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": [
+    "AdversarialDT"
+  ],
+  "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": 1.0,
+  "lambda2": 1.0,
+  "layer_norm_epsilon": 1e-05,
+  "log_interval_steps": 100,
+  "max_ep_len": 999,
+  "max_ep_return": 1289,
+  "max_obs_len": 999,
+  "max_obs_return": 1289,
+  "min_ep_return": -96.82057464200004,
+  "min_obs_return": -96.82057464200004,
+  "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": [
+    1.1575770125674947,
+    0.16765135255888688,
+    -0.10125198559126013,
+    -0.048161959123527304,
+    0.5866500870318963,
+    -0.08899718695545204,
+    -0.061814936780544304,
+    0.4317695335151212,
+    0.6883041465189006,
+    -0.16324784546275664,
+    0.19899619193857912,
+    -0.20914407994362408,
+    -0.13819401972845488,
+    -0.16121668431453506,
+    -0.18814624439028538,
+    -0.22432487455365363,
+    -0.4176328378528237
+  ],
+  "state_std": [
+    0.07913902740370016,
+    0.31899006345311365,
+    0.2802079737062644,
+    0.21292672390975065,
+    0.4762548011641742,
+    0.255352584176118,
+    0.28210621331296115,
+    0.6152170140751688,
+    0.845679173234231,
+    0.45206512647207403,
+    1.7121676393489746,
+    1.8135847754595518,
+    2.064758437346931,
+    2.6553273730109574,
+    1.7935814824240828,
+    1.9478784717081086,
+    2.8018697185903454
+  ],
+  "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:bac73415f0744738825f554a680c3516af3eddea52f38b799b3a417888cfa5a7
+size 6624829

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:925b98d4689e3d694ab7bb02b12c8a6db977e6b9a0b1d5c46bfe1074fe35ad2f
+size 4728