---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: csv_text
    dtype: string
  - name: viz_image
    dtype: image
  - name: viz_type
    dtype: string
  splits:
  - name: train
    num_bytes: 94402356
    num_examples: 134
  download_size: 9264813
  dataset_size: 94402356
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
license: mit
language:
- en
pretty_name: Question Answering from CSV Data vs Data Visualization
size_categories:
- n<1K
---

## CSVvsVizQA

Dataset to compare question answering ability from CSV Data vs Data Visualization images.