c-gohlke commited on
Commit
baddf25
·
1 Parent(s): b85dbde

Upload new model run

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ models/LunarLander-v2/Sac/92409bddba5240b98564ef3fdffccb22/demo.mp4 filter=lfs diff=lfs merge=lfs -text
models/LunarLander-v2/Sac/92409bddba5240b98564ef3fdffccb22/README.md ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ model-index:
3
+ - name: mock
4
+ results:
5
+ - task:
6
+ type: reinforcement-learning
7
+ name: reinforcement-learning
8
+ dataset:
9
+ name: mock
10
+ type: mock
11
+ metrics:
12
+ - type: mean_reward
13
+ value: mock +/- mock
14
+ name: mean_reward
15
+ verified: false
16
+ tags:
17
+ - reinforcement-learning
18
+ - connect-four
19
+ - pytorch
20
+ - custom-implementation
21
+ - generic
22
+ run_info:
23
+ litrl:
24
+ version: 0.0.9
25
+ repo_url: git@github.com:c-gohlke/LightningRL.git
26
+ commit_hash: 3a8eee368ecc3618d243299d19278175bc846adf
27
+ ---
models/LunarLander-v2/Sac/92409bddba5240b98564ef3fdffccb22/config.yaml ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ run_id: ${mlflow:null}
2
+ tags:
3
+ model: ${model._target_}
4
+ env: ${model.cfg.env_id}
5
+ run_id: ${run_id}
6
+ load_path: latest
7
+ model:
8
+ _target_: litrl.algo.sac.model.Sac
9
+ cfg:
10
+ seed: 123
11
+ val_env_seed: 10000
12
+ env_id: LunarLander-v2
13
+ env_fabric:
14
+ _target_: litrl.make
15
+ id: ${model.cfg.env_id}
16
+ val_env_fabric:
17
+ _target_: litrl.make
18
+ id: ${model.cfg.env_id}
19
+ val: true
20
+ render_each_n_episodes: 1
21
+ lr: 0.0001
22
+ gamma: 0.99
23
+ warm_start_steps: 10000.0
24
+ hidden_size: 64
25
+ n_hidden_layers: 2
26
+ epsilon: 0
27
+ tau: 0.005
28
+ target_entropy: -2
29
+ buffer:
30
+ batch_size: 256
31
+ max_size: 1000000.0
32
+ trainer:
33
+ _target_: lightning.Trainer
34
+ log_every_n_steps: 500
35
+ max_epochs: 3
36
+ limit_train_batches: 5000
37
+ logger:
38
+ _target_: litrl.common.logger.LitRLLogger
39
+ log_model: true
40
+ tags: ${tags}
41
+ run_id: ${run_id}
42
+ callbacks:
43
+ - _target_: litrl.common.huggingface.HuggingfaceCallback
44
+ mode: max
45
+ monitor: val_reward
46
+ filename: '{val_reward:.2f}-{step}'
models/LunarLander-v2/Sac/92409bddba5240b98564ef3fdffccb22/demo.mp4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0458a405fcd8d8d98aa4a1ae445239b5607b75764b8b09db893b9ed710837d72
3
+ size 3205381
models/LunarLander-v2/Sac/92409bddba5240b98564ef3fdffccb22/model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b6a1e55a27bff46aa6aff938c43c016b6c541bf979092fa5a80a77c01142410
3
+ size 20790
models/LunarLander-v2/Sac/92409bddba5240b98564ef3fdffccb22/model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae84bdd459a1d53eee5357f1098abb2dca65ee5400248563f28aa193788b34dd
3
+ size 248974
models/LunarLander-v2/Sac/92409bddba5240b98564ef3fdffccb22/model_config.yaml ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ root: !!python/object:litrl.common.schema.ModelConfigSchema
2
+ buffer: !!python/object:litrl.common.schema.BufferSchema
3
+ batch_size: 256
4
+ max_size: 1000000
5
+ env_fabric: !!python/object:litrl.common.schema.EnvSchema
6
+ _target_: litrl.make
7
+ id: LunarLander-v2
8
+ opponent: null
9
+ val: null
10
+ epsilon: 0.0
11
+ gamma: 0.99
12
+ hidden_size: 64
13
+ lr: 0.0001
14
+ n_hidden_layers: 2
15
+ seed: 123
16
+ target_entropy: -2.0
17
+ tau: 0.005
18
+ val_env_fabric: !!python/object:litrl.common.schema.EnvSchema
19
+ _target_: litrl.make
20
+ id: LunarLander-v2
21
+ opponent: null
22
+ val: true
23
+ warm_start_steps: 10000
models/LunarLander-v2/Sac/92409bddba5240b98564ef3fdffccb22/overrides.yaml ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ - model=sac
2
+ - env=lunar_lander
3
+ - trainer.max_epochs=3