timxiaohangt
commited on
Commit
•
7d1c697
1
Parent(s):
4e80d49
Training in progress, step 200000
Browse files- .gitignore +1 -0
- config.json +78 -0
- pytorch_model.bin +3 -0
- training_args.bin +3 -0
.gitignore
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
checkpoint-*/
|
config.json
ADDED
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"act_dim": 6,
|
3 |
+
"action_tanh": true,
|
4 |
+
"activation_function": "relu",
|
5 |
+
"adv_act_dim": 6,
|
6 |
+
"architectures": [
|
7 |
+
"TrainableDT"
|
8 |
+
],
|
9 |
+
"attn_pdrop": 0.1,
|
10 |
+
"bos_token_id": 50256,
|
11 |
+
"context_size": 20,
|
12 |
+
"embd_pdrop": 0.1,
|
13 |
+
"eos_token_id": 50256,
|
14 |
+
"hidden_size": 128,
|
15 |
+
"initializer_range": 0.02,
|
16 |
+
"lambda1": 0.01,
|
17 |
+
"lambda2": 1.0,
|
18 |
+
"layer_norm_epsilon": 1e-05,
|
19 |
+
"max_ep_len": 1000,
|
20 |
+
"max_ep_return": 15000,
|
21 |
+
"max_obs_len": 1000,
|
22 |
+
"max_obs_return": 11252.035989716649,
|
23 |
+
"model_type": "decision_transformer",
|
24 |
+
"n_head": 1,
|
25 |
+
"n_inner": null,
|
26 |
+
"n_layer": 3,
|
27 |
+
"n_positions": 1024,
|
28 |
+
"pr_act_dim": 6,
|
29 |
+
"reorder_and_upcast_attn": false,
|
30 |
+
"resid_pdrop": 0.1,
|
31 |
+
"returns_scale": 1000,
|
32 |
+
"scale_attn_by_inverse_layer_idx": false,
|
33 |
+
"scale_attn_weights": true,
|
34 |
+
"state_dim": 17,
|
35 |
+
"state_mean": [
|
36 |
+
-0.04489211615979194,
|
37 |
+
0.03232611992451431,
|
38 |
+
0.06034820791204714,
|
39 |
+
-0.1708161758114941,
|
40 |
+
-0.1947702336464969,
|
41 |
+
-0.057516805782271266,
|
42 |
+
0.09701419538351468,
|
43 |
+
0.032391781978295245,
|
44 |
+
11.047338665669947,
|
45 |
+
-0.07997213109949118,
|
46 |
+
-0.3236324481697188,
|
47 |
+
0.3629688906533557,
|
48 |
+
0.4232352315326554,
|
49 |
+
0.4083653650984332,
|
50 |
+
1.10850102882522,
|
51 |
+
-0.48743752095133835,
|
52 |
+
-0.07375080715086205
|
53 |
+
],
|
54 |
+
"state_std": [
|
55 |
+
0.04003438547801023,
|
56 |
+
0.41146983308045787,
|
57 |
+
0.542171076653844,
|
58 |
+
0.4154335823272121,
|
59 |
+
0.23797831126847824,
|
60 |
+
0.6205308852047869,
|
61 |
+
0.301055996310825,
|
62 |
+
0.21742004892182604,
|
63 |
+
2.2114253628640745,
|
64 |
+
0.5726970124299755,
|
65 |
+
1.7259367582375336,
|
66 |
+
11.845533609872767,
|
67 |
+
12.067533717608823,
|
68 |
+
7.052666254859624,
|
69 |
+
13.50640651853262,
|
70 |
+
7.197609616290412,
|
71 |
+
5.027523940529379
|
72 |
+
],
|
73 |
+
"torch_dtype": "float32",
|
74 |
+
"transformers_version": "4.29.2",
|
75 |
+
"use_cache": true,
|
76 |
+
"vocab_size": 1,
|
77 |
+
"warmup_steps": 1000
|
78 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a97e88c4bab189926baf23c9af2d99b33d4ded4113e74085edd969dd0e6cd9a1
|
3 |
+
size 6609064
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cfbf35addce7e86e407e5168a1f2f39507feecda2a30f61b68ad0f47c949235f
|
3 |
+
size 4091
|