hierarchical-decision-transformer

README.md

@@ -38,7 +38,7 @@ The following hyperparameters were used during training:
 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 200
+- num_epochs: 100
 
 ### Training results
 

config.json

@@ -1,5 +1,5 @@
 {
-  "act_dim": 6,
+  "act_dim": 3,
   "action_tanh": true,
   "activation_function": "relu",
   "architectures": [
@@ -24,7 +24,7 @@
   "resid_pdrop": 0.1,
   "scale_attn_by_inverse_layer_idx": false,
   "scale_attn_weights": true,
-  "state_dim": 17,
+  "state_dim": 11,
   "torch_dtype": "float32",
   "transformers_version": "4.46.2",
   "use_cache": true,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae38aa338356b360a9ee46b2beab0f65981e5ccfc4e293599827e8534d402224
-size 5049232
+oid sha256:19cfbea8ea33df6b0617a21144ffca3eaa1feb950d53c58f0ea1651cb432e17c
+size 5036908

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6ba6d84c457b26cd8e66709749e8469938848f5e0a06a33b7a7bb28a89b48192
+oid sha256:66fce4f66a8b2f180f8e95d5c632051c2332fd526f2cd268dac8ceec5b411f52
 size 5240