Hugging Face's logo

pasto2003
/
dqn-SpaceInvadersNoFrameskip-v4

Reinforcement Learning
stable-baselines3
SpaceInvadersNoFrameskip-v4
deep-reinforcement-learning
Eval Results
Model card Files Community
pasto2003 commited on
Commit
b9af72a
·
1 Parent(s): de112fa

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: 161.00 +/- 61.35
20
  name: mean_reward
21
  verified: false
22
  ---
@@ -63,19 +63,19 @@ 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', 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', 10000000.0),
76
  ('optimize_memory_usage', False),
77
  ('policy', 'CnnPolicy'),
78
- ('target_update_interval', 1000),
79
  ('train_freq', 4),
80
  ('normalize', False)])
81
  ```
 
16
  type: SpaceInvadersNoFrameskip-v4
17
  metrics:
18
  - type: mean_reward
19
+ value: 497.50 +/- 227.36
20
  name: mean_reward
21
  verified: false
22
  ---
 
63
  ## Hyperparameters
64
  ```python
65
  OrderedDict([('batch_size', 32),
66
+ ('buffer_size', 200000),
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', 0.0002),
74
  ('learning_starts', 100000),
75
  ('n_timesteps', 10000000.0),
76
  ('optimize_memory_usage', False),
77
  ('policy', 'CnnPolicy'),
78
+ ('target_update_interval', 750),
79
  ('train_freq', 4),
80
  ('normalize', False)])
81
  ```
args.yml CHANGED
@@ -54,7 +54,7 @@
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
- - 3379052623
58
  - - storage
59
  - null
60
  - - study_name
 
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
+ - 1176898861
58
  - - storage
59
  - null
60
  - - study_name
config.yml CHANGED
@@ -2,7 +2,7 @@
2
  - - - batch_size
3
  - 32
4
  - - buffer_size
5
- - 100000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
@@ -14,7 +14,7 @@
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
- - 0.0001
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
@@ -24,6 +24,6 @@
24
  - - policy
25
  - CnnPolicy
26
  - - target_update_interval
27
- - 1000
28
  - - train_freq
29
  - 4
 
2
  - - - batch_size
3
  - 32
4
  - - buffer_size
5
+ - 200000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
 
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
+ - 0.0002
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
 
24
  - - policy
25
  - CnnPolicy
26
  - - target_update_interval
27
+ - 750
28
  - - train_freq
29
  - 4
dqn-SpaceInvadersNoFrameskip-v4.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7efcfbc77d825b426a2341ec53b46c1f6846a12631c03a19f81892f8daefc50e
3
- size 27218511
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fa2b16dd294f26beadf23454c79a71518464e64a28ddf105ff85cf5759db069
3
+ size 27218499
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 sha256:8905fce74aeab15d09240bf91dee31a173702064d661d682d390032b7bd11aab
3
  size 13505739
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c3025b1f0737c9d8c3046d236700eb272426e07a8d4274d4c3c5cdf121ef30c
3
  size 13505739
dqn-SpaceInvadersNoFrameskip-v4/policy.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:84510446485b2e2f9de1b5f01ef89e48962a06afc36b6596a52abad4f8bf8b88
3
  size 13504937
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d2c00f2fd1a807fc90b595d486d70b6b6b014d803f07f20e9e0a04e5752fedb
3
  size 13504937
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9d97320299436dd0b6c1626ee8492e44f8cbf97d0d0f03c4eaffc0143f3e241b
3
- size 262403
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5382d7408a71c81da8fb6e0744cd6a2fae2aec7935c7e1977570c2f6586a371c
3
+ size 255649
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 161.0, "std_reward": 61.35144660071187, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-08-20T07:41:39.605637"}
 
1
+ {"mean_reward": 497.5, "std_reward": 227.35709797584943, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-08-20T08:47:19.047686"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b250907427106ac5ada6014e824259fe5e86798d7afbdc9bdd79704de90baf77
3
- size 12908
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12ebf239391dcef9686702299f48af9874d138bd3b5eb9d177141d327340da66
3
+ size 35614