Hugging Face's logo

mnavas
/
spaceinv-1

Reinforcement Learning
stable-baselines3
SpaceInvadersNoFrameskip-v4
deep-reinforcement-learning
Eval Results
Model card Files Community
María Navas Loro commited on
Commit
a2d24ae
·
1 Parent(s): de1354f

Initial commit

Browse files
Files changed (10) 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. 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: 484.50 +/- 189.89
20
  name: mean_reward
21
  verified: false
22
  ---
@@ -57,17 +57,17 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
57
 
58
  ## Hyperparameters
59
  ```python
60
- OrderedDict([('batch_size', 16),
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: 660.50 +/- 150.22
20
  name: mean_reward
21
  verified: false
22
  ---
 
57
 
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', 0.02),
66
  ('frame_stack', 4),
67
  ('gradient_steps', 1),
68
  ('learning_rate', 0.0001),
69
  ('learning_starts', 100000),
70
+ ('n_timesteps', 2000000),
71
  ('optimize_memory_usage', False),
72
  ('policy', 'CnnPolicy'),
73
  ('target_update_interval', 1000),
args.yml CHANGED
@@ -54,7 +54,7 @@
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
- - 2582964334
58
  - - storage
59
  - null
60
  - - study_name
 
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
+ - 218349997
58
  - - storage
59
  - null
60
  - - study_name
config.yml CHANGED
@@ -1,6 +1,6 @@
1
  !!python/object/apply:collections.OrderedDict
2
  - - - batch_size
3
- - 16
4
  - - buffer_size
5
  - 100000
6
  - - env_wrapper
@@ -8,7 +8,7 @@
8
  - - exploration_final_eps
9
  - 0.01
10
  - - exploration_fraction
11
- - 0.1
12
  - - frame_stack
13
  - 4
14
  - - gradient_steps
@@ -18,7 +18,7 @@
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
- - 1000000.0
22
  - - optimize_memory_usage
23
  - false
24
  - - policy
 
1
  !!python/object/apply:collections.OrderedDict
2
  - - - batch_size
3
+ - 32
4
  - - buffer_size
5
  - 100000
6
  - - env_wrapper
 
8
  - - exploration_final_eps
9
  - 0.01
10
  - - exploration_fraction
11
+ - 0.02
12
  - - frame_stack
13
  - 4
14
  - - gradient_steps
 
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
+ - 2000000
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 sha256:068d975e2c0ebceb647074bcc889a425e210e1ae7351fad8baf5e5ad69734471
3
- size 27224794
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc96635814eaac3b1cb3001d7c243981e221a50269fc28b950d421a1d9eb1ca2
3
+ size 27224795
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 sha256:8e5c2564023dd54bd17f7cf2ddb5e1d02dde30219bb3d18df82625a66e2dddc6
3
  size 13505739
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fbba84a82c76e7bbea717c8f20aa8383e3f413c8b300bc3001798dd3326e9e1d
3
  size 13505739
dqn-SpaceInvadersNoFrameskip-v4/policy.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ac756b6f4fbffb222f837eefcb13abc0897660c02b6210e90b58375f302d06e5
3
  size 13504937
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e38a6cac1f8d679e5c8946d5b4dfc6057197c2bbce5468a169535baaefdfd3cd
3
  size 13504937
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bbef6a5c7add83ade35ab14d504bf32ea8eed94eeb3fbce3267fda6f2ffe4afd
3
- size 195104
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a21d1c3b750f6ecaaa06c667720ca0afbb3e57d93885a5b07c830fe4461cdf76
3
+ size 191441
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 484.5, "std_reward": 189.88746667434373, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-23T19:50:55.848344"}
 
1
+ {"mean_reward": 660.5, "std_reward": 150.22399941420812, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-24T18:00:39.428122"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e95c635a389a19905ca96c1b27f6de504ca20a6e79290c62b3e8bb8af6c92393
3
- size 39815
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc8d7a26874b980843290ab181919bbad8369d88d9aba2a2f3c8e954a8585f21
3
+ size 64412