Upload folder using huggingface_hub

config.json

@@ -1,83 +1,83 @@
 {
-    "type": "pi05",
-    "n_obs_steps": 1,
-    "input_features": {
-        "observation.state": {
-            "type": "STATE",
-            "shape": [
-                6
-            ]
-        },
-        "observation.images.gripper": {
-            "type": "VISUAL",
-            "shape": [
-                3,
-                360,
-                640
-            ]
-        },
-        "observation.images.scene": {
-            "type": "VISUAL",
-            "shape": [
-                3,
-                480,
-                640
-            ]
-        }
+  "type": "pi05",
+  "n_obs_steps": 1,
+  "input_features": {
+    "observation/state": {
+      "type": "STATE",
+      "shape": [
+        6
+      ]
     },
-    "output_features": {
-        "action": {
-            "type": "ACTION",
-            "shape": [
-                6
-            ]
-        }
+    "observation/wrist_image": {
+      "type": "VISUAL",
+      "shape": [
+        3,
+        360,
+        640
+      ]
     },
-    "device": "cuda",
-    "use_amp": false,
-    "push_to_hub": true,
-    "repo_id": "s3y/pi05-10-30",
-    "private": null,
-    "tags": null,
-    "license": null,
-    "pretrained_path": "/root/pi05_libero_finetuned",
-    "paligemma_variant": "gemma_2b",
-    "action_expert_variant": "gemma_300m",
-    "dtype": "bfloat16",
-    "chunk_size": 50,
-    "n_action_steps": 50,
-    "max_state_dim": 32,
-    "max_action_dim": 32,
-    "num_inference_steps": 10,
-    "time_sampling_beta_alpha": 1.5,
-    "time_sampling_beta_beta": 1.0,
-    "time_sampling_scale": 0.999,
-    "time_sampling_offset": 0.001,
-    "min_period": 0.004,
-    "max_period": 4.0,
-    "image_resolution": [
-        224,
-        224
-    ],
-    "empty_cameras": 0,
-    "tokenizer_max_length": 200,
-    "normalization_mapping": {
-        "VISUAL": "IDENTITY",
-        "STATE": "QUANTILES",
-        "ACTION": "QUANTILES"
-    },
-    "gradient_checkpointing": true,
-    "compile_model": true,
-    "compile_mode": "max-autotune",
-    "optimizer_lr": 2.5e-05,
-    "optimizer_betas": [
-        0.9,
-        0.95
-    ],
-    "optimizer_eps": 1e-08,
-    "optimizer_weight_decay": 0.01,
-    "optimizer_grad_clip_norm": 1.0,
-    "scheduler_warmup_steps": 1000,
-    "scheduler_decay_steps": 30000,
-    "scheduler_decay_lr": 2.5e-06
+    "observation/image": {
+      "type": "VISUAL",
+      "shape": [
+        3,
+        480,
+        640
+      ]
+    }
+  },
+  "output_features": {
+    "actions": {
+      "type": "ACTION",
+      "shape": [
+        6
+      ]
+    }
+  },
+  "device": "cuda",
+  "use_amp": false,
+  "push_to_hub": true,
+  "repo_id": "s3y/pi05-10-30",
+  "private": null,
+  "tags": null,
+  "license": null,
+  "pretrained_path": "/root/pi05_libero_finetuned",
+  "paligemma_variant": "gemma_2b",
+  "action_expert_variant": "gemma_300m",
+  "dtype": "bfloat16",
+  "chunk_size": 50,
+  "n_action_steps": 50,
+  "max_state_dim": 32,
+  "max_action_dim": 32,
+  "num_inference_steps": 10,
+  "time_sampling_beta_alpha": 1.5,
+  "time_sampling_beta_beta": 1.0,
+  "time_sampling_scale": 0.999,
+  "time_sampling_offset": 0.001,
+  "min_period": 0.004,
+  "max_period": 4.0,
+  "image_resolution": [
+    224,
+    224
+  ],
+  "empty_cameras": 0,
+  "tokenizer_max_length": 200,
+  "normalization_mapping": {
+    "VISUAL": "IDENTITY",
+    "STATE": "QUANTILES",
+    "ACTION": "QUANTILES"
+  },
+  "gradient_checkpointing": true,
+  "compile_model": true,
+  "compile_mode": "max-autotune",
+  "optimizer_lr": 2.5e-05,
+  "optimizer_betas": [
+    0.9,
+    0.95
+  ],
+  "optimizer_eps": 1e-08,
+  "optimizer_weight_decay": 0.01,
+  "optimizer_grad_clip_norm": 1.0,
+  "scheduler_warmup_steps": 1000,
+  "scheduler_decay_steps": 30000,
+  "scheduler_decay_lr": 2.5e-06
 }

policy_preprocessor.json

@@ -16,13 +16,13 @@
       "config": {
         "eps": 1e-08,
         "features": {
-          "observation.state": {
+          "observation/state": {
             "type": "STATE",
             "shape": [
               6
             ]
           },
-          "observation.images.gripper": {
+          "observation/wrist_image": {
             "type": "VISUAL",
             "shape": [
               3,
@@ -30,7 +30,7 @@
               640
             ]
           },
-          "observation.images.scene": {
+          "observation/image": {
             "type": "VISUAL",
             "shape": [
               3,