Add puzzle and mesh index configurations

.gitattributes

@@ -1,3 +1,3 @@
 *.ply filter=lfs diff=lfs merge=lfs -text
 *.json filter=lfs diff=lfs merge=lfs -text
-*.parquet -filter -diff -merge
+*.parquet filter=lfs diff=lfs merge=lfs -text

README.md

@@ -16,6 +16,10 @@ configs:
     path: data/validation-*.parquet
   - split: test
     path: data/test-*.parquet
+- config_name: meshes
+  data_files:
+  - split: train
+    path: data/meshes-*.parquet
 ---
 
 # CloudAVR Point-Cloud Matrix Reasoning
@@ -60,6 +64,10 @@ CloudAVR does not contain one single mesh per puzzle: the actual assets are the
 `candidate_paths`. All six rule names are encoded in `id`, `sample_path`, and
 the asset paths.
 
+The optional `meshes` configuration is a conventional one-row-per-PLY view.
+Its `mesh_path` column contains all 360,000 concrete PLY paths across all six
+rules; use it with `load_dataset("ShowRison/CloudAVR", "meshes")`.
+
 ## Loading the annotation index
 
 ```python

data/meshes-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d3339afb0192cd4e11c3db8bb50caee986af90e8c44c56c982349beecaf7d3f
+size 1766414

dataset_summary.json

@@ -1,64 +1,3 @@
-{
-  "schema_version": "cloudavr-hub-index-v1",
-  "source": "final5000easy",
-  "seed": 42,
-  "rules": [
-    "Conservation",
-    "Count",
-    "Distribute-three",
-    "Permutation",
-    "Progression",
-    "Symmetry"
-  ],
-  "total_examples": 30000,
-  "examples_per_rule": {
-    "Conservation": 5000,
-    "Count": 5000,
-    "Distribute-three": 5000,
-    "Permutation": 5000,
-    "Progression": 5000,
-    "Symmetry": 5000
-  },
-  "examples_per_split": {
-    "train": 24000,
-    "validation": 3000,
-    "test": 3000
-  },
-  "examples_per_split_and_rule": {
-    "train/Conservation": 4000,
-    "train/Count": 4000,
-    "train/Distribute-three": 4000,
-    "train/Permutation": 4000,
-    "train/Progression": 4000,
-    "train/Symmetry": 4000,
-    "validation/Conservation": 500,
-    "validation/Count": 500,
-    "validation/Distribute-three": 500,
-    "validation/Permutation": 500,
-    "validation/Progression": 500,
-    "validation/Symmetry": 500,
-    "test/Conservation": 500,
-    "test/Count": 500,
-    "test/Distribute-three": 500,
-    "test/Permutation": 500,
-    "test/Progression": 500,
-    "test/Symmetry": 500
-  },
-  "assets_per_example": 12,
-  "asset_validation": "not_run",
-  "asset_counts_checked": {},
-  "parquet": {
-    "train": {
-      "rows": 24000,
-      "bytes": 2160616
-    },
-    "validation": {
-      "rows": 3000,
-      "bytes": 291556
-    },
-    "test": {
-      "rows": 3000,
-      "bytes": 290170
-    }
-  }
-}
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8ff555de0eafea6fd1f3550165c263bdfe162f78a10a95910123e64dfb87b2a
+size 1630