Training in progress, step 3334

.gitignore

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

config.json

@@ -0,0 +1,84 @@
+{
+  "act_dim": 6,
+  "action_tanh": true,
+  "activation_function": "relu",
+  "adv_act_dim": 6,
+  "architectures": [
+    "MultipartARDT"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "context_size": 20,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "flag": 0,
+  "hidden_size": 128,
+  "initializer_range": 0.02,
+  "lambda1": 0.1,
+  "lambda2": 1.0,
+  "layer_norm_epsilon": 1e-05,
+  "log_interval_steps": 100,
+  "max_ep_len": 1000,
+  "max_ep_return": 1877.9450329321019,
+  "max_obs_len": 1000,
+  "max_obs_return": 1877.9450329321019,
+  "min_ep_return": -190.01259551200002,
+  "min_obs_return": -190.01259551200002,
+  "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,
+  "rtg_shift": 190.01269551200002,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "state_dim": 17,
+  "state_mean": [
+    -0.11599756189776794,
+    0.06560846819735384,
+    0.10920993021869002,
+    -0.1429371624743709,
+    -0.2901137395566367,
+    0.3850552577167159,
+    -0.15487860075363918,
+    0.017648110348436993,
+    1.1784440975010764,
+    -0.02345524509133069,
+    -0.0314115102416058,
+    0.026543268847608603,
+    0.010301741866859983,
+    0.01022627527729616,
+    -0.005871335910026125,
+    0.043837974735514124,
+    0.02332982818133526
+  ],
+  "state_std": [
+    0.09609994075024549,
+    0.3551870219888492,
+    0.4352159618226794,
+    0.2600141710357705,
+    0.2431119131985146,
+    0.437137970893003,
+    0.40224724909726645,
+    0.37690022746069324,
+    0.7656356396229258,
+    0.7916682336864219,
+    1.5525011838693283,
+    6.497787228344542,
+    4.533413148119782,
+    3.57088860444648,
+    3.4983678330452213,
+    5.185370152218882,
+    5.452662567393045
+  ],
+  "torch_dtype": "float32",
+  "total_train_steps": 10000,
+  "transformers_version": "4.31.0",
+  "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:b2ec7c3cc860a75516a24836716907db155798488ca0e104c07d28597250f433
+size 10398657

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb5c95b56533a3ebf38bc550abb6a442473ac30d757fe9c886391072ebc977a9
+size 4347