Add files using upload-large-folder tool

README.md

@@ -0,0 +1,166 @@
+---
+configs:
+- config_name: fluxreason
+  data_files:
+  - split: train
+    path: "fluxreason/train-*.parquet"
+  default: true
+- config_name: gptedit
+  data_files:
+  - split: train
+    path: "gptedit/train-*.parquet"
+- config_name: imagenet22k
+  data_files:
+  - split: train
+    path: "imagenet22k/train-*.parquet"
+- config_name: inaturalist
+  data_files:
+  - split: train
+    path: "inaturalist/train-*.parquet"
+- config_name: megalith10m
+  data_files:
+  - split: train
+    path: "megalith10m/train-*.parquet"
+- config_name: midjourneyv6
+  data_files:
+  - split: train
+    path: "midjourneyv6/train-*.parquet"
+- config_name: pexels
+  data_files:
+  - split: train
+    path: "pexels/train-*.parquet"
+- config_name: places365-challenge2016
+  data_files:
+  - split: train
+    path: "places365-challenge2016/train-*.parquet"
+- config_name: redcaps
+  data_files:
+  - split: train
+    path: "redcaps/train-*.parquet"
+- config_name: rendered_text
+  data_files:
+  - split: train
+    path: "rendered_text/train-*.parquet"
+- config_name: textatlas
+  data_files:
+  - split: train
+    path: "textatlas/train-*.parquet"
+- config_name: yfcc
+  data_files:
+  - split: train
+    path: "yfcc/train-*.parquet"
+task_categories:
+- text-to-image
+size_categories:
+- 100M<n<1B
+---
+
+# i1-captions-selected: one complete caption per image
+
+A derivative of [`zlab-princeton/i1-captions`](https://huggingface.co/datasets/zlab-princeton/i1-captions)
+that (a) **repairs** the captions that the original pipeline cut off mid-sentence, and
+(b) **collapses** the up-to-five alternative captions per image down to a single chosen one.
+
+The original dataset stores `caption1`..`caption5` and the i1 recipe samples one at random per
+training iteration. Here that choice is made once, up front, and the chosen caption is
+guaranteed to be a complete sentence — so the data can be consumed directly as `(key, caption)`.
+
+## 1. What changed vs. the original
+
+**Repaired truncation.** 1,152,605 rows (0.69%) had *no* complete caption in any column — an
+artifact of the original captioner running with a 256-token generation cap. Those images were
+re-captioned from source with the i1 recipe: **`Qwen/Qwen3-VL-30B-A3B-Instruct-FP8`** in FP8,
+the same prompts (including the TextAtlas ground-truth-annotation prompt and the RenderedText
+OCR-line prompt), images RGB → shorter-edge-512 → center-crop, and a raised token budget
+(640 → 1536, and 3072 for the OCR-transcription-heavy `textatlas`, where 640 was badly
+insufficient: repaired captions there average 293 words vs ~160 under the old cap).
+
+**One caption per row.** The `caption1`..`caption5` columns are replaced by a single `caption`.
+
+## 2. Layout
+
+Same structure and `key` values as the original: one directory per subset, `train-000NN.parquet`
+shards of 1M rows.
+
+```
+<subset>/train-000NN.parquet     # columns: key (string), caption (string)
+```
+
+| subset | rows | vs. original |
+| --- | ---: | --- |
+| `yfcc` | 97,942,438 | −2,848 |
+| `imagenet22k` | 13,673,542 | −2 |
+| `rendered_text` | 11,977,780 | −36 |
+| `megalith10m` | 9,393,285 | −686 |
+| `places365-challenge2016` | 7,220,913 | −684 |
+| `fluxreason` | 5,890,279 | = |
+| `textatlas` | 5,396,767 | −123 |
+| `redcaps` | 4,817,430 | −1 |
+| `inaturalist` | 4,812,852 | −691 |
+| `pexels` | 2,810,634 | = |
+| `gptedit` | 1,553,575 | = |
+| `midjourneyv6` | 1,240,185 | = |
+
+**Total: 166,729,680 rows** — 5,071 fewer than the original 166,734,751 (0.003%).
+
+Those 5,071 rows were **dropped**, not left broken. Their source images are dead links, so no
+clean caption exists for them anywhere and repeated resampling could not produce one. Breakdown
+of what was dropped: 4,117 captions under 25 words (on inspection these are not merely terse but
+collapsed — broken grammar, language mixing, and outright non-answers like "I cannot describe an
+image that was not provided"), 918 that could not be made to end on a sentence boundary, 21
+repetition loops, and 15 assistant refusals.
+
+```python
+from datasets import load_dataset
+ds = load_dataset("<your-namespace>/i1-captions-selected", "yfcc", split="train")
+```
+
+## 3. How the caption was chosen
+
+1. **Repaired rows win** — used verbatim.
+2. **Otherwise the best of `caption1`..`caption5`.** The five columns are statistically
+   interchangeable (measured: near-identical completion rates and length distributions), so
+   there is no "better column"; candidates are filtered on content. A candidate must end on
+   sentence-final punctuation, contain no structural junk (` ``` `, `bbox_2d`, `"image_path"`,
+   `<html>`, …), not open with a refusal, contain no repetition loop, and be ≥25 words. Among
+   survivors the longest wins — the i1 paper finds longer caption sets train better.
+3. **Salvage** — if nothing passes, every candidate is trimmed to its last complete sentence and
+   the richest survivor is used (0.08% of rows).
+4. **Drop** — if even that fails, the row is omitted (5,071 rows total, 0.003%).
+
+A "repetition loop" means the captioner degenerated into repeating itself (e.g. one caption had
+`ME` 211 times in 421 words). The detector is deliberately strict — a single token appearing
+≥50 times *and* making up >25% of the caption, or a 5-gram repeated ≥20 times — because a looser
+rule wrongly flags correct OCR transcription of images that genuinely contain long runs of a
+character.
+
+`imagenet22k`'s extra columns (`qwen2vl_2b`, `qwen2.5vl_3b`, `qwen3vl_2b`, `qwen3vl_4b`,
+`short`, `no_center_crop`) are **excluded** — they are ablation artifacts from weaker captioners
+and different preprocessing/prompt settings, not the captions used for i1's final training.
+
+## 4. Quality
+
+Audited over all 166.7M rows, every subset: **0 incomplete, 0 repetition loops, 0 refusals,
+0 empty, 0 under 25 words.** Mean caption length 128–174 words depending on subset.
+
+## 5. One thing to know before training
+
+i1 right-truncates text-encoder input to **256 tokens**. The repaired captions are longer than
+the original corpus, which was implicitly bounded by the old 256-token generation cap: measured
+with the `google/t5gemma-2b-2b-ul2` tokenizer, the original `yfcc` `caption1` is median 172
+tokens with 0.1% over 256, whereas the repaired ones are median 246 with ~41% over 256. Those
+rows will be cut at the encoder. Trimming to the last complete sentence within 256 tokens is a
+lossless offline fix if that matters for your run; it is deliberately **not** applied here.
+
+## 6. Citation
+
+The underlying dataset and recipe are from:
+
+```bibtex
+@article{zeng2026i1,
+  title={i1: A Simple and Fully Open Recipe for Strong Text-to-Image Models},
+  author={Zeng, Boya and Luo, Tianze and Pu, Shu and Shen, Jucheng and Lu, Taiming and Sarch, Gabriel and Liu, Zhuang},
+  journal={arXiv preprint arXiv:2606.11289},
+  year={2026}
+}
+```

fluxreason/train-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bddbe3990e311cc9013eae3d806701771601b43915a941eb7c28b4dd014ac71c
+size 389758929

fluxreason/train-00002.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c0e1aa0041390e401cb73f5fca5f65bc26277c05dbb8b2a4f29dd6f703f4c07
+size 389780465

fluxreason/train-00003.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46834cb420b5d2040291fd13f7fd4d8cccfe018bc2def3250b3f1208083bfd27
+size 388664746

fluxreason/train-00004.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:730bf9342c4ce35b0b9dffab2df2fd8898998dfdb214f63384be3f2a09756b5c
+size 388700765

fluxreason/train-00005.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c3626794de25ce3a3ff0778a5af84b573e45f52220a713f1337c3dceb27aa9d
+size 377634867

fluxreason/train-00006.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46f8e49a84b53cee0bd348ba2bbdfba6224fb649522b2444d74112b6b743fed2
+size 354889181

gptedit/train-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b9c968a60f27261da7f208286c5388df360d943ce3749a808059fc81d8af817
+size 363098046

gptedit/train-00002.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce81c9852a3da4a9f31c695d0ba8ed987ff49e87dbca471ee798b08046ec82ff
+size 200747946

imagenet22k/train-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa92d29ab77c7ef32c9eeb99966d587464e9f355169a31dfaaadc9da9fef429c
+size 346588655

imagenet22k/train-00002.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ed700e23cfdd73d84146b98ce52a0a79350495393d90355d2bba2074013e03d
+size 346544162

imagenet22k/train-00003.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec353ed64bb1ed20cbedfb9a9f1095e31869e4ce13046d32a0b4420deb591f07
+size 350102268

imagenet22k/train-00004.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87ac3540ce3e79a95b08303a181e2258e5c864495c9e335fb209789e1bc95620
+size 346649353

imagenet22k/train-00005.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2565c8d059236750c9fa65c9893bdc28f65c33b4b77aa642712fde23e07c025c
+size 346756301

imagenet22k/train-00006.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8e22c92274538c07c2a47c7b6622a3b70fc4f4643561b51cbdc67541e09e9c1
+size 346814949

imagenet22k/train-00007.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:318754966b788e5711a6f5d471b3b5dcaa7b416ed4d17e50a86a35372ffab88f
+size 346633053

imagenet22k/train-00008.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:149f8665bd938a5a42b327d90cc3c0a39e2ebe0d3398e4b5b4420340efbddb48
+size 346670168

imagenet22k/train-00009.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91e130aa7818e3768e37e3492211da4f137e2b2ada8aad1c215f8d5f97308537
+size 346722108

imagenet22k/train-00010.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23e981ca01267702a055832fcc64b9ec9d29b0f0e002ddcf16f533169fac9a7e
+size 346749828

imagenet22k/train-00011.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db194d4a501cc18f222b073cebe0e86f23865154c00641e97dff9dfd8528036c
+size 346624183

imagenet22k/train-00012.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b01ce0ef1b4179ace68617106f118b9b2d6ea78b97cbb33dd430331ba88d9d36
+size 346646087

imagenet22k/train-00013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf0730215c298df551f3a6b427a25b85dec3b0f9f9e3fbd2204c3debfb23994b
+size 346729432

imagenet22k/train-00014.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4be4b6c9a1e1f30f099dacd5a8b872cca5efb86e9ba65db02ab3bd61801d8d01
+size 230042758

inaturalist/train-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57f333a729f30cfcbc900ae8e1516a826121014ab5df19ceb4a0c47066cbf768
+size 230365768

inaturalist/train-00002.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd49f24a41403a4399a41bd42a3f8fa7a33020e12a50d4c6ab6cc6be5ddb03ba
+size 230179334

inaturalist/train-00003.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a0ea94d24a0e34c59e67b2b847a03b8db6662cdd39154923127134d57e7c6cf
+size 230465963

inaturalist/train-00004.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:945ab1b16c7aa0f21a9b272d8c47d5e043cff4da00c839262396f503c67a5c82
+size 229977943

inaturalist/train-00005.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:206438e4124dede7d015cb61acbd34e0d66060c8f8cbeee60d565c2b41c38fc2
+size 187095776

megalith10m/train-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8198b2ba4dad145380652e2656946b9ae7f037d2009608f34fff914711c8ae0e
+size 325216009

megalith10m/train-00002.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b20acc2f385a68d4ffc6ccbc50b8e6a62e524385a2822ee2f2efce031b844fe2
+size 325321044

megalith10m/train-00003.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc60cd57737e154856bd46027703f63b58fba54405151939c4e88ad0132c0870
+size 325277689

megalith10m/train-00004.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c144ed76437eba5ccb85f7e313edc28444429f2f3212f5b11d9545166f4d9c2e
+size 325215531

megalith10m/train-00005.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85bf2fe3316719a720fd448adcde4a116a3b81c604e9a13e1c8c33a568a71f33
+size 325218434

megalith10m/train-00006.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59ab1b445f1ab6721d2745fd654b8a9cac8a117b46b61f42278f1b89b17f18f6
+size 325358333

megalith10m/train-00007.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e7ff423c338785427dd58497127fbd3fa68931b7ebfa2f36aaca894d38f69cd
+size 325201226

megalith10m/train-00008.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60d75afe49fea541c28b58e0e39826d1544306abf6b2440bde046ab599bce747
+size 325261801

megalith10m/train-00009.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0b5a203670db037371a73d69b902aabe8ddad6d2ad9877b2ef0bc905039c74b
+size 325122222

megalith10m/train-00010.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a71dbaf94bbb50d466520812583f0ab8f5bd5498329e4f3744d7082afafee5ce
+size 128201414

midjourneyv6/train-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c91e975011d6a56301518c510c0ab78f078b1e06e5603dc4ce54f81453142cf5
+size 361781880

midjourneyv6/train-00002.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f26591f97b34e862af03367a75219eef28067b919ed4b8e64ad72d47575d7a77
+size 87068620

pexels/train-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd0a2d05fafc8223edcb1b46bbf946a570b4f4e2e24e95d3f331074f7023c767
+size 350768210

pexels/train-00002.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57f4d1aa791c3de8181b59eb3b95899bcb627c24458ee6fcdfb8c558926e67c9
+size 350716172

pexels/train-00003.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6678ef77f0a93be98f4f2307096c1c099849b3f6176cebdefe47dbdbc9adcba5
+size 284300117

places365-challenge2016/train-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7824f87d1f5180bd9470f076f01c4a61fc9973750aa3fc78a8fd82883e84e108
+size 285797567

places365-challenge2016/train-00002.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87665a90d5efb9d358f632cef643d06323a19555b74e98b303ab6064c1b189cf
+size 284384957

places365-challenge2016/train-00003.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5468d1eaf346737f8f688f83d4829990e052c7506ee63a45d06aa9cf49e3c98a
+size 284259748

places365-challenge2016/train-00004.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f4d69490c21a9ba4faaa167ddc4da3c03837622ab73871099fe3ab2594ce490
+size 284426269

places365-challenge2016/train-00005.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbca2b12308cdef4d66d0d14c9ddf71cb6a3cab95d56473635a1e2ab9e8620f5
+size 283688947

places365-challenge2016/train-00006.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:226bacfba74455d2474ffedcdb65e53c66a5589a96e87bc89d0317e7442db338
+size 285455812

places365-challenge2016/train-00007.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f02ee151baac243b8d0ff284c9049d5ff54adbeff55e47430dedc22c5c2032a5
+size 285217630