Training in progress, step 2000

.gitignore

@@ -0,0 +1 @@
+checkpoint-*/

config.json

@@ -0,0 +1,83 @@
+{
+  "act_dim": 6,
+  "action_tanh": true,
+  "activation_function": "relu",
+  "adv_act_dim": 6,
+  "architectures": [
+    "MultipartADT"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "context_size": 20,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "flag": false,
+  "hidden_size": 128,
+  "initializer_range": 0.02,
+  "lambda1": 0.7749999999999693,
+  "lambda2": 250.0,
+  "layer_norm_epsilon": 1e-05,
+  "log_interval_steps": 100,
+  "max_ep_len": 999,
+  "max_ep_return": 2055,
+  "max_obs_len": 999,
+  "max_obs_return": 2055,
+  "min_ep_return": -31.996896582000012,
+  "min_obs_return": -31.996896582000012,
+  "model_type": "decision_transformer",
+  "n_head": 1,
+  "n_inner": null,
+  "n_layer": 3,
+  "n_positions": 1024,
+  "pr_act_dim": 6,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "returns_scale": 1000,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "state_dim": 17,
+  "state_mean": [
+    -0.11173252361577966,
+    0.06332421416752207,
+    0.10642913798176151,
+    0.016696281865103354,
+    0.01292316632921915,
+    0.0252802808143255,
+    -0.163862748115139,
+    -0.2785617364192604,
+    1.800042231687943,
+    -0.023767931373515298,
+    0.0033658001950578565,
+    0.016833230392745077,
+    -0.05654895540758762,
+    0.057139100560966265,
+    -0.0686476457401574,
+    0.04198423127298879,
+    0.005969089117189445
+  ],
+  "state_std": [
+    0.0985398737294268,
+    0.19583829223764262,
+    0.3315948858702887,
+    0.32217133624379024,
+    0.39319405952592607,
+    0.4463002698189581,
+    0.3517920731882343,
+    0.2777303946401783,
+    0.7745754765732835,
+    0.9800573190633699,
+    1.365188942289809,
+    4.87113384669823,
+    5.058535039406536,
+    6.395587223733579,
+    5.947895650994883,
+    6.4933930314513075,
+    5.279823051159204
+  ],
+  "torch_dtype": "float32",
+  "total_train_steps": 10000,
+  "transformers_version": "4.29.2",
+  "use_cache": true,
+  "vocab_size": 1,
+  "warmup_steps": 1000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b35c71b87bf49b3cf3bfc376367d762900624b878babe10603242aa61839532c
+size 19857646

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d794518eeff1443ccd2a7df5ae56735908930556a9c76eba20be7f6808717c0
+size 4664