IngoTB303 commited on
Commit
9c0b89d
·
1 Parent(s): 7bf16c0

Initial commit

Browse files
README.md CHANGED
@@ -16,7 +16,7 @@ model-index:
16
  type: SpaceInvadersNoFrameskip-v4
17
  metrics:
18
  - type: mean_reward
19
- value: 726.50 +/- 266.77
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', 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),
 
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
  ## 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),
args.yml CHANGED
@@ -44,7 +44,7 @@
44
  - - optimize_hyperparameters
45
  - false
46
  - - progress
47
- - false
48
  - - pruner
49
  - median
50
  - - sampler
@@ -54,7 +54,7 @@
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
- - 3188977004
58
  - - storage
59
  - null
60
  - - study_name
@@ -64,7 +64,7 @@
64
  - - track
65
  - false
66
  - - trained_agent
67
- - ''
68
  - - truncate_last_trajectory
69
  - true
70
  - - uuid
 
44
  - - optimize_hyperparameters
45
  - false
46
  - - progress
47
+ - true
48
  - - pruner
49
  - median
50
  - - sampler
 
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
+ - 924316529
58
  - - storage
59
  - null
60
  - - study_name
 
64
  - - track
65
  - false
66
  - - trained_agent
67
+ - logs/dqn/SpaceInvadersNoFrameskip-v4_1/SpaceInvadersNoFrameskip-v4.zip
68
  - - truncate_last_trajectory
69
  - true
70
  - - uuid
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
 
2
  - - - batch_size
3
  - 32
4
  - - buffer_size
5
+ - 120000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
 
14
  - - gradient_steps
15
  - 1
16
  - - learning_rate
17
+ - 0.00015
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 sha256:465db1134f583d8694d10cd01de1d393e46a2051e5defec8a5aceb9f472528d6
3
- size 27224788
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15bc91ffba3f8ddd0b4208ccde4d549e71639d837851fafebb40710b1aa66fc5
3
+ size 27224509
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 sha256:70b08acccc0eeed786ed4883f18026a05f20c2567abbdf0256b44760242b203b
3
- size 13505739
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2e045707c3cd126842a902e25dec20823089f9b8e8ba95d9d3910ce20f7c2a6
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:fc5d80291519693e16e4756ecc5003c24c636eb00b7247c7c8f5b8b4783ff67b
3
- size 13504937
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a6cd58380a235acbac72c098e350c544eb77e7135de46032c24729ecd47e73f8
3
+ size 13504745
dqn-SpaceInvadersNoFrameskip-v4/system_info.txt CHANGED
@@ -1,7 +1,7 @@
1
- OS: Linux-5.10.133+-x86_64-with-glibc2.27 #1 SMP Fri Aug 26 08:44:51 UTC 2022
2
- Python: 3.8.16
3
  Stable-Baselines3: 1.6.2
4
- PyTorch: 1.13.0+cu116
5
- GPU Enabled: True
6
- Numpy: 1.21.6
7
  Gym: 0.21.0
 
1
+ OS: Linux-5.15.0-56-generic-x86_64-with-glibc2.29 #62~20.04.1-Ubuntu SMP Tue Nov 22 21:24:20 UTC 2022
2
+ Python: 3.8.10
3
  Stable-Baselines3: 1.6.2
4
+ PyTorch: 1.13.1+cu117
5
+ GPU Enabled: False
6
+ Numpy: 1.24.0
7
  Gym: 0.21.0
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8c1ab88c8945f8788b846f2aadaff2d8e370185f83028a64dc8dd23cb032ad66
3
- size 204407
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4f78fa4a20c9164eea3b20b1e344bec7824739f6791b745b4f4134ca17258fd
3
+ size 203689
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 726.5, "std_reward": 266.76815777000076, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-20T19:13:49.522113"}
 
1
+ {"mean_reward": 632.5, "std_reward": 131.68617998863814, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2022-12-21T12:36:29.112498"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:38bb7d56c215cd6c6778687763857bfdf709d8aa21cb2279be57405ddfbc706a
3
- size 36895
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a38f5ac818970ae681b437b22171cae233c9eda69347bbf5d2f586a2ad5fa61
3
+ size 29418