ValentinGuigon commited on
Commit
a5d254d
1 Parent(s): 7235a9a

Initial commit

Browse files
README.md CHANGED
@@ -16,7 +16,7 @@ model-index:
16
  type: SpaceInvadersNoFrameskip-v4
17
  metrics:
18
  - type: mean_reward
19
- value: 29.00 +/- 64.30
20
  name: mean_reward
21
  verified: false
22
  ---
@@ -63,16 +63,16 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
63
  ## Hyperparameters
64
  ```python
65
  OrderedDict([('batch_size', 32),
66
- ('buffer_size', 1000000),
67
  ('env_wrapper',
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', 0.01),
74
- ('learning_starts', 10000),
75
- ('n_timesteps', 100000.0),
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: 609.00 +/- 78.38
20
  name: mean_reward
21
  verified: false
22
  ---
 
63
  ## Hyperparameters
64
  ```python
65
  OrderedDict([('batch_size', 32),
66
+ ('buffer_size', 100000),
67
  ('env_wrapper',
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),
77
  ('policy', 'CnnPolicy'),
78
  ('target_update_interval', 1000),
args.yml CHANGED
@@ -56,7 +56,7 @@
56
  - - save_replay_buffer
57
  - false
58
  - - seed
59
- - 3875727083
60
  - - storage
61
  - null
62
  - - study_name
 
56
  - - save_replay_buffer
57
  - false
58
  - - seed
59
+ - 2032044146
60
  - - storage
61
  - null
62
  - - study_name
config.yml CHANGED
@@ -2,7 +2,7 @@
2
  - - - batch_size
3
  - 32
4
  - - buffer_size
5
- - 1000000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
@@ -14,11 +14,11 @@
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
- - 0.01
18
  - - learning_starts
19
- - 10000
20
  - - n_timesteps
21
- - 100000.0
22
  - - optimize_memory_usage
23
  - false
24
  - - policy
 
2
  - - - batch_size
3
  - 32
4
  - - buffer_size
5
+ - 100000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
 
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
+ - 0.0001
18
  - - learning_starts
19
+ - 100000
20
  - - n_timesteps
21
+ - 1000000.0
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 sha256:64ac5d6fb452aec5a499256e8eb29b41de48581a0a00162069d5f973773fd644
3
- size 27220722
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f31f6e1033475bd2b739b81fa6570249d69d1686ecd1402ab6971f8d18a3833a
3
+ size 27220728
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:b79d4d52a7733a851b736e4f63b0f2007f8a75d206eb5b3919cdf98902126716
3
  size 13506172
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:53411f0b62c81a7a7f1bcb397d7d1eb1db8212bf006c79f91cc437d7b1fbeff7
3
  size 13506172
dqn-SpaceInvadersNoFrameskip-v4/policy.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:47759d3bdddbe23dead13d3f560e6fce89246c6f73f5ac292bf2428b99c3daaf
3
  size 13505370
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8e089e98098a8454b2fae4ee9768ee0dc2d8ff7bc44a4d68deaa014758318f7
3
  size 13505370
dqn-SpaceInvadersNoFrameskip-v4/system_info.txt CHANGED
@@ -1,7 +1,7 @@
1
  - OS: Linux-6.1.85+-x86_64-with-glibc2.35 # 1 SMP PREEMPT_DYNAMIC Thu Jun 27 21:05:47 UTC 2024
2
  - Python: 3.10.12
3
  - Stable-Baselines3: 2.4.0a7
4
- - PyTorch: 2.4.0+cu121
5
  - GPU Enabled: True
6
  - Numpy: 1.26.4
7
  - Cloudpickle: 2.2.1
 
1
  - OS: Linux-6.1.85+-x86_64-with-glibc2.35 # 1 SMP PREEMPT_DYNAMIC Thu Jun 27 21:05:47 UTC 2024
2
  - Python: 3.10.12
3
  - Stable-Baselines3: 2.4.0a7
4
+ - PyTorch: 2.4.1+cu121
5
  - GPU Enabled: True
6
  - Numpy: 1.26.4
7
  - Cloudpickle: 2.2.1
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6332f13a1224ba686797bd770b10af974d632c130bd192c46a69bdf79b1811c4
3
- size 294871
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74c7b629f17d0597029de94c16fdc6f3deaa2c95e4e7a860e2855af40d3e90eb
3
+ size 215094
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 29.0, "std_reward": 64.29618962271404, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2024-09-14T17:01:21.733283"}
 
1
+ {"mean_reward": 609.0, "std_reward": 78.38367176906169, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2024-09-22T15:48:28.126075"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6250f705c487a6f738b7d8c0a9378c656c488b4cc28136d0ea5e19ca40be8667
3
- size 4904
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:feeaa014634ad80f8d488830a56b4e784838d0ada2fa366a3c997966be170c88
3
+ size 36001