Commit
·
442a4aa
1
Parent(s):
e6652e3
Training in progress, step 3334
Browse files- .gitignore +1 -0
- config.json +84 -0
- pytorch_model.bin +3 -0
- training_args.bin +3 -0
.gitignore
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
checkpoint-*/
|
config.json
ADDED
|
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"act_dim": 6,
|
| 3 |
+
"action_tanh": true,
|
| 4 |
+
"activation_function": "relu",
|
| 5 |
+
"adv_act_dim": 6,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"MultipartARDT"
|
| 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": 0,
|
| 15 |
+
"hidden_size": 128,
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"lambda1": 0.1,
|
| 18 |
+
"lambda2": 1.0,
|
| 19 |
+
"layer_norm_epsilon": 1e-05,
|
| 20 |
+
"log_interval_steps": 100,
|
| 21 |
+
"max_ep_len": 1000,
|
| 22 |
+
"max_ep_return": 2054.7955639899983,
|
| 23 |
+
"max_obs_len": 1000,
|
| 24 |
+
"max_obs_return": 2054.7955639899983,
|
| 25 |
+
"min_ep_return": -136.11629783300003,
|
| 26 |
+
"min_obs_return": -136.11629783300003,
|
| 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 |
+
"rtg_shift": 136.11639783300004,
|
| 37 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 38 |
+
"scale_attn_weights": true,
|
| 39 |
+
"state_dim": 17,
|
| 40 |
+
"state_mean": [
|
| 41 |
+
-0.11105085755165309,
|
| 42 |
+
0.03261898734092991,
|
| 43 |
+
0.09503744437931684,
|
| 44 |
+
-0.04665634126127169,
|
| 45 |
+
-0.1422736866683895,
|
| 46 |
+
0.11284505177282138,
|
| 47 |
+
-0.14719247154011844,
|
| 48 |
+
-0.20084285335399846,
|
| 49 |
+
1.6364819749735804,
|
| 50 |
+
-0.023428430703471302,
|
| 51 |
+
-0.011453850687947402,
|
| 52 |
+
0.028144657537117004,
|
| 53 |
+
-0.03049390117503992,
|
| 54 |
+
0.04205502023233107,
|
| 55 |
+
-0.047717887152182,
|
| 56 |
+
0.036133203776659344,
|
| 57 |
+
0.004414099071640206
|
| 58 |
+
],
|
| 59 |
+
"state_std": [
|
| 60 |
+
0.09256462065202084,
|
| 61 |
+
0.25102800281725657,
|
| 62 |
+
0.39216763873921784,
|
| 63 |
+
0.3014480322749965,
|
| 64 |
+
0.3594363079388843,
|
| 65 |
+
0.49460765306751897,
|
| 66 |
+
0.3719440989195892,
|
| 67 |
+
0.3381944556471006,
|
| 68 |
+
0.8053914288925733,
|
| 69 |
+
0.8879606428983204,
|
| 70 |
+
1.4583408884879427,
|
| 71 |
+
5.654339376346348,
|
| 72 |
+
4.917451449281009,
|
| 73 |
+
5.191545845751596,
|
| 74 |
+
5.222851276565701,
|
| 75 |
+
6.242544228826246,
|
| 76 |
+
5.14957441497535
|
| 77 |
+
],
|
| 78 |
+
"torch_dtype": "float32",
|
| 79 |
+
"total_train_steps": 10000,
|
| 80 |
+
"transformers_version": "4.31.0",
|
| 81 |
+
"use_cache": true,
|
| 82 |
+
"vocab_size": 1,
|
| 83 |
+
"warmup_steps": 1000
|
| 84 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4d7e21924109ebc42ef6e55beba1761b986e37cf813fe5d7c3182f4ff131694f
|
| 3 |
+
size 10398657
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4f4e337117cb4a1203fffbf08d17b408f94210a812a8b9227ac50f8ba94983a
|
| 3 |
+
size 4283
|