894
60
  - - storage
61
  - null
62
  - - study_name
 
56
  - - save_replay_buffer
57
  - false
58
  - - seed
59
+ - 3478458756
60
  - - storage
61
  - null
62
  - - study_name
config.yml CHANGED
@@ -2,7 +2,7 @@
2
  - - - batch_size
3
  - 128
4
  - - buffer_size
5
- - 300000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
@@ -16,7 +16,7 @@
16
  - - learning_rate
17
  - 0.00025
18
  - - learning_starts
19
- - 50000
20
  - - n_timesteps
21
  - 1000000.0
22
  - - optimize_memory_usage
@@ -26,4 +26,4 @@
26
  - - target_update_interval
27
  - 1000
28
  - - train_freq
29
- - 4
 
2
  - - - batch_size
3
  - 128
4
  - - buffer_size
5
+ - 500000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
 
16
  - - learning_rate
17
  - 0.00025
18
  - - learning_starts
19
+ - 100000
20
  - - n_timesteps
21
  - 1000000.0
22
  - - optimize_memory_usage
 
26
  - - target_update_interval
27
  - 1000
28
  - - train_freq
29
+ - 8
dqn-SpaceInvadersNoFrameskip-v4.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:86f960bb0651a207527fd8dd558c589238295601434ec7fbb73147ce9ec0537e
3
- size 27221024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:871f15a826c221989f9fffbb32266b7a26f48f901beff9dc309abef44cc7d52d
3
+ size 27221026
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:e9d391aaad9d50fcc792655841b027bf03b4427da37e6d272ba167d6bb70172b
3
  size 13506172
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:28df8eaaa722014d4e4d0f98df3698440ca567cc09ebde4ab31dfcd1ef0df3fa
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:c28b4060bbb35a8739d64510d279933f4b61a75c850bccb136207d7a40bce3e4
3
  size 13505370
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23bead6a79e0ca3b752e95d3ab43f9eb6f3c24ccfbf51477b3d7230417e810d8
3
  size 13505370
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 361.0, "std_reward": 162.29294500994183, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2025-01-18T21:48:19.130994"}
 
1
+ {"mean_reward": 666.0, "std_reward": 267.6079221547823, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2025-01-19T13:52:52.192576"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a1cd5b9c9f6639f9afe88dde5e3c3702cd396a30de0d9756a90e16307847eeb6
3
- size 9428
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c43a9a1e8398f523b54751a8c62c6838e37c255481921cb4bf4d2c106eabe89
3
+ size 29122