Add corpus split (images and IDs)

README.md

@@ -0,0 +1,19 @@
+---
+dataset_info:
+  features:
+  - name: corpus_id
+    dtype: int32
+  - name: image
+    dtype: image
+  splits:
+  - name: corpus
+    num_bytes: 2442728
+    num_examples: 10
+  download_size: 2441782
+  dataset_size: 2442728
+configs:
+- config_name: default
+  data_files:
+  - split: corpus
+    path: data/corpus-*
+---

data/corpus-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a89a00166310dc51cde607c9e728eda6d4682338868e958aa3d9c7fe054e7b5
+size 2441782