---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: objectID
    dtype: int64
  - name: isHighlight
    dtype: bool
  - name: accessionNumber
    dtype: string
  - name: accessionYear
    dtype: int64
  - name: isPublicDomain
    dtype: bool
  - name: primaryImage
    dtype: string
  - name: primaryImageSmall
    dtype: string
  - name: department
    dtype: string
  - name: objectName
    dtype: string
  - name: title
    dtype: string
  - name: culture
    dtype: float64
  - name: period
    dtype: float64
  - name: dynasty
    dtype: float64
  - name: reign
    dtype: float64
  - name: portfolio
    dtype: float64
  - name: artistRole
    dtype: string
  - name: artistPrefix
    dtype: float64
  - name: artistDisplayName
    dtype: string
  - name: artistDisplayBio
    dtype: string
  - name: artistSuffix
    dtype: float64
  - name: artistAlphaSort
    dtype: string
  - name: artistNationality
    dtype: string
  - name: artistBeginDate
    dtype: float64
  - name: artistEndDate
    dtype: float64
  - name: artistGender
    dtype: float64
  - name: artistWikidata_URL
    dtype: string
  - name: artistULAN_URL
    dtype: string
  - name: objectDate
    dtype: string
  - name: objectBeginDate
    dtype: int64
  - name: objectEndDate
    dtype: int64
  - name: medium
    dtype: string
  - name: dimensions
    dtype: string
  - name: creditLine
    dtype: string
  - name: geographyType
    dtype: float64
  - name: city
    dtype: float64
  - name: state
    dtype: float64
  - name: county
    dtype: float64
  - name: country
    dtype: float64
  - name: region
    dtype: float64
  - name: subregion
    dtype: float64
  - name: locale
    dtype: float64
  - name: locus
    dtype: float64
  - name: excavation
    dtype: float64
  - name: river
    dtype: float64
  - name: classification
    dtype: string
  - name: rightsAndReproduction
    dtype: float64
  - name: linkResource
    dtype: float64
  - name: metadataDate
    dtype: string
  - name: repository
    dtype: string
  - name: objectURL
    dtype: string
  - name: objectWikidata_URL
    dtype: float64
  - name: isTimelineWork
    dtype: bool
  - name: GalleryNumber
    dtype: float64
  splits:
  - name: train
    num_bytes: 74129545
    num_examples: 942
  download_size: 73571351
  dataset_size: 74129545
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
license: cc0-1.0
task_categories:
- unconditional-image-generation
size_categories:
- n<1K
---
# Dataset Card for Brewster Carriages

![](samples.png)

<!-- Provide a quick summary of the dataset. -->

A permissively licensed dataset of original [Brewster & Co.](https://en.wikipedia.org/wiki/Brewster_%26_Co.) carriage illustrations (and associated metadata) from the public domain. While this dataset could be used for a variety of tasks and purposes, it was originally collated by [Jonathan Dinu](https://jonathandinu.com) to train an unconditional image generation model.

## Dataset Description

- **Curated by:** [Jonathan Dinu](https://jonathandinu.com)
- **License:** Dataset [CC0](https://creativecommons.org/public-domain/cc0/) licensed, source images are in the Public Domain.

## Dataset Sources

<!-- Provide the basic links for the dataset. -->

Dataset is sourced from The Metropolitan Museum of Art [open access archives](https://www.metmuseum.org/about-the-met/policies-and-documents/open-access) using a query for ["brewster+co"](https://collectionapi.metmuseum.org/public/collection/v1/search?q=brewster+co). Results are then:

1. Filtered to make sure the source images are all in the public domain.
2. Manually filtered to remove images of low quality.
3. Center cropped and resized to 1024x1024.