---
configs:
- config_name: default
  data_files: "*.tsv"
  names: ["kind", "sound"]
size_categories:
- n<1K
---

# [doc] format - tsv - 3

This dataset contains one tsv file at the root. We provide the name of the columns in the YAML config, since the data file has no header. The delimiter is set to `"\t"` (tabulation) due to the extension of the file. The reference for the options is the [documentation of pandas.read_csv()](https://pandas.pydata.org/docs/reference/api/pandas.read_csv.html).