Commit
•
d6212e1
1
Parent(s):
e9348bc
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": 4227,
|
| 23 |
+
"max_obs_len": 1000,
|
| 24 |
+
"max_obs_return": 4227,
|
| 25 |
+
"min_ep_return": -40.860974691808224,
|
| 26 |
+
"min_obs_return": -40.860974691808224,
|
| 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 |
+
1.218962501630123,
|
| 41 |
+
0.14163528947484982,
|
| 42 |
+
-0.037049690418079276,
|
| 43 |
+
-0.13816565862287417,
|
| 44 |
+
0.5150924442392095,
|
| 45 |
+
-0.047190008896185874,
|
| 46 |
+
-0.4728192310128083,
|
| 47 |
+
0.04225571020440272,
|
| 48 |
+
2.394827945466415,
|
| 49 |
+
-0.03143244895598562,
|
| 50 |
+
0.04466346992652861,
|
| 51 |
+
-0.023907474073160472,
|
| 52 |
+
-0.10133977389419448,
|
| 53 |
+
0.09091033149985288,
|
| 54 |
+
-0.004192652549242934,
|
| 55 |
+
-0.12120530698001407,
|
| 56 |
+
-0.5496955076793073
|
| 57 |
+
],
|
| 58 |
+
"state_std": [
|
| 59 |
+
0.12314899816939005,
|
| 60 |
+
0.3243013276243093,
|
| 61 |
+
0.11457279298201731,
|
| 62 |
+
0.26230205605383794,
|
| 63 |
+
0.5641955092260325,
|
| 64 |
+
0.22723931119215887,
|
| 65 |
+
0.383823465370718,
|
| 66 |
+
0.7373887431190398,
|
| 67 |
+
1.23903653745177,
|
| 68 |
+
0.798316986357,
|
| 69 |
+
1.5671374062610388,
|
| 70 |
+
1.8108464431690363,
|
| 71 |
+
3.027126524702137,
|
| 72 |
+
4.064707182931284,
|
| 73 |
+
1.4600719051825006,
|
| 74 |
+
3.7459188910671553,
|
| 75 |
+
5.584324011405262
|
| 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 sha256:6c1b953bd5dc2f53764a06f365a1c36f0edcb6d683d3044ec05f6f314cd4ea22
|
| 3 |
+
size 3461398
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e8634a6ca7086b5af1926317e21d7c91a4b557393e3e0d0375ba5bf00d04a83c
|
| 3 |
+
size 4219
|