metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
dataset_info:
features:
- name: image
dtype: image
- name: annotation
struct:
- name: description
dtype: string
- name: objects
list:
- name: bitmap
struct:
- name: data
dtype: string
- name: origin
sequence: int64
- name: classId
dtype: int64
- name: classTitle
dtype: string
- name: createdAt
dtype: string
- name: description
dtype: string
- name: geometryType
dtype: string
- name: id
dtype: int64
- name: labelerLogin
dtype: string
- name: points
struct:
- name: exterior
sequence:
sequence: int64
- name: interior
sequence:
sequence:
sequence: int64
- name: tags
sequence: 'null'
- name: updatedAt
dtype: string
- name: size
struct:
- name: height
dtype: int64
- name: width
dtype: int64
- name: tags
list:
- name: createdAt
dtype: string
- name: id
dtype: int64
- name: labelerLogin
dtype: string
- name: name
dtype: string
- name: tagId
dtype: int64
- name: updatedAt
dtype: string
- name: value
dtype: 'null'
- name: filename
dtype: string
- name: embedding
sequence: float32
- name: cropped
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 1083796984
num_examples: 358
download_size: 939274079
dataset_size: 1083796984