Hugging Face's logo

IngoTB303
/
dqn-SpaceInvadersNoFrameskip-v4

Reinforcement Learning
stable-baselines3
SpaceInvadersNoFrameskip-v4
deep-reinforcement-learning
Eval Results
Model card Files Community
IngoTB303 commited on
Commit
5a8bc40
·
1 Parent(s): 9c0b89d

Initial commit

Browse files
Files changed (10) hide show
  1. README.md +4 -4
  2. args.yml +2 -2
  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: 632.50 +/- 131.69
20
  name: mean_reward
21
  verified: false
22
  ---
@@ -58,14 +58,14 @@ 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', 120000),
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', 0.00015),
69
  ('learning_starts', 100000),
70
  ('n_timesteps', 1000000.0),
71
  ('optimize_memory_usage', False),
 
16
  type: SpaceInvadersNoFrameskip-v4
17
  metrics:
18
  - type: mean_reward
19
+ value: 792.50 +/- 265.08
20
  name: mean_reward
21
  verified: false
22
  ---
 
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
  ('optimize_memory_usage', False),
args.yml CHANGED
@@ -54,7 +54,7 @@
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
- - 924316529
58
  - - storage
59
  - null
60
  - - study_name
@@ -64,7 +64,7 @@
64
  - - track
65
  - false
66
  - - trained_agent
67
- - logs/dqn/SpaceInvadersNoFrameskip-v4_1/SpaceInvadersNoFrameskip-v4.zip
68
  - - truncate_last_trajectory
69
  - true
70
  - - uuid
 
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
+ - 533738464
58
  - - storage
59
  - null
60
  - - study_name
 
64
  - - track
65
  - false
66
  - - trained_agent
67
+ - logs/dqn/SpaceInvadersNoFrameskip-v4_2/SpaceInvadersNoFrameskip-v4.zip
68
  - - truncate_last_trajectory
69
  - true
70
  - - uuid
config.yml CHANGED
@@ -2,19 +2,19 @@
2
  - - - batch_size
3
  - 32
4
  - - buffer_size
5
- - 120000
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
17
- - 0.00015
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
 
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
dqn-SpaceInvadersNoFrameskip-v4.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:15bc91ffba3f8ddd0b4208ccde4d549e71639d837851fafebb40710b1aa66fc5
3
- size 27224509
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1fb5faf87466dc2ddc332743adbc12f14c0e5b0db142e307cbfa22572cb7670
3
+ size 27224510
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:a2e045707c3cd126842a902e25dec20823089f9b8e8ba95d9d3910ce20f7c2a6
3
  size 13505419
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e63fd4d06b0a903e71bd19ef73b6b058efa4897d5d89d2e04960679bd40495ce
3
  size 13505419
dqn-SpaceInvadersNoFrameskip-v4/policy.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a6cd58380a235acbac72c098e350c544eb77e7135de46032c24729ecd47e73f8
3
  size 13504745
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cbe4392bfa8b35678ef63a1777edb69f447097d5f568bac61ab244018643d850
3
  size 13504745
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d4f78fa4a20c9164eea3b20b1e344bec7824739f6791b745b4f4134ca17258fd
3
- size 203689
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94ffe08a588941b46066b788f88a75612498c6207e652c7ceb2e16301bf47d2b
3
+ size 210111
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 632.5, "std_reward": 131.68617998863814, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-21T12:36:29.112498"}
 
1
+ {"mean_reward": 792.5, "std_reward": 265.0778187627173, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-21T14:08:17.092615"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7a38f5ac818970ae681b437b22171cae233c9eda69347bbf5d2f586a2ad5fa61
3
- size 29418
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4eadc8dbbddd83f4bfa94d93ed097f9aec94dbf489feb900272132c62e622f48
3
+ size 28484