Commit
•
d2200f2
1
Parent(s):
5c69757
Training in progress, step 200
Browse files- .gitignore +1 -0
- config.json +80 -0
- pytorch_model.bin +3 -0
- training_args.bin +3 -0
.gitignore
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
checkpoint-*/
|
config.json
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 1.0,
|
| 17 |
+
"lambda2": 10.0,
|
| 18 |
+
"layer_norm_epsilon": 1e-05,
|
| 19 |
+
"log_interval_steps": 100,
|
| 20 |
+
"max_ep_len": 1001,
|
| 21 |
+
"max_ep_return": 2000.0,
|
| 22 |
+
"max_obs_len": 1001,
|
| 23 |
+
"max_obs_return": 1666.0162529458535,
|
| 24 |
+
"min_obs_return": -327.4452760369768,
|
| 25 |
+
"model_type": "decision_transformer",
|
| 26 |
+
"n_head": 1,
|
| 27 |
+
"n_inner": null,
|
| 28 |
+
"n_layer": 3,
|
| 29 |
+
"n_positions": 1024,
|
| 30 |
+
"pr_act_dim": 6,
|
| 31 |
+
"reorder_and_upcast_attn": false,
|
| 32 |
+
"resid_pdrop": 0.1,
|
| 33 |
+
"returns_scale": 1000,
|
| 34 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 35 |
+
"scale_attn_weights": true,
|
| 36 |
+
"state_dim": 17,
|
| 37 |
+
"state_mean": [
|
| 38 |
+
-0.47949490638902126,
|
| 39 |
+
2.9392855446868635,
|
| 40 |
+
0.2599355675435394,
|
| 41 |
+
-0.13339093314086728,
|
| 42 |
+
0.1662935979860262,
|
| 43 |
+
-0.07257338276852067,
|
| 44 |
+
0.017435688300063465,
|
| 45 |
+
0.003208991217454027,
|
| 46 |
+
1.7938188390218883,
|
| 47 |
+
0.030872478249036942,
|
| 48 |
+
0.07502310908913958,
|
| 49 |
+
0.324547227176298,
|
| 50 |
+
-0.27233690514262826,
|
| 51 |
+
-0.04487928956840291,
|
| 52 |
+
-0.16896564328141353,
|
| 53 |
+
0.0648626864260328,
|
| 54 |
+
0.12811998530688132
|
| 55 |
+
],
|
| 56 |
+
"state_std": [
|
| 57 |
+
0.130504818046764,
|
| 58 |
+
1.52517394990146,
|
| 59 |
+
0.5243136462888118,
|
| 60 |
+
0.4580009500375659,
|
| 61 |
+
0.3207804699740532,
|
| 62 |
+
0.6207726011744169,
|
| 63 |
+
0.37194518983551855,
|
| 64 |
+
0.2779030180591103,
|
| 65 |
+
1.1986000369857561,
|
| 66 |
+
0.8974155842189676,
|
| 67 |
+
1.6281378735321663,
|
| 68 |
+
8.853020115717245,
|
| 69 |
+
11.611960972109499,
|
| 70 |
+
8.194997861866108,
|
| 71 |
+
11.215696331272747,
|
| 72 |
+
10.835559173586182,
|
| 73 |
+
6.144604000512429
|
| 74 |
+
],
|
| 75 |
+
"torch_dtype": "float32",
|
| 76 |
+
"transformers_version": "4.31.0",
|
| 77 |
+
"use_cache": true,
|
| 78 |
+
"vocab_size": 1,
|
| 79 |
+
"warmup_steps": 1000
|
| 80 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52bdb95d8c2d2172649c8d4544341150d10261d2ea6cfe91122926e330368ef2
|
| 3 |
+
size 3461910
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d1368aac1fc1ba948a598c91e133a71a9f8c8fc2a2ed21fe32d471b83c1d1318
|
| 3 |
+
size 4283
|