Shuffle image groups within native splits with seed 0; preserve all examples

README.md

@@ -24,7 +24,7 @@ dataset_info:
   - name: train
     num_bytes: 2286302065
     num_examples: 49820
-  download_size: 2542979726
+  download_size: 2543457015
   dataset_size: 2542962524
 configs:
 - config_name: default
@@ -47,9 +47,10 @@ questions are omitted; no distractors are generated. Options retain source
 order, equivalent duplicates collapse, and gold indices are remapped.
 Embedded option blocks are removed before Jev permutation. Screenshots are
 grouped across QAs and test images take precedence over training copies.
+Image groups are shuffled within each native split with seed 0; QA order is retained.
 Procedural game labels are used; generated rationales are excluded from inputs.
 Original source declares MIT. No validation split is fabricated.
 
-Original data: [OpenMOSS-Team/GameQA-140K](https://huggingface.co/datasets/OpenMOSS-Team/GameQA-140K). Repackaged as parquet for [tasksource](https://github.com/sileod/tasksource) by [scripts/repackage_dataset/games.py](https://github.com/sileod/tasksource/blob/main/scripts/repackage_dataset/games.py#L105).
+Original data: [OpenMOSS-Team/GameQA-140K](https://huggingface.co/datasets/OpenMOSS-Team/GameQA-140K). Repackaged as parquet for [tasksource](https://github.com/sileod/tasksource) by [scripts/repackage_dataset/games.py](https://github.com/sileod/tasksource/blob/main/scripts/repackage_dataset/games.py#L110).
 
 [Conversion provenance](provenance.json): source revisions, hashes, configuration and exclusions.

data/test-00000-of-00002.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:60c81c9ec37cabb22c740f7181f5a3b08266dcd892e7975f93ae876977248afd
-size 121981704
+oid sha256:db4300fb2c277b2974aa5c80a42415c316a1d475460d5bde9200d7d2e3f5fb01
+size 127088934

data/test-00001-of-00002.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c6e01f9f98bf4aafab3ab0b477cedf16ee840fc9f9cd7697e12ff93b2037b53
-size 134650369
+oid sha256:b7cdba6ee23857c11b47ea974ef3aa658caeec280f2c4e470206206b365a257e
+size 129586825

data/train-00000-of-00009.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cf466bca7c8975161b211a14680d5bedc229bd2a2b6453e55f330e038656b56d
-size 239038391
+oid sha256:d75c27c71f709af424a5fb42f033466e6f176f0e34e0d638c9a7d50d65dc9007
+size 251611323

data/train-00001-of-00009.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e4dd08263262d8d1df10a8a4d1326690df93c6f4ba65b701fddd4241170ae53
-size 51118560
+oid sha256:ac0ae5c99752403d3719ae02013f81bfea135b345e5519f48dafec71f0e0a105
+size 253649853

data/train-00002-of-00009.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:73e921609d1bd7efb0e606c65ae942e3c6557f4d25469004519f22ff0b0d8908
-size 338077683
+oid sha256:f0fd8acfc890af54c4ad3fb3636d37c81e48dd85ac0f898d705d92f135b4d3a1
+size 253893553

data/train-00003-of-00009.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb767cc16d59acda5c903c223982a35503718f2e82d07733083b66b5e3aa4299
-size 111927137
+oid sha256:15319878e12f0d3c0e6d4ba5206e060a33c3cf1939cf3ac98617d4c8e97fc605
+size 254132214

data/train-00004-of-00009.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:37047d249f1a584479ab91b5ffdbb26f075054cb7679b0ca58ab137de2f9aec3
-size 186361704
+oid sha256:e6d6df7997e01376f362c503c4e3cd8f389c1b5e4eea4e9e9903d08afeac5e60
+size 253046734

data/train-00005-of-00009.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5f9beed6b28e93d2311f8c1bd62148f05674c87352e8635979b9dc08828e6d56
-size 311724675
+oid sha256:77dc99d5f188256757579715b50d54738ebfb8c83f48d21d2a3d3e604b45af9c
+size 257561785

data/train-00006-of-00009.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c30ce349cdb84e89c1774be96d22ab130368d3aa4c542a465440b7c247ac9350
-size 293029541
+oid sha256:65819d64fb06c5b95e9f0e39b785837351a672ee45e212c5dc62c7fcad9307e7
+size 250508286

data/train-00007-of-00009.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:670f0a1b7d6f060eee59d1d18d81387664f9a5df2508cf7043ce0955455653b5
-size 366449205
+oid sha256:b3fc7bbbe80155130b83f0ea882a21ae0f316f7a9176dded387058a2118dd262
+size 255531255

data/train-00008-of-00009.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d9774b3aa443834db780c2b584cb5495e810f0f89a3a78aae6028c60f14ea9e2
-size 388620757
+oid sha256:369fd24412325b1f5e79b3b0b5bf8d3af69a79e6d754211529abfd20d892b940
+size 256846253

provenance.json

@@ -12,7 +12,7 @@
       "questions": 80379
     }
   },
-  "conversion_sha256": "c71452577e9a5f5e8a1b4e6a448f8412cb1c4a2cb62af44b0600c5187c2dc441",
+  "conversion_sha256": "9f0edc15ccd6aae253686be09e32542cca5f5c7e0e7f86e8ecfd5d67e5057b3d",
   "excluded_rows": 50935,
   "license": "mit",
   "license_url": "https://huggingface.co/datasets/OpenMOSS-Team/GameQA-140K",
@@ -29,5 +29,11 @@
     "misaligned_source_options": 668,
     "heldout_image_or_layout": 4818
   },
-  "annotation_provenance": "procedural game/solver labels; source rationales and hidden states are never model inputs"
+  "annotation_provenance": "procedural game/solver labels; source rationales and hidden states are never model inputs",
+  "ordering": {
+    "method": "datasets.shuffle",
+    "seed": 0,
+    "unit": "image_group",
+    "within": "native_split"
+  }
 }