Hugging Face's logo

RamonAnkersmit
/
dqn-SpaceInvadersNoFrameskip-v4

Reinforcement Learning
stable-baselines3
SpaceInvadersNoFrameskip-v4
deep-reinforcement-learning
Eval Results (legacy)
Model card Files
xet
 Community
RamonAnkersmit commited on
Commit
06c8c34
·
1 Parent(s): 97f4765

Initial commit

Browse files
Files changed (10) hide show
  1. README.md +6 -6
  2. args.yml +1 -1
  3. config.yml +5 -5
  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. 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: 493.00 +/- 211.86
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', 64),
61
- ('buffer_size', 100000),
62
  ('env_wrapper',
63
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
64
  ('exploration_final_eps', 0.01),
65
- ('exploration_fraction', 0.025),
66
- ('frame_stack', 3),
67
  ('gradient_steps', 1),
68
  ('learning_rate', 0.0001),
69
  ('learning_starts', 100000),
70
- ('n_timesteps', 2500000.0),
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: 591.50 +/- 117.71
20
  name: mean_reward
21
  verified: false
22
  ---
 
57
 
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', 0.1),
66
+ ('frame_stack', 4),
67
  ('gradient_steps', 1),
68
  ('learning_rate', 0.0001),
69
  ('learning_starts', 100000),
70
+ ('n_timesteps', 1000000),
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
- - 665862491
58
  - - storage
59
  - null
60
  - - study_name
 
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
+ - 680273506
58
  - - storage
59
  - null
60
  - - study_name
config.yml CHANGED
@@ -1,16 +1,16 @@
1
  !!python/object/apply:collections.OrderedDict
2
  - - - batch_size
3
- - 64
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
- - 0.025
12
  - - frame_stack
13
- - 3
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
@@ -18,7 +18,7 @@
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
- - 2500000.0
22
  - - optimize_memory_usage
23
  - false
24
  - - policy
 
1
  !!python/object/apply:collections.OrderedDict
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
+ - 0.1
12
  - - frame_stack
13
+ - 4
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
 
18
  - - learning_starts
19
  - 100000
20
  - - n_timesteps
21
+ - 1000000
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 sha256:3054fddc2a9340c2706543e36efa823fb8c57c5d92885fb46406c8f6aba01015
3
- size 27144084
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5897a2732b277b6a3fb897fadd8567df6b57bfd774ba4bdd4ae8b4fb38d9c65c
3
+ size 27224794
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:c4fd9c2ea8c0d6ada31a314ff513ce991766aada324aaecddab246db33bb36df
3
- size 13489355
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc61e9e0f53f13d457ed9b896cee38cc3c084aa44cb1225d9fc9d2c056ec88d4
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:a112295547df644d83f076a7203552265068b49af626082f4023817c88b3cc92
3
- size 13488553
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b77fa5493dde9afad62e8e4002832884feaa3a9c4093fe166c22f25d43e84532
3
+ size 13504937
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a06bc80efefaea257067fcd81ea754205edd812c9f9bc4d7b7bb178e360f277e
3
- size 220484
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a8db153ebffe68c84da0628cbe14517973c02731270d34002b8a5ccdcb68a7c
3
+ size 175245
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 493.0, "std_reward": 211.8631633861819, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-30T22:08:52.104860"}
 
1
+ {"mean_reward": 591.5, "std_reward": 117.70832595870183, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-30T23:34:44.336315"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4135019886216595b7d9f54bc773ebb25269aff4f5114fed9a1a933680cf545d
3
- size 79360
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5c0882310e6cc2969a39b3d1cb690f86098aec9d6b7b547b3b4cbc2137d7afe
3
+ size 36379