Add Croissant JSON-LD metadata with Tabular modality

dataset-metadata.json

@@ -24,6 +24,7 @@
     "name": "Deepcell Inc.",
     "url": "https://huggingface.co/Deepcell"
   },
+  "modality": ["Tabular"],                ←─ change this from “Image” to “Tabular”
   "distribution": [
     {
       "@type": "cr:FileObject",
@@ -38,7 +39,11 @@
       "@id": "parquet-files",
       "containedIn": { "@id": "repo" },
       "encodingFormat": "application/x-parquet",
-      "includes": "*.parquet"
+      "includes": [
+        "embeddings_pcs_shape.parquet",
+        "embeddings_pcs_texture.parquet",
+        "embeddings_pcs_width.parquet"
+      ]
     }
   ],
   "recordSet": [
@@ -46,23 +51,15 @@
       "@type": "RecordSet",
       "@id": "default_splits",
       "name": "Dataset splits",
-      "description": "Available splits for the default configuration.",
       "field": [
-        {
-          "@type": "Field",
-          "@id": "default_splits/split_name",
-          "dataType": "Text"
-        }
+        { "@type": "Field", "@id": "default_splits/split_name", "dataType": "Text" }
       ],
-      "data": [
-        { "default_splits/split_name": "train" }
-      ]
+      "data": [{ "default_splits/split_name": "train" }]
     },
     {
       "@type": "RecordSet",
       "@id": "default",
-      "name": "Default configuration records",
-      "description": "All rows in the 'train' split.",
+      "name": "Default config records",
       "field": [
         {
           "@type": "Field",
@@ -70,29 +67,20 @@
           "dataType": "Text",
           "source": {
             "fileSet": { "@id": "parquet-files" },
-            "extract": {
-              "fileProperty": "fullpath",
-              "regex": ".*train.*\\.parquet$"
-            }
+            "extract": { "fileProperty": "fullpath", "regex": ".*\\.parquet$" }
           }
         },
         {
           "@type": "Field",
           "@id": "default/cluster",
           "dataType": "cr:Int64",
-          "source": {
-            "fileSet": { "@id": "parquet-files" },
-            "extract": { "column": "cluster" }
-          }
+          "source": { "fileSet": { "@id": "parquet-files" }, "extract": { "column": "cluster" } }
         },
         {
           "@type": "Field",
           "@id": "default/number",
           "dataType": "cr:Int64",
-          "source": {
-            "fileSet": { "@id": "parquet-files" },
-            "extract": { "column": "number" }
-          }
+          "source": { "fileSet": { "@id": "parquet-files" }, "extract": { "column": "number" } }
         },
         {
           "@type": "Field",
@@ -100,10 +88,7 @@
           "dataType": "cr:Float32",
           "isArray": true,
           "arrayShape": "-1",
-          "source": {
-            "fileSet": { "@id": "parquet-files" },
-            "extract": { "column": "hfm" }
-          }
+          "source": { "fileSet": { "@id": "parquet-files" }, "extract": { "column": "hfm" } }
         },
         {
           "@type": "Field",
@@ -111,10 +96,7 @@
           "dataType": "cr:Float32",
           "isArray": true,
           "arrayShape": "-1",
-          "source": {
-            "fileSet": { "@id": "parquet-files" },
-            "extract": { "column": "imagenet" }
-          }
+          "source": { "fileSet": { "@id": "parquet-files" }, "extract": { "column": "imagenet" } }
         },
         {
           "@type": "Field",
@@ -122,10 +104,7 @@
           "dataType": "cr:Float64",
           "isArray": true,
           "arrayShape": "-1",
-          "source": {
-            "fileSet": { "@id": "parquet-files" },
-            "extract": { "column": "cp" }
-          }
+          "source": { "fileSet": { "@id": "parquet-files" }, "extract": { "column": "cp" } }
         }
       ]
     }