Datasets:

CheneyWong
/

table_recognition_datasets_mixed

sample_id stringlengths 9 122	cohort stringclasses 35 values	cohorts listlengths 1 7	n_cohorts_merged int32 1 7	dataset stringclasses 81 values	origin stringclasses 3 values	group stringclasses 9 values	image unknown	img_format stringclasses 2 values	has_image bool 2 classes	img_w int32 0 936	img_h int32 0 923	megapixels float32 0 0.81	res_bucket stringclasses 2 values	gt_otsl stringlengths 10 30.5k	gt_html stringlengths 33 46k	gt_format_native stringclasses 2 values	metrics listlengths 0 2	n_rows int32 1 111	n_cols int32 1 115	n_cells int32 1 5.06k	n_rowspan int32 0 355	n_colspan int32 0 316	max_rowspan int32 1 99	max_colspan int32 1 51	n_merge_cells int32 0 495	complexity stringclasses 3 values	tags listlengths 0 5	numeric_cell_frac float32 0 1	has_numeric bool 2 classes	img_sha1 stringlengths 24 40	gt_conflict bool 1 class	otsl_legal bool 1 class	otsl_roundtrip_ok bool 2 classes
synth::image_095464_1634649566.836397	pub_synth	[ "pub_synth" ]	1	SynthTabNet	public	pub	[ 137, 80, 78, 71, 13, 10, 26, 10, 0, 0, 0, 13, 73, 72, 68, 82, 0, 0, 1, 75, 0, 0, 2, 0, 8, 2, 0, 0, 0, 132, 162, 217, 251, 0, 1, 0, 0, 73, 68, 65, 84, 120, 156, 236, 157, 101, 124, 20, 231, 254, 246, 103, 221, 3...	png	true	331	512	0.169472	<0.5MP	<ecel><fcel>Amortized Cost<fcel>Low<fcel>10-K<nl><fcel>Interest Rate<fcel>June 30,<fcel>Revenues:<lcel><nl><fcel>Cash<fcel>Cash provided by operating activities<fcel>Income tax expense benefit<fcel>Other assets<nl><fcel>Time deposits<fcel>Operating income loss<fcel>Net earnings<fcel>Net investment income<nl><fcel>Forei...	<table><tr><td></td><td>Amortized Cost</td><td>Low</td><td>10-K</td></tr><tr><td>Interest Rate</td><td>June 30,</td><td colspan="2">Revenues:</td></tr><tr><td>Cash</td><td>Cash provided by operating activities</td><td>Income tax expense benefit</td><td>Other assets</td></tr><tr><td>Time deposits</td><td>Operating incom...	otsl	[]	19	4	75	0	1	1	2	1	medium	[]	0.144928	true	10eba9d2a5d0edb12c48a05a92f8061f5ae0a995	false	true	true
pt1m::PMC6117860_table_2	pub_pt1m	[ "pub_pt1m" ]	1	PubTables-1M	public	pub	[ 255, 216, 255, 224, 0, 16, 74, 70, 73, 70, 0, 1, 1, 0, 0, 1, 0, 1, 0, 0, 255, 219, 0, 67, 0, 8, 6, 6, 7, 6, 5, 8, 7, 7, 7, 9, 9, 8, 10, 12, 20, 13, 12, 11, 11, 12, 25, 18, 19, 15, 20, 29, 26, 31, 30, 29, ...	jpeg	true	491	115	0.056465	<0.5MP	<fcel>Study<fcel>Treatment<fcel>Patients, number<fcel>Response<fcel>Adverse events<nl><fcel>Retrospective44<fcel>Daratumumab<fcel>25<fcel>HR: ORR 76%, CR 36%<fcel>Infusion reaction GI-II<nl><fcel>Retrospective60<fcel>Doxycyclinea<fcel>30<fcel>HR 56% versus 33%<fcel>Photosensitive rash 3<nl><fcel>Phase I/II52<fcel>NEOD0...	<table><tr><td>Study</td><td>Treatment</td><td>Patients, number</td><td>Response</td><td>Adverse events</td></tr><tr><td>Retrospective44</td><td>Daratumumab</td><td>25</td><td>HR: ORR 76%, CR 36%</td><td>Infusion reaction GI-II</td></tr><tr><td>Retrospective60</td><td>Doxycyclinea</td><td>30</td><td>HR 56% versus 33%</...	otsl	[]	6	5	30	0	0	1	1	0	simple	[]	0.633333	true	f50ccf0f25501a85b906b48d989fa84546fa0a90	false	true	true
pt1m::PMC4815196_table_1	pub_pt1m	[ "pub_pt1m" ]	1	PubTables-1M	public	pub	[ 255, 216, 255, 224, 0, 16, 74, 70, 73, 70, 0, 1, 1, 0, 0, 1, 0, 1, 0, 0, 255, 219, 0, 67, 0, 8, 6, 6, 7, 6, 5, 8, 7, 7, 7, 9, 9, 8, 10, 12, 20, 13, 12, 11, 11, 12, 25, 18, 19, 15, 20, 29, 26, 31, 30, 29, ...	jpeg	true	611	384	0.234624	<0.5MP	<fcel>Dependent variable<fcel>Independent variables<fcel>Beta<fcel>R2<fcel>p<nl><fcel>Pulse frequency<fcel>BMI SDS<fcel>0,326<fcel>0,188<fcel><0,001<nl><ucel><fcel>Weight loss<fcel>−0,199<fcel>0,070<fcel><0,001<nl><ucel><fcel>Final rate of weight loss<fcel>−0,189<fcel>0,027<fcel>p = 0,003<nl><fcel>Systolic blood pressu...	<table><tr><td>Dependent variable</td><td>Independent variables</td><td>Beta</td><td>R2</td><td>p</td></tr><tr><td rowspan="3">Pulse frequency</td><td>BMI SDS</td><td>0,326</td><td>0,188</td><td><0,001</td></tr><tr><td>Weight loss</td><td>−0,199</td><td>0,070</td><td><0,001</td></tr><tr><td>Final rate of weight l...	otsl	[]	22	5	96	7	0	3	1	7	medium	[]	0.458333	true	e91e97150fb760f2d63d81f422b2d11004964478	false	true	true
real5c::Screen-Photography::eastmoney_f72b162baf25f1a2060e4ce877071ec27e490ab16e9de1a2b18572b596d2bde8.pdf_2.jpg::t1	ood_full_select	[ "ood_full_select", "ood_real5c" ]	2	Real5C-Screen-Photography	ood	ood	null	null	false	0	0	0	<0.5MP	<fcel>合肥<fcel>上海<nl><fcel>地址：安徽省合肥市梅山路18号安徽国际金融中心A座国元证券<fcel>地址：上海市浦东新区民生路1199号证大五道口广场16楼国元证券<nl><fcel>邮编： 230000<fcel>邮编： 200135<nl><fcel>传真：（0551）62207952<fcel>传真：（021）68869125<nl><ecel><fcel>电话：（021）51097188<nl>	<table><tr><td>合肥</td><td>上海</td></tr><tr><td>地址：安徽省合肥市梅山路18号安徽国际金融中心A座国元证券</td><td>地址：上海市浦东新区民生路1199号证大五道口广场16楼国元证券</td></tr><tr><td>邮编： 230000</td><td>邮编： 200135</td></tr><tr><td>传真：（0551）62207952</td><td>传真：（021）68869125</td></tr><tr><td></td><td>电话：（021）51097188</td></tr></table>	html	[]	5	2	10	0	0	1	1	0	simple	[]	0.777778	true	missing:09fa8f2ed56a57cf	false	true	true
pt1m::PMC3571826_table_0	pub_pt1m	[ "pub_pt1m" ]	1	PubTables-1M	public	pub	"/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0aHBwgJC4nICIsIxwcKDcpLDA(...TRUNCATED)	jpeg	true	437	125	0.054625	<0.5MP	"<fcel>Sensor<fcel>c0×103 [mbar/s]<fcel>δc0×103 [mbar/s]<fcel>c1×103 [s−1]<fcel>δc1×103 [s(...TRUNCATED)	"<table><tr><td>Sensor</td><td>c0×103 [mbar/s]</td><td>δc0×103 [mbar/s]</td><td>c1×103 [s−1]</(...TRUNCATED)	otsl	[]	6	6	36	0	0	1	1	0	simple	[]	0.972222	true	e152421214f9b15b72973a0844b0b5f49c8bcbac	false	true	true
ooddeg::PubTabNet::train::PMC2567350_006_00	ooddeg_probe	[ "ooddeg_probe" ]	1	OODSynth3	ood	opd_pool	null	null	false	0	0	0	<0.5MP	"<fcel>Data type\\Mapping source<fcel>SSMap<fcel>DR PDB<fcel>SIFTS<fcel>PDBSWS<nl><fcel>1. Number of(...TRUNCATED)	"<table><tr><td>Data typeMapping source</td><td>SSMap</td><td>DR PDB</td><td>SIFTS</td><td>PDBSWS</t(...TRUNCATED)	otsl	[]	3	5	15	0	0	1	1	0	simple	[]	0.666667	true	missing:9a8d0f40e41a24cd	false	true	true
pt1m::PMC4408140_table_2	pub_pt1m	[ "pub_pt1m" ]	1	PubTables-1M	public	pub	"/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0aHBwgJC4nICIsIxwcKDcpLDA(...TRUNCATED)	jpeg	true	683	205	0.140015	<0.5MP	"<fcel>ξ<fcel>R = 0.7<lcel><fcel>R = 0.9<lcel><fcel>R = 1.0<lcel><nl><ucel><fcel>Solution<fcel>Resi(...TRUNCATED)	"<table><tr><td rowspan=\"2\">ξ</td><td colspan=\"2\">R = 0.7</td><td colspan=\"2\">R = 0.9</td><td(...TRUNCATED)	otsl	[]	13	7	87	1	3	2	2	4	medium	[]	0.91954	true	d5f080ed8dc50f13d33892bb89d3f28bd07db315	false	true	true
synth::image_072171_1634638050.697725	pub_synth	[ "pub_synth" ]	1	SynthTabNet	public	pub	"iVBORw0KGgoAAAANSUhEUgAAAgAAAACkCAIAAABuCAF8AABNs0lEQVR4nO3dZ0AU1/ow8GlbWXaXDtKLNBWkiKiAXaPEjordKP4(...TRUNCATED)	png	true	512	164	0.083968	<0.5MP	"<fcel>N %<fcel>Description<lcel><lcel><lcel><lcel><fcel>Reference<fcel>Median<fcel>SNP<fcel>F<fcel>(...TRUNCATED)	"<table><tr><td>N %</td><td colspan=\"5\">Description</td><td>Reference</td><td>Median</td><td>SNP</(...TRUNCATED)	otsl	[]	9	15	120	1	2	2	6	2	complex	[ "wide" ]	0.333333	true	546552f955635eb5d338a050b3944ffb4a14dd44	false	true	true
cxr::controlled_20261006_00770	cx_render	[ "cx_render" ]	1	ControlledRender	selfbuilt	cx	null	null	false	0	0	0	<0.5MP	"<fcel>Item<fcel>Code<fcel>Jan<fcel>Feb<fcel>Mar<fcel>Group subtotal<lcel><fcel>Ratio<nl><fcel>East<(...TRUNCATED)	"<table><tr><td>Item</td><td>Code</td><td>Jan</td><td>Feb</td><td>Mar</td><td colspan=\"2\">Group su(...TRUNCATED)	otsl	[]	17	8	135	0	1	1	2	1	medium	[]	0.820313	true	missing:af1a4e0b46248617	false	true	true
pt1m::PMC3185044_table_1	pub_pt1m	[ "pub_pt1m" ]	1	PubTables-1M	public	pub	"/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgKDBQNDAsLDBkSEw8UHRofHh0aHBwgJC4nICIsIxwcKDcpLDA(...TRUNCATED)	jpeg	true	597	181	0.108057	<0.5MP	"<fcel>Variables<fcel>Non-breast-fed (n = 34) n (%)<fcel>breastfed (n = 73) n (%)<fcel>Non-adjusted<(...TRUNCATED)	"<table><tr><td rowspan=\"2\">Variables</td><td rowspan=\"2\">Non-breast-fed (n = 34) n (%)</td><td (...TRUNCATED)	otsl	[]	11	7	72	3	2	2	2	5	medium	[]	0.833333	true	54b40c8a85aa2095246a2ced0ddd484349c4262f	false	true	true

End of preview. Expand in Data Studio

Table Recognition Datasets (Mixed)

Evaluation data for table recognition (OCR → table markup), packaged as parquet shards with the images embedded inline.

Every table is stored in two dialects — OTSL and canonical HTML — so a model can be scored against either without re-deriving one.

Dataset	Rows	With image	Files	What it is
`table_eval_paddleocr_vl`	240,075	152,982	32 shards	The main evaluation set
`table_eval_paddleocr_vl_excluded_toolong`	3,828	3,633	4 shards	Samples withheld from evaluation because their tables are too long

Start with the first. The second is an archive of exclusions, not a scoring target.

Loading

import pyarrow.parquet as pq

t = pq.read_table("table_eval_paddleocr_vl/shard_0000.parquet")
layout_only = t.filter(t.column("has_numeric") == False).to_pylist()

Images are the original encoded bytes, never re-encoded — recompressing would introduce artefacts that contaminate the metric. When serving many requests, resolve images lazily per request; do not call pq.read_table(shard).to_pylist() on an inference path, since it re-decodes an entire row group per request and collapses under concurrency.

Scoring

Normalise prediction and ground truth to canonical HTML before computing TEDS, so a model that answers in OTSL and one that answers in HTML are compared on equal terms. The full column reference and distributions are in the main dataset card.

For the SPRINT-MUSTARD family the ground truth is deliberately all-empty <td>, so only TEDS-S is meaningful there; its TEDS column reads artificially low.

Downloads last month: 94