Commit
·
b17c15c
1
Parent(s):
23d0c4f
Training in progress, step 2000
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 |
+
"MultipartADT"
|
| 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": 0.7749999999999693,
|
| 18 |
+
"lambda2": 250.0,
|
| 19 |
+
"layer_norm_epsilon": 1e-05,
|
| 20 |
+
"log_interval_steps": 100,
|
| 21 |
+
"max_ep_len": 999,
|
| 22 |
+
"max_ep_return": 3741,
|
| 23 |
+
"max_obs_len": 999,
|
| 24 |
+
"max_obs_return": 3741,
|
| 25 |
+
"min_ep_return": -67.36698192600001,
|
| 26 |
+
"min_obs_return": -67.36698192600001,
|
| 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 |
+
1.1741498260053231,
|
| 41 |
+
0.1504549391892248,
|
| 42 |
+
-0.15121413627675384,
|
| 43 |
+
-0.07285067597832379,
|
| 44 |
+
0.39519318778781415,
|
| 45 |
+
-0.16927935776702405,
|
| 46 |
+
-0.036412978118961305,
|
| 47 |
+
0.37994817283499194,
|
| 48 |
+
1.1008757847362347,
|
| 49 |
+
-0.0760672642486118,
|
| 50 |
+
0.07804146275485414,
|
| 51 |
+
-0.15959546188219698,
|
| 52 |
+
-0.10067056116685251,
|
| 53 |
+
-0.366385471959961,
|
| 54 |
+
-0.09145403023169597,
|
| 55 |
+
-0.13521694087550387,
|
| 56 |
+
-0.43988015960374044
|
| 57 |
+
],
|
| 58 |
+
"state_std": [
|
| 59 |
+
0.0770901694018985,
|
| 60 |
+
0.3221843631575875,
|
| 61 |
+
0.32393220476272716,
|
| 62 |
+
0.2628725732908372,
|
| 63 |
+
0.6108049302372695,
|
| 64 |
+
0.29287376018786104,
|
| 65 |
+
0.25754502919037525,
|
| 66 |
+
0.6474974908473997,
|
| 67 |
+
1.1693254095664798,
|
| 68 |
+
0.521658948120785,
|
| 69 |
+
2.241845841362767,
|
| 70 |
+
2.53300963250326,
|
| 71 |
+
2.3769917279315886,
|
| 72 |
+
4.296415165084856,
|
| 73 |
+
2.5227004058659124,
|
| 74 |
+
1.9587142628721583,
|
| 75 |
+
3.7512497512493335
|
| 76 |
+
],
|
| 77 |
+
"torch_dtype": "float32",
|
| 78 |
+
"total_train_steps": 10000,
|
| 79 |
+
"transformers_version": "4.29.2",
|
| 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 sha256:3775cf6f960ebeb3d84ac73fa21c8156ee1b07c1c430f6020ec267c2683e4b32
|
| 3 |
+
size 19857646
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d80646f544f4d45467bc41ff12f9312b70642f74221f3d61a79915f0952570d
|
| 3 |
+
size 4664
|