Hugging Face's logo

jmsalvi
/
dqn-SpaceInvadersNoFrameskip-v4

Reinforcement Learning stable-baselines3 SpaceInvadersNoFrameskip-v4 deep-reinforcement-learning Eval Results
Model card Files Community
jmsalvi commited on
Commit
027338d
•
1 Parent(s): 488dca8

Initial commit

Browse files
Files changed (11) hide show
  1. README.md +5 -5
  2. args.yml +3 -1
  3. config.yml +4 -4
  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 +3 -3
  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: 171.00 +/- 32.92
20
  name: mean_reward
21
  verified: false
22
  ---
@@ -58,16 +58,16 @@ 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', 100000),
62
  ('env_wrapper',
63
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
64
  ('exploration_final_eps', 0.01),
65
- ('exploration_fraction', 0.1),
66
- ('frame_stack', 4),
67
  ('gradient_steps', 1),
68
  ('learning_rate', 0.0001),
69
  ('learning_starts', 100000),
70
- ('n_timesteps', 1000000.0),
71
  ('optimize_memory_usage', False),
72
  ('policy', 'CnnPolicy'),
73
  ('target_update_interval', 1000),
16
  type: SpaceInvadersNoFrameskip-v4
17
  metrics:
18
  - type: mean_reward
19
+ value: 457.50 +/- 157.18
20
  name: mean_reward
21
  verified: false
22
  ---
58
  ## Hyperparameters
59
  ```python
60
  OrderedDict([('batch_size', 32),
61
+ ('buffer_size', 150000),
62
  ('env_wrapper',
63
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
64
  ('exploration_final_eps', 0.01),
65
+ ('exploration_fraction', 0.15),
66
+ ('frame_stack', 3),
67
  ('gradient_steps', 1),
68
  ('learning_rate', 0.0001),
69
  ('learning_starts', 100000),
70
+ ('n_timesteps', 1200000.0),
71
  ('optimize_memory_usage', False),
72
  ('policy', 'CnnPolicy'),
73
  ('target_update_interval', 1000),
args.yml CHANGED
@@ -1,6 +1,8 @@
1
  !!python/object/apply:collections.OrderedDict
2
  - - - algo
3
  - dqn
 
 
4
  - - device
5
  - auto
6
  - - env
@@ -52,7 +54,7 @@
52
  - - save_replay_buffer
53
  - false
54
  - - seed
55
- - 2176844926
56
  - - storage
57
  - null
58
  - - study_name
1
  !!python/object/apply:collections.OrderedDict
2
  - - - algo
3
  - dqn
4
+ - - conf_file
5
+ - null
6
  - - device
7
  - auto
8
  - - env
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
+ - 184500454
58
  - - storage
59
  - null
60
  - - study_name
config.yml CHANGED
@@ -2,15 +2,15 @@
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
- - 0.1
12
  - - frame_stack
13
- - 4
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
@@ -18,7 +18,7 @@
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
- - 1000000.0
22
  - - optimize_memory_usage
23
  - false
24
  - - policy
2
  - - - batch_size
3
  - 32
4
  - - buffer_size
5
+ - 150000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
9
  - 0.01
10
  - - exploration_fraction
11
+ - 0.15
12
  - - frame_stack
13
+ - 3
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
+ - 1200000.0
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:b5cd8be34a3f3427f5be4415acf97b039ae924bb2f315a1deb2272efbb98cca7
3
- size 27224959
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0dfa399c7837af47c4513692d132d6183e1c88610e8c527eb8c1bfa50ae4a16
3
+ size 27144077
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:beba3d931a1b4525c54da44efe65a28d06c565815bae5a753aaa7246aa320f0f
3
- size 13505739
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f53aaf5d0ee3ab9baba5d69fd3ad6ba136c1b5a49904cb521f7899d75561b6a
3
+ size 13489355
dqn-SpaceInvadersNoFrameskip-v4/policy.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:550edc451702fe2adcb97dd077569118dd721d6b482f894e75930c087f7319d6
3
- size 13504937
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:703e2c80fd154688015d3d57d846cb38d417b5d4c933cffe9e9e12c4d4c1558b
3
+ size 13488553
dqn-SpaceInvadersNoFrameskip-v4/system_info.txt CHANGED
@@ -1,7 +1,7 @@
1
- OS: Linux-5.10.133+-x86_64-with-Ubuntu-18.04-bionic #1 SMP Fri Aug 26 08:44:51 UTC 2022
2
- Python: 3.7.15
3
  Stable-Baselines3: 1.6.2
4
- PyTorch: 1.12.1+cu113
5
  GPU Enabled: True
6
  Numpy: 1.21.6
7
  Gym: 0.21.0
1
+ OS: Linux-5.10.133+-x86_64-with-glibc2.27 #1 SMP Fri Aug 26 08:44:51 UTC 2022
2
+ Python: 3.8.16
3
  Stable-Baselines3: 1.6.2
4
+ PyTorch: 1.13.0+cu116
5
  GPU Enabled: True
6
  Numpy: 1.21.6
7
  Gym: 0.21.0
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f89b24ca84a29ab210f6488d87d909d392ae823c1b3af7956137e3e302aaf735
3
- size 243904
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:10940277f02f144afa99a6332ba7587a6cc247604e860e40cddca02269ae407a
3
+ size 199091
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 171.0, "std_reward": 32.92415526630866, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-11-12T05:00:16.212793"}
1
+ {"mean_reward": 457.5, "std_reward": 157.18221909618148, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-09T07:29:37.576281"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c5f14088058ac066596a74c81e106cb52eb1bf56934eaeebc54c737ebe8cf4b8
3
- size 9393
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8016b282b5ddb1f1e9348f0c7cc1485c229ca87f4868ddda7029367b4bcc468d
3
+ size 43228