Upload policy weights, train config and readme

README.md

@@ -6,8 +6,8 @@ model_name: reward_classifier
 pipeline_tag: robotics
 tags:
 - robotics
-- lerobot
 - reward_classifier
+- lerobot
 ---
 
 # Model Card for reward_classifier

config.json

@@ -44,8 +44,8 @@
     "latent_dim": 256,
     "image_embedding_pooling_dim": 8,
     "dropout_rate": 0.1,
-    "model_name": "helper2424/resnet10",
-    "model_type": "cnn",
+    "model_name": "openai/clip-vit-base-patch32",
+    "model_type": "transformer",
     "num_cameras": 1,
     "learning_rate": 0.0003,
     "weight_decay": 0.01,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eec7fe990d05e17c032c0cb250f59e1af1b3c2aab73e917b572236c0a22cffe5
-size 24358564
+oid sha256:174034fa15024461897747db635b4d17c04efbede9aa24f84226186e34763cd4
+size 350638652

train_config.json

@@ -71,14 +71,14 @@
         "latent_dim": 256,
         "image_embedding_pooling_dim": 8,
         "dropout_rate": 0.1,
-        "model_name": "helper2424/resnet10",
-        "model_type": "cnn",
+        "model_name": "openai/clip-vit-base-patch32",
+        "model_type": "transformer",
         "num_cameras": 1,
         "learning_rate": 0.0003,
         "weight_decay": 0.01,
         "grad_clip_norm": 1.0
     },
-    "output_dir": "outputs/train/2025-09-05/16-36-46_reward-classifier",
+    "output_dir": "outputs/train/2025-09-05/16-56-32_reward-classifier",
     "job_name": "reward-classifier",
     "resume": false,
     "seed": 2,
@@ -91,7 +91,7 @@
     "save_freq": 1000,
     "use_validation": true,
     "val_split": 0.05,
-    "val_freq": 10000,
+    "val_freq": 500,
     "val_batch_size": 16,
     "use_policy_training_preset": true,
     "optimizer": {
@@ -117,7 +117,7 @@
         "project": "reward-classifier",
         "entity": null,
         "notes": null,
-        "run_id": "7mcbm58f",
+        "run_id": "9scb7mxz",
         "mode": null
     },
     "use_peft": false,