Commit
•
ac5f56f
1
Parent(s):
434f3f5
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": 1000,
|
| 22 |
+
"max_ep_return": 1878,
|
| 23 |
+
"max_obs_len": 1000,
|
| 24 |
+
"max_obs_return": 1878,
|
| 25 |
+
"min_ep_return": -190.01259551200002,
|
| 26 |
+
"min_obs_return": -190.01259551200002,
|
| 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 |
+
-0.11599756189776794,
|
| 41 |
+
0.06560846819735384,
|
| 42 |
+
0.10920993021869002,
|
| 43 |
+
-0.1429371624743709,
|
| 44 |
+
-0.2901137395566367,
|
| 45 |
+
0.3850552577167159,
|
| 46 |
+
-0.15487860075363918,
|
| 47 |
+
0.017648110348436993,
|
| 48 |
+
1.1784440975010764,
|
| 49 |
+
-0.02345524509133069,
|
| 50 |
+
-0.0314115102416058,
|
| 51 |
+
0.026543268847608603,
|
| 52 |
+
0.010301741866859983,
|
| 53 |
+
0.01022627527729616,
|
| 54 |
+
-0.005871335910026125,
|
| 55 |
+
0.043837974735514124,
|
| 56 |
+
0.02332982818133526
|
| 57 |
+
],
|
| 58 |
+
"state_std": [
|
| 59 |
+
0.09609994075024549,
|
| 60 |
+
0.3551870219888492,
|
| 61 |
+
0.4352159618226794,
|
| 62 |
+
0.2600141710357705,
|
| 63 |
+
0.2431119131985146,
|
| 64 |
+
0.437137970893003,
|
| 65 |
+
0.40224724909726645,
|
| 66 |
+
0.37690022746069324,
|
| 67 |
+
0.7656356396229258,
|
| 68 |
+
0.7916682336864219,
|
| 69 |
+
1.5525011838693283,
|
| 70 |
+
6.497787228344542,
|
| 71 |
+
4.533413148119782,
|
| 72 |
+
3.57088860444648,
|
| 73 |
+
3.4983678330452213,
|
| 74 |
+
5.185370152218882,
|
| 75 |
+
5.452662567393045
|
| 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 sha256:3480eb82df3930ea10f7d3c110639ee559e46c8c969f185c630af8ecc6c0b4cd
|
| 3 |
+
size 19859182
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b841683a48ad7596e5d5004eada910d35c323001a3204bc10d4e60d32c6d5c30
|
| 3 |
+
size 4728
|