Dataset Open Access

im2latex-100k , arXiv:1609.04938

Kanervisto, Anssi


MARC21 XML Export

<?xml version='1.0' encoding='UTF-8'?>
<record xmlns="http://www.loc.gov/MARC21/slim">
  <leader>00000nmm##2200000uu#4500</leader>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">im2latex</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">latex</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">tex</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">formula</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">openai</subfield>
  </datafield>
  <controlfield tag="005">20200124192546.0</controlfield>
  <controlfield tag="001">56198</controlfield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">237383</subfield>
    <subfield code="z">md5:1bc17b865796dca5df15250b4da7804f</subfield>
    <subfield code="u">https://zenodo.org/record/56198/files/im2latex_test.lst</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">1923255</subfield>
    <subfield code="z">md5:d5607c37aa00576098a9e4bad84a7040</subfield>
    <subfield code="u">https://zenodo.org/record/56198/files/im2latex_train.lst</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">12323106</subfield>
    <subfield code="z">md5:974c0a14f0daa6d91ecd0e625f1ddf52</subfield>
    <subfield code="u">https://zenodo.org/record/56198/files/im2latex_formulas.lst</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">292150187</subfield>
    <subfield code="z">md5:cf25f2408f1ea09bbd096890a6361533</subfield>
    <subfield code="u">https://zenodo.org/record/56198/files/formula_images.tar.gz</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">213665</subfield>
    <subfield code="z">md5:cf6eeee02bc443b1b9557685fbfe7ea5</subfield>
    <subfield code="u">https://zenodo.org/record/56198/files/im2latex_validate.lst</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">924</subfield>
    <subfield code="z">md5:3d4cb64d8c403148ff06370d71072cdc</subfield>
    <subfield code="u">https://zenodo.org/record/56198/files/readme.txt</subfield>
  </datafield>
  <datafield tag="542" ind1=" " ind2=" ">
    <subfield code="l">open</subfield>
  </datafield>
  <datafield tag="260" ind1=" " ind2=" ">
    <subfield code="c">2016-06-21</subfield>
  </datafield>
  <datafield tag="909" ind1="C" ind2="O">
    <subfield code="p">openaire_data</subfield>
    <subfield code="o">oai:zenodo.org:56198</subfield>
  </datafield>
  <datafield tag="100" ind1=" " ind2=" ">
    <subfield code="u">University of Eastern Finland</subfield>
    <subfield code="a">Kanervisto, Anssi</subfield>
  </datafield>
  <datafield tag="245" ind1=" " ind2=" ">
    <subfield code="a">im2latex-100k ,  arXiv:1609.04938</subfield>
  </datafield>
  <datafield tag="540" ind1=" " ind2=" ">
    <subfield code="u">http://creativecommons.org/publicdomain/zero/1.0/legalcode</subfield>
    <subfield code="a">Creative Commons Zero v1.0 Universal</subfield>
  </datafield>
  <datafield tag="650" ind1="1" ind2="7">
    <subfield code="a">cc-by</subfield>
    <subfield code="2">opendefinition.org</subfield>
  </datafield>
  <datafield tag="520" ind1=" " ind2=" ">
    <subfield code="a">&lt;p&gt;A prebuilt dataset for OpenAI's task for image-2-latex system. Includes total of ~100k formulas and images splitted into train, validation and test sets. Formulas were parsed from LaTeX sources provided here: http://www.cs.cornell.edu/projects/kddcup/datasets.html(originally fromĀ  arXiv)&lt;/p&gt;

&lt;p&gt;Each image is a PNG image of fixed size. Formula is in black and rest of the image is transparent.&lt;/p&gt;

&lt;p&gt;For related tools (eg. tokenizer) check out this repository: https://github.com/Miffyli/im2latex-dataset&lt;br&gt;
For pre-made evaluation scripts and built im2latex system check this repository: https://github.com/harvardnlp/im2markup&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Newlines used in &lt;em&gt;formulas_im2latex.lst &lt;/em&gt;are UNIX-style newlines (\n). Reading file with other type of newlines results to slightly wrong amount of lines (104563 instead of 103558), and thus breaks the structure used by this dataset. Python 3.x reads files using newlines of the running system by default, and to avoid this file must be opened with newlines&lt;/strong&gt;&lt;strong&gt;="\n" (eg. open("formulas_im2latex.lst", newline="\n")).&lt;/strong&gt;&lt;/p&gt;</subfield>
  </datafield>
  <datafield tag="773" ind1=" " ind2=" ">
    <subfield code="n">arxiv</subfield>
    <subfield code="i">isPartOf</subfield>
    <subfield code="a">arXiv:1609.04938</subfield>
  </datafield>
  <datafield tag="024" ind1=" " ind2=" ">
    <subfield code="a">10.5281/zenodo.56198</subfield>
    <subfield code="2">doi</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">dataset</subfield>
  </datafield>
</record>
8,946
12,697
views
downloads
All versions This version
Views 8,9468,972
Downloads 12,69712,697
Data volume 2.0 TB2.0 TB
Unique views 7,9767,998
Unique downloads 2,9612,961

Share

Cite as