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": 1000,
+  "max_ep_return": 2055,
+  "max_obs_len": 1000,
+  "max_obs_return": 2055,
+  "min_ep_return": -136.11629783300003,
+  "min_obs_return": -136.11629783300003,
+  "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": [
+    -0.11105085755165309,
+    0.03261898734092991,
+    0.09503744437931684,
+    -0.04665634126127169,
+    -0.1422736866683895,
+    0.11284505177282138,
+    -0.14719247154011844,
+    -0.20084285335399846,
+    1.6364819749735804,
+    -0.023428430703471302,
+    -0.011453850687947402,
+    0.028144657537117004,
+    -0.03049390117503992,
+    0.04205502023233107,
+    -0.047717887152182,
+    0.036133203776659344,
+    0.004414099071640206
+  ],
+  "state_std": [
+    0.09256462065202084,
+    0.25102800281725657,
+    0.39216763873921784,
+    0.3014480322749965,
+    0.3594363079388843,
+    0.49460765306751897,
+    0.3719440989195892,
+    0.3381944556471006,
+    0.8053914288925733,
+    0.8879606428983204,
+    1.4583408884879427,
+    5.654339376346348,
+    4.917451449281009,
+    5.191545845751596,
+    5.222851276565701,
+    6.242544228826246,
+    5.14957441497535
+  ],
+  "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:a8681c2eed10265064afb0b0ff059ee5740121ccaf0f35658752d1a3ae5f929f
+size 19859182

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0df64e88ef1489bb41af150bfd4052b5720c4ca01939929733e4b76e9ca6eba8
+size 4728