Datasets:
id string | format string | source string | stage string | token_count int64 | image image | target_text string | provenance string |
|---|---|---|---|---|---|---|---|
dot_github_search_0012429 | dot | github_search | stage_8192-12288 | 9,863 | Not supported with pagination yet | digraph neomutt
{
graph [
rankdir="TB"
nodesep="0.5"
ranksep="0.5"
];
node [
shape="plain"
];
edge [
penwidth="4.5"
arrowsize="1.0"
color="#c0c0c0"
];
obj_0x1dba5e0 [
label=<<table cellspacing="0" border="1" rows="*" color="#d0d0d0">
<tr>
<td border="0" align="left" bgcolor="#9347de" por... | |
dot_github_search_0003926 | dot | github_search | stage_32768_plus | 40,945 | Not supported with pagination yet | digraph RNN {
labelloc="t";
label="Genome Fitness: 2477.07% MAE";
graph [pad="0.01", nodesep="0.05", ranksep="0.9"];
{
rank = source;
node1 [shape=box,color=green,label="input 1\ndepth 0\nLatitude"];
node2 [shape=box,color=green,label="input 2\ndepth 0\nLongitude"];
node3 [shape=box,color=green,label="input ... | |
dot_github_search_0001917 | dot | github_search | stage_4096-6144 | 4,267 | Not supported with pagination yet | digraph {
"Function- _ZNSaIcEC2Ev"
"Function- _ZNSt6chrono8durationIxSt5ratioILx1ELx1000000000EEEC2IxvEERKT_"
"Function- _ZNSt6chrono8durationIxSt5ratioILx1ELx1000000000EEEC2IxvEERKT_" -> "Function- _ZNSaIcEC2Ev" [dir=back]
"Function- _ZNSt6chrono15duration_valuesIxE4zeroEv"
"Function- _ZNSt6chrono15duration_value... | |
dot_github_search_0001985 | dot | github_search | stage_2048-4096 | 2,498 | Not supported with pagination yet | digraph boys1foursThayer {BUAcademy [URL="boys1foursThayerBUAcademy.html"];MarianapolisPrep [URL="boys1foursThayerMarianapolisPrep.html"];Medford [URL="boys1foursThayerMedford.html"];StMarysLynn [URL="boys1foursThayerStMarysLynn.html"];BrewsterAcademy [URL="boys1foursThayerBrewsterAcademy.html"];Thayer [URL="boys1fours... | |
dot_github_search_0006268 | dot | github_search | stage_8192-12288 | 8,466 | Not supported with pagination yet | digraph "pkg_resources._vendor.pyparsing.core.ParserElement.split"
{
// LATEX_PDF_SIZE
bgcolor="transparent";
edge [fontname=Helvetica,fontsize=10,labelfontname=Helvetica,labelfontsize=10];
node [fontname=Helvetica,fontsize=10,shape=box,height=0.2,width=0.4];
rankdir="RL";
Node1 [id="Node000001",label="pkg_r... | |
dot_github_search_0003508 | dot | github_search | stage_2048-4096 | 3,232 | Not supported with pagination yet | digraph {
node [fontsize=9 height=0.2 shape=circle width=0.2]
"dintancia projetil 1" [fillcolor=lightgray shape=box style=filled]
"distancia projetil 2" [fillcolor=lightgray shape=box style=filled]
"distancia projetil 3" [fillcolor=lightgray shape=box style=filled]
"distancia projetil 4" [fillcolor=lightgray shape... | |
dot_thestack_restructuredtext_0000221 | dot | thestack_restructuredtext | stage_0-2048 | 44 | Not supported with pagination yet | digraph {
"poliastro.coordinates" -> "body_centered_to_icrs", "icrs_to_body_centered",
"inertial_body_centered_to_pqw", "transform" ;
} | |
dot_thestack_graphviz_dot_0010543 | dot | thestack_graphviz_dot | stage_16384-20480 | 18,948 | Not supported with pagination yet | "digraph G {\nnode[shape=rectangle,style=filled,fillcolor=lightsteelblue,color=lightsteelblue]\nbgco(...TRUNCATED) | |
dot_thestack_md_0000461 | dot | thestack_md | stage_0-2048 | 730 | Not supported with pagination yet | "digraph agents {\nrankdir=LR;\n\ncontributions [label=\"CONTRIBUTIONS\" shape=box penwidth=0 fontsi(...TRUNCATED) | |
dot_thestack_graphviz_dot_0004003 | dot | thestack_graphviz_dot | stage_6144-8192 | 8,164 | Not supported with pagination yet | "graph {\n 0;\n 1;\n 2;\n 3;\n 4;\n 5;\n 6;\n 7;\n 8;\n 9;\n 10;\n 11;\n 12;\n 13;\n (...TRUNCATED) |
End of preview. Expand in Data Studio
StructCrawl Sample — Structured Image Understanding
A uniform random subsample (~500 per format×stage×source cell) of a larger web-crawled corpus for structured image understanding: converting a rendered image of a diagram/chart into the editable structured-text code that produces it.
This is a sharing sample, not the full dataset — it is stratified so every length band and every source is represented for inspection.
Configs (one per code representation)
| Config | Rows | Description |
|---|---|---|
svg |
~21.7K | Diagrams/charts/schematics as SVG XML |
dot |
~9.7K | Graphs as Graphviz DOT |
plantuml |
~3.7K | UML/sequence diagrams as PlantUML |
vega |
~3.5K | Charts as Vega-Lite JSON specs |
mermaid |
~3.3K | Flowcharts/sequence/class/ER as Mermaid |
(Exact per-config counts in _manifest.json.)
Usage
from datasets import load_dataset
ds = load_dataset("<your-username>/<repo-name>", "svg", split="train")
ex = ds[0]
ex["image"] # PIL.Image — the rendered diagram
ex["target_text"] # the SVG/Mermaid/DOT/PlantUML/Vega source to reproduce it
ex["stage"] # token-length band, e.g. "stage_0-2048"
ex["source"] # crawl origin, e.g. "commoncrawl_inline"
ex["token_count"] # target_text length in SFT tokenizer tokens
Fields
| Field | Type | Notes |
|---|---|---|
id |
string | unique sample id |
format |
string | svg | mermaid | dot | plantuml | vega |
source |
string | crawl origin (commoncrawl_inline, github_repos, github_search, thestack_*, web_linked, wikimedia, ...) |
stage |
string | token-length band (2048-token steps to 8192, 4096-steps to 32768, then stage_32768_plus) |
token_count |
int64 | target_text length (SFT tokenizer) |
image |
Image | rendered diagram (native dimensions, white background) |
target_text |
string | the structured-text code that renders to image |
provenance |
string | original url / repo / path when available |
Sampling & processing
- Stratified: up to 500 samples drawn uniformly at random (seed 42, reservoir sampling) from every (format × stage × source) cell.
- Washed: samples whose code cannot be reproduced from the image alone
were removed upstream — base64-embedded rasters/fonts, external image/data
references, scripts,
<foreignObject>, PlantUML!include/sprites. - Deduplicated: exact content-level dedup (whitespace-normalized SHA-256), filename-independent. Near-duplicate template families may still remain.
- Render-validated: every
target_textwas executed and produced a non-blank image; the pairedimageis that render.
Caveats
- Licensing is mixed (public web, license-filtered source corpora, CC-BY
Commons) — intended for research; per-sample
provenanceretained for attribution. - SVG length is heavy-tailed; use
stage/token_countto filter. - This is a small representative slice of a ~860K-pair processed corpus.
- Downloads last month
- 3