Hugging Face's logo

mgmeskill
/
dqn-SpaceInvadersNoFrameskip-v4

Reinforcement Learning
stable-baselines3
SpaceInvadersNoFrameskip-v4
deep-reinforcement-learning
Eval Results (legacy)
Model card Files
xet
Community
mgmeskill commited on
Commit
84d49c2
·
1 Parent(s): 3641a2f

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 +1 -1
  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: 212.00 +/- 182.21
20
  name: mean_reward
21
  verified: false
22
  ---
@@ -63,7 +63,7 @@ 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', 10000000),
67
  ('env_wrapper',
68
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
69
  ('exploration_final_eps', 0.01),
@@ -71,11 +71,11 @@ OrderedDict([('batch_size', 32),
71
  ('frame_stack', 4),
72
  ('gradient_steps', 1),
73
  ('learning_rate', 0.0001),
74
- ('learning_starts', 1000),
75
  ('n_timesteps', 100000.0),
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: 217.00 +/- 77.24
20
  name: mean_reward
21
  verified: false
22
  ---
 
63
  ## Hyperparameters
64
  ```python
65
  OrderedDict([('batch_size', 32),
66
+ ('buffer_size', 1000000),
67
  ('env_wrapper',
68
  ['stable_baselines3.common.atari_wrappers.AtariWrapper']),
69
  ('exploration_final_eps', 0.01),
 
71
  ('frame_stack', 4),
72
  ('gradient_steps', 1),
73
  ('learning_rate', 0.0001),
74
+ ('learning_starts', 5000),
75
  ('n_timesteps', 100000.0),
76
  ('optimize_memory_usage', False),
77
  ('policy', 'CnnPolicy'),
78
+ ('target_update_interval', 5000),
79
  ('train_freq', 4),
80
  ('normalize', False)])
81
  ```
args.yml CHANGED
@@ -54,7 +54,7 @@
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
- - 3693068319
58
  - - storage
59
  - null
60
  - - study_name
 
54
  - - save_replay_buffer
55
  - false
56
  - - seed
57
+ - 1663148039
58
  - - storage
59
  - null
60
  - - study_name
config.yml CHANGED
@@ -2,7 +2,7 @@
2
  - - - batch_size
3
  - 32
4
  - - buffer_size
5
- - 10000000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
@@ -16,7 +16,7 @@
16
  - - learning_rate
17
  - 0.0001
18
  - - learning_starts
19
- - 1000
20
  - - n_timesteps
21
  - 100000.0
22
  - - optimize_memory_usage
@@ -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
+ - 1000000
6
  - - env_wrapper
7
  - - stable_baselines3.common.atari_wrappers.AtariWrapper
8
  - - exploration_final_eps
 
16
  - - learning_rate
17
  - 0.0001
18
  - - learning_starts
19
+ - 5000
20
  - - n_timesteps
21
  - 100000.0
22
  - - optimize_memory_usage
 
24
  - - policy
25
  - CnnPolicy
26
  - - target_update_interval
27
+ - 5000
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 sha256:35e066511ec9c1bedb0812368967730b8967efe0384f3df0d21cdeb54b3e67a7
3
  size 27218385
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc8b5fb58a3b0c0136dea5f501da04eac78ad06197496be7690f60d1f15a14f8
3
  size 27218385
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 sha256:37c0d95817772440ea4a90c2daeeec5cccb248d63bd9bc7b4ce1f3609d8c7c59
3
  size 13505739
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d15ca27d3c77a5d81fc20052266a86f79f7fffb42daaaeed3028d49fbfcdd799
3
  size 13505739
dqn-SpaceInvadersNoFrameskip-v4/policy.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:70d4b48ca5c446960dd75e820ece7f04917f4310366d8434ea4128b85c7212f6
3
  size 13504937
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:090cdca8d2e52e2cb402c69955972a511b3176cdd3e19b49a55b9af9af2b9adb
3
  size 13504937
replay.mp4 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b21dc9868122616364c25e79be6a6636e5db13406d138cc95c5a23d1e7541e6a
3
- size 262458
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfd460b4842fc0a7b94d46788ceda5e96f859cfd5e57ca0e6a6090aced9c9aab
3
+ size 259354
results.json CHANGED
@@ -1 +1 @@
1
- {"mean_reward": 212.0, "std_reward": 182.21141566872257, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-06-17T21:53:02.264173"}
 
1
+ {"mean_reward": 217.0, "std_reward": 77.23988606931007, "is_deterministic": false, "n_eval_episodes": 10, "eval_datetime": "2023-06-18T02:09:41.523982"}
train_eval_metrics.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8bdbb2b2870ed21788fe8af736e6ac63f7f4fbc4d8d7edd2b0af4ad62ccc7061
3
- size 4966
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e8678aaeb9df50a1b6e7f9d5ed75939cfc97214e57f38b899157241b8573ed2
3
+ size 5297