Commit
•
96fce73
1
Parent(s):
58a4188
Training in progress, step 3334
Browse files- .gitignore +1 -0
- config.json +83 -0
- pytorch_model.bin +3 -0
- training_args.bin +3 -0
.gitignore
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
checkpoint-*/
|
config.json
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"act_dim": 6,
|
| 3 |
+
"action_tanh": true,
|
| 4 |
+
"activation_function": "relu",
|
| 5 |
+
"adv_act_dim": 6,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"VanillaDT"
|
| 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 |
+
"flag": false,
|
| 15 |
+
"hidden_size": 128,
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"lambda1": 1.0,
|
| 18 |
+
"lambda2": 1.0,
|
| 19 |
+
"layer_norm_epsilon": 1e-05,
|
| 20 |
+
"log_interval_steps": 100,
|
| 21 |
+
"max_ep_len": 1000,
|
| 22 |
+
"max_ep_return": 5313,
|
| 23 |
+
"max_obs_len": 1000,
|
| 24 |
+
"max_obs_return": 5313,
|
| 25 |
+
"min_ep_return": -310.2342068772123,
|
| 26 |
+
"min_obs_return": -310.2342068772123,
|
| 27 |
+
"model_type": "decision_transformer",
|
| 28 |
+
"n_head": 1,
|
| 29 |
+
"n_inner": null,
|
| 30 |
+
"n_layer": 3,
|
| 31 |
+
"n_positions": 1024,
|
| 32 |
+
"pr_act_dim": 6,
|
| 33 |
+
"reorder_and_upcast_attn": false,
|
| 34 |
+
"resid_pdrop": 0.1,
|
| 35 |
+
"returns_scale": 1000,
|
| 36 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 37 |
+
"scale_attn_weights": true,
|
| 38 |
+
"state_dim": 17,
|
| 39 |
+
"state_mean": [
|
| 40 |
+
-0.06845804518791063,
|
| 41 |
+
0.016414620763427302,
|
| 42 |
+
-0.18355838881698158,
|
| 43 |
+
-0.27624860700003606,
|
| 44 |
+
-0.3411387304685636,
|
| 45 |
+
-0.09339610047292654,
|
| 46 |
+
-0.21321190551996644,
|
| 47 |
+
-0.087743576042083,
|
| 48 |
+
5.173194196145483,
|
| 49 |
+
-0.04275185415106432,
|
| 50 |
+
-0.036108797591958,
|
| 51 |
+
0.14053658038970707,
|
| 52 |
+
0.060498910548357605,
|
| 53 |
+
0.09550849713411831,
|
| 54 |
+
0.06739012912888284,
|
| 55 |
+
0.005627354056507024,
|
| 56 |
+
0.013382561512544935
|
| 57 |
+
],
|
| 58 |
+
"state_std": [
|
| 59 |
+
0.07475442136308201,
|
| 60 |
+
0.3025072066045333,
|
| 61 |
+
0.30215810644332014,
|
| 62 |
+
0.3442407887586211,
|
| 63 |
+
0.17633869582496478,
|
| 64 |
+
0.5072853517004952,
|
| 65 |
+
0.2567592370005554,
|
| 66 |
+
0.32956261124858977,
|
| 67 |
+
1.257819020001254,
|
| 68 |
+
0.7602306988844978,
|
| 69 |
+
1.9805893025503742,
|
| 70 |
+
6.568254204917612,
|
| 71 |
+
7.467826340639571,
|
| 72 |
+
4.474474568084098,
|
| 73 |
+
10.569660245277557,
|
| 74 |
+
5.673224909433867,
|
| 75 |
+
7.500414247830719
|
| 76 |
+
],
|
| 77 |
+
"torch_dtype": "float32",
|
| 78 |
+
"total_train_steps": 10000,
|
| 79 |
+
"transformers_version": "4.31.0",
|
| 80 |
+
"use_cache": true,
|
| 81 |
+
"vocab_size": 1,
|
| 82 |
+
"warmup_steps": 1000
|
| 83 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b35c33cb4a7c15c3def53b8c1a5a1fa93022a59945908c53ec096c98ae19f9c
|
| 3 |
+
size 3461398
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a5a377023c5a8fa1b019c73b30ba1a169029b5a22d88bf4480259fce8f8073ce
|
| 3 |
+
size 4219
|