---
dataset_info:
  features:
  - name: id
    dtype: int32
  - name: image
    dtype: image
  - name: conversations
    list:
    - name: role
      dtype: string
    - name: content
      dtype: string
  - name: metadata
    struct:
    - name: image_id
      dtype: int32
    - name: data_category
      dtype: string
    - name: chart_source
      dtype: string
    - name: chart_color
      dtype: string
    - name: chart_multi
      dtype: string
    - name: chart_year
      dtype: string
    - name: chart_main
      dtype: string
    - name: chart_sub
      dtype: string
    - name: width
      dtype: int32
    - name: height
      dtype: int32
    - name: annotations
      list:
      - name: image_id
        dtype: int32
      - name: is_title
        dtype: bool
      - name: is_legend
        dtype: bool
      - name: is_datalabel
        dtype: bool
      - name: is_unit
        dtype: bool
      - name: is_base
        dtype: bool
      - name: is_axis_label_x_axis
        dtype: bool
      - name: is_axis_label_y_axis
        dtype: bool
      - name: title
        dtype: string
      - name: legend
        list: string
      - name: unit
        dtype: string
      - name: base
        dtype: string
      - name: axis_title
        struct:
        - name: x_axis
          dtype: string
        - name: y_axis
          dtype: string
      - name: axis_label
        struct:
        - name: x_axis
          list: string
        - name: y_axis
          list: string
      - name: data_label
        list:
          list: string
  - name: summary
    list: string
  splits:
  - name: train
    num_bytes: 11945231599.75
    num_examples: 240002
  - name: validation
    num_bytes: 1485438923.0
    num_examples: 30000
  download_size: 12565577510
  dataset_size: 13430670522.75
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---