Below are datasets that we have published as part of various publications. If you are looking for additional code or datasets used in our various projects, please contact us.

• Documerica Captions Automatically generated captions for 15,911 photographs from the Documerica collection, digitized and held by the U.S. National Archives. We also include metadata taken from the National Archives that can be joined to the captions.
  [documerica_img2txt.csv.bz2 (3.8 MB)]
  [documerica_metadata.csv.bz2 (727 KB)]
• Documerica Images Automatically color corrected images of the Documerica collection, digitized and held by the U.S. National Archives.
  [documerica_mhm.tar (5.8 GB)].