Hugging Face's logo

Alex48
/
SpaceInvadersNoFrameskip-v4

Reinforcement Learning
stable-baselines3
SpaceInvadersNoFrameskip-v4
deep-reinforcement-learning
Eval Results (legacy)
Model card Files
xet
Community
Alex48 commited on
Commit
f581432
·
1 Parent(s): a336155

Initial commit

Browse files
Files changed (8) hide show
  1. README.md +2 -2
  2. args.yml +9 -5
  3. config.yml +2 -2
  4. dqn-SpaceInvadersNoFrameskip-v4.zip +2 -2
  5. dqn-SpaceInvadersNoFrameskip-v4/data +0 -0
  6. dqn-SpaceInvadersNoFrameskip-v4/policy.pth +1 -1
  7. results.json +1 -1
  8. train_eval_metrics.zip +2 -2
README.md CHANGED
@@ -62,7 +62,7 @@ python -m rl_zoo3.push_to_hub --algo dqn --env SpaceInvadersNoFrameskip-v4 -f lo
62
 
63
  ## Hyperparameters
64
  ```python
65
- OrderedDict([('batch_size', 32),
66
  ('buffer_size', 100000),
67
  ('env_wrapper',
68
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
@@ -70,7 +70,7 @@ OrderedDict([('batch_size', 32),
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),
 
62
 
63
  ## Hyperparameters
64
  ```python
65
+ OrderedDict([('batch_size', 84),
66
  ('buffer_size', 100000),
67
  ('env_wrapper',
68
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
 
70
  ('exploration_fraction', 0.1),
71
  ('frame_stack', 4),
72
  ('gradient_steps', 1),
73
+ ('learning_rate', 0.001),
74
  ('learning_starts', 100000),
75
  ('n_timesteps', 10000000.0),
76
  ('optimize_memory_usage', False),
args.yml CHANGED
@@ -1,6 +1,8 @@
1
  !!python/object/apply:collections.OrderedDict
2
  - - - algo
3
  - dqn
 
 
4
  - - conf_file
5
  - null
6
  - - device
@@ -16,7 +18,9 @@
16
  - - gym_packages
17
  - []
18
  - - hyperparams
19
- - null
 
 
20
  - - log_folder
21
  - logs/
22
  - - log_interval
@@ -32,7 +36,7 @@
32
  - - n_startup_trials
33
  - 10
34
  - - n_timesteps
35
- - 1000000
36
  - - n_trials
37
  - 500
38
  - - no_optim_plots
@@ -50,11 +54,11 @@
50
  - - sampler
51
  - tpe
52
  - - save_freq
53
- - 100000
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
- - 3242211519
58
  - - storage
59
  - null
60
  - - study_name
@@ -64,7 +68,7 @@
64
  - - track
65
  - false
66
  - - trained_agent
67
- - ''
68
  - - truncate_last_trajectory
69
  - true
70
  - - uuid
 
1
  !!python/object/apply:collections.OrderedDict
2
  - - - algo
3
  - dqn
4
+ - - batch_size
5
+ - -1
6
  - - conf_file
7
  - null
8
  - - device
 
18
  - - gym_packages
19
  - []
20
  - - hyperparams
21
+ - batch_size: 84
22
+ buffer_size: 100000
23
+ learning_rate: 0.001
24
  - - log_folder
25
  - logs/
26
  - - log_interval
 
36
  - - n_startup_trials
37
  - 10
38
  - - n_timesteps
39
+ - 100000
40
  - - n_trials
41
  - 500
42
  - - no_optim_plots
 
54
  - - sampler
55
  - tpe
56
  - - save_freq
57
+ - -1
58
  - - save_replay_buffer
59
  - false
60
  - - seed
61
+ - 274124340
62
  - - storage
63
  - null
64
  - - study_name
 
68
  - - track
69
  - false
70
  - - trained_agent
71
+ - logs/dqn/SpaceInvadersNoFrameskip-v4_1/SpaceInvadersNoFrameskip-v4.zip
72
  - - truncate_last_trajectory
73
  - true
74
  - - uuid
config.yml CHANGED
@@ -1,6 +1,6 @@
1
  !!python/object/apply:collections.OrderedDict
2
  - - - batch_size
3
- - 32
4
  - - buffer_size
5
  - 100000
6
  - - env_wrapper
@@ -14,7 +14,7 @@
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
- - 0.0001
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
 
1
  !!python/object/apply:collections.OrderedDict
2
  - - - batch_size
3
+ - 84
4
  - - buffer_size
5
  - 100000
6
  - - env_wrapper
 
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
+ - 0.001
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 sha256:54bb5b358f87aafb8246d279df3b2218a18fc1dd8e01f3e696f424a89428838e
3
- size 27224807
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd1650c5a90766289b63f09fe8645f73647ae46eb2256adb5157c107436faee9
3
+ size 27224804
dqn-SpaceInvadersNoFrameskip-v4/data CHANGED
The diff for this file is too large to render. See raw diff
 
dqn-SpaceInvadersNoFrameskip-v4/policy.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5884d000da089a1d98d2f9284741a6ac57d52037035966483b1dc0f7ebebbc61
3
  size 13504937
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:007d3dd8a4fd60ba48ac4391bdb32a76b5c19153042cda913c91551ff594bcfb
3
  size 13504937
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 794.0, "std_reward": 247.2326839234651, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-02-21T01:34:09.309806"}
 
1
+ {"mean_reward": 794.0, "std_reward": 247.2326839234651, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-02-21T01:51:30.055701"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:14d915c56248fdebdce8c0310d15a3aa0f7c7ba9cfecb6f43cb35f65efbb59a9
3
- size 37873
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94d25b9d422cb8e0bf36ce3db93b5caa5950e032db31d26e3b41b84d60f85b7b
3
+ size 5552