Initial upload of synthetic OKVQA dataset with key objects

README.md

@@ -476,10 +476,10 @@ dataset_info:
     dtype: string
   splits:
   - name: train
-    num_bytes: 57110.0
-    num_examples: 2
-  download_size: 32836
-  dataset_size: 57110.0
+    num_bytes: 2345651.0
+    num_examples: 50
+  download_size: 848772
+  dataset_size: 2345651.0
 - config_name: from300_400
   features:
   - name: idx

data/from2200_2300/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db874126579c95308e6940ca086e4c80ecdc105c2ade2c3e5b8403895dfca9d7
+size 848772