Include encoded SS targets for sage in standard restoration

README.md

@@ -12,7 +12,7 @@ pretty_name: PartProbe3D compact reproduction data
 # PartProbe3D reproduction data
 
 This release supports SS, generative SLAT segmentation, and 32³ SLAT shape training.
-The default transfer is **197.57 GB**, including observations, SS targets, and available shape targets.
+The default transfer is **201.74 GB**, including observations, SS targets, and available shape targets.
 Default downloads exclude optional and deprecated folders.
 
 <!-- partprobe3d-required-files:start -->
@@ -72,7 +72,7 @@ Add `--keep-archives` only when you need temporary transfer files.
 | Dataset | SS/segmentation training views | Shape training views | Monitoring views | Transfer, GB |
 | --- | ---: | ---: | ---: | ---: |
 | 3dfront | 39,930 | 39,930 | 16 | 123.01 |
-| sage | 17,456 | 17,453 | 16 | 74.56 |
+| sage | 17,456 | 17,453 | 16 | 78.73 |
 | foundationpose | 42,256 | Unavailable | 16 | 61.76 |
 
 FoundationPose requires explicit selection and supports SS and segmentation.
@@ -204,6 +204,7 @@ Model weights remain separate dependencies.
 | Dataset | Target views | Transfer, GB |
 | --- | ---: | ---: |
 | 3dfront | 49,996 | 31.249 |
+| sage | 19,408 | 4.175 |
 
 Use the current project checkout and download the selected datasets:
 

training/index.json

@@ -37,9 +37,10 @@
     "sage": {
       "required": [
         "training/observations/sage",
-        "training/shape/32/sage"
+        "training/shape/32/sage",
+        "training/ss/sage"
       ],
-      "bytes": 74558238720,
+      "bytes": 78733731840,
       "source_manifest_sha256": "384f03647a4a54de36edb1465b1142604d34eb3e8f757e48e0ccbf5c936a7f9b",
       "resolution": 64,
       "source_views": 19408,
@@ -50,7 +51,19 @@
         "sage10k/23adeb2d2638a873e5d9",
         "sage10k/7b2d68afd83bf89364ea",
         "sage10k/ff0a565cad070014b1b8"
-      ]
+      ],
+      "ss_targets": {
+        "component": "training/ss/sage",
+        "bytes": 4175493120,
+        "views": 19408,
+        "latent_shape": [
+          8,
+          16,
+          16,
+          16
+        ],
+        "target_scope": "masked_objects_complete_occupancy"
+      }
     },
     "foundationpose": {
       "required": [