README.md

---
license: cc-by-4.0
dataset_info:
- config_name: CPA
  features:
  - name: better_image
    dtype: image
  - name: worse_image
    dtype: image
  - name: dataset
    dtype: string
  - name: task
    dtype: string
  - name: category
    dtype: string
  - name: subcategory
    dtype: string
  - name: chart_type
    dtype: string
  - name: pair_id
    dtype: string
  - name: preferred_label
    dtype: string
  - name: rejected_label
    dtype: string
  splits:
  - name: train
    num_bytes: 167884676
    num_examples: 1003
  download_size: 167948561
  dataset_size: 167884676
- config_name: ChartEditing
  features:
  - name: reference
    dtype: image
  - name: edited
    dtype: image
  - name: instruction
    dtype: string
  - name: decision
    dtype: string
  - name: task
    dtype: string
  - name: chart_type
    dtype: string
  - name: example_id
    dtype: string
  splits:
  - name: train
    num_bytes: 42524042
    num_examples: 335
  download_size: 42506804
  dataset_size: 42524042
- config_name: ChartReproduction
  features:
  - name: input
    dtype: image
  - name: target
    dtype: image
  - name: decision
    dtype: string
  - name: task
    dtype: string
  - name: chart_type
    dtype: string
  - name: example_id
    dtype: string
  splits:
  - name: train
    num_bytes: 54659426
    num_examples: 315
  download_size: 54659904
  dataset_size: 54659426
configs:
- config_name: CPA
  data_files:
  - split: train
    path: CPA/train-*
  default: true
- config_name: ChartEditing
  data_files:
  - split: train
    path: ChartEditing/train-*
- config_name: ChartReproduction
  data_files:
  - split: train
    path: ChartReproduction/train-*
---