Hugging Face's logo

dbaibak
/
dqn-SpaceInvadersNoFrameskip-v4

Reinforcement Learning
stable-baselines3
SpaceInvadersNoFrameskip-v4
deep-reinforcement-learning
Eval Results
Model card Files Community
dbaibak commited on
Commit
a0ce003
•
1 Parent(s): 43385c5

Initial commit

Browse files
Files changed (10) hide show
  1. README.md +7 -7
  2. args.yml +1 -1
  3. config.yml +6 -6
  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. 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: 506.50 +/- 134.22
20
  name: mean_reward
21
  verified: false
22
  ---
@@ -58,18 +58,18 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
58
  ## Hyperparameters
59
  ```python
60
  OrderedDict([('batch_size', 32),
61
- ('buffer_size', 80000),
62
  ('env_wrapper',
63
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
64
  ('exploration_final_eps', 0.01),
65
- ('exploration_fraction', 0.05),
66
- ('frame_stack', 4),
67
  ('gradient_steps', 1),
68
- ('learning_rate', 9.5e-05),
69
  ('learning_starts', 100000),
70
- ('n_timesteps', 1000000.0),
71
  ('normalize', False),
72
- ('optimize_memory_usage', False),
73
  ('policy', 'CnnPolicy'),
74
  ('target_update_interval', 1000),
75
  ('train_freq', 4)])
 
16
  type: SpaceInvadersNoFrameskip-v4
17
  metrics:
18
  - type: mean_reward
19
+ value: 760.50 +/- 315.33
20
  name: mean_reward
21
  verified: false
22
  ---
 
58
  ## Hyperparameters
59
  ```python
60
  OrderedDict([('batch_size', 32),
61
+ ('buffer_size', 100000),
62
  ('env_wrapper',
63
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
64
  ('exploration_final_eps', 0.01),
65
+ ('exploration_fraction', 0.025),
66
+ ('frame_stack', 3),
67
  ('gradient_steps', 1),
68
+ ('learning_rate', 0.0001),
69
  ('learning_starts', 100000),
70
+ ('n_timesteps', 10000000.0),
71
  ('normalize', False),
72
+ ('optimize_memory_usage', None),
73
  ('policy', 'CnnPolicy'),
74
  ('target_update_interval', 1000),
75
  ('train_freq', 4)])
args.yml CHANGED
@@ -54,7 +54,7 @@
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
- - 3831049508
58
  - - storage
59
  - null
60
  - - study_name
 
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
+ - 3052815607
58
  - - storage
59
  - null
60
  - - study_name
config.yml CHANGED
@@ -2,27 +2,27 @@
2
  - - - batch_size
3
  - 32
4
  - - buffer_size
5
- - 80000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
9
  - 0.01
10
  - - exploration_fraction
11
- - 0.05
12
  - - frame_stack
13
- - 4
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
- - 9.5e-05
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
- - 1000000.0
22
  - - normalize
23
  - false
24
  - - optimize_memory_usage
25
- - false
26
  - - policy
27
  - CnnPolicy
28
  - - target_update_interval
 
2
  - - - batch_size
3
  - 32
4
  - - buffer_size
5
+ - 100000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
9
  - 0.01
10
  - - exploration_fraction
11
+ - 0.025
12
  - - frame_stack
13
+ - 3
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
+ - 0.0001
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
+ - 10000000.0
22
  - - normalize
23
  - false
24
  - - optimize_memory_usage
25
+ - null
26
  - - policy
27
  - CnnPolicy
28
  - - target_update_interval
dqn-SpaceInvadersNoFrameskip-v4.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1434cd3094356b04514ce4058142c1c4279166ee64b8166f3c8250b11f0e8dea
3
- size 27224789
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:156a5d5e6c3bbb61a7ffdee9b6aa2fbcdf7872030ccd37c0ee1b212be9fc2c7c
3
+ size 27144082
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:abf9a116f1e5668936f81100253272c0d541b74604f6a032f88968101aad96bd
3
- size 13505739
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:972c8ff8b9a67f89209595388207f11aa778d3939f92a873b899fc231eef69bb
3
+ size 13489355
dqn-SpaceInvadersNoFrameskip-v4/policy.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1d348c3a5a2b49f217d4a8b68d97244b7a6b981f816ed314eea48fd2060b0e96
3
- size 13504937
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e803fb0909dfde1e6534b7a02d1b0f6add81ce950fafc8746225b0080d73f5b0
3
+ size 13488553
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a57908d30cc0c53d4f4faca71ef59c2e6b685d8838f979711bbb32c9160089ec
3
- size 224313
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62d0f0f8582dfccfd66e3fcbcc075bf489151ab6e3f1e9952e0c29f7ce0f94a2
3
+ size 191908
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 506.5, "std_reward": 134.2209000118834, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-22T12:36:22.603647"}
 
1
+ {"mean_reward": 760.5, "std_reward": 315.3287966551739, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-23T22:08:57.758691"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2863c9b19c058b21acb8107fc51aaa2e27bc465920abc7ac46392cb3e697b1f8
3
- size 37100
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b792b71cee9fb2468c313cb537f5622066b9ede39647301199ce7ffe3e24d7d5
3
+ size 266469