---
configs:
- config_name: default
  data_files: "data.csv"
  delimiter: "|"
  header: 0
  names: ["kind", "sound"]
size_categories:
- n<1K
---

# [doc] formats - csv - 3

This dataset contains one csv file at the root. We provide several options in the YAML config (`"|"` delimiter, name of the columns, index of the header row). The reference for the options is the [documentation of pandas.read_csv()](https://pandas.pydata.org/docs/reference/api/pandas.read_csv.html).