Read from Kaggle — tl_read_kaggle • tidylearn

Downloads a dataset file from Kaggle using the Kaggle CLI and reads it into a tidylearn_data object. Requires the Kaggle CLI to be installed and configured (pip install kaggle).

Usage

tl_read_kaggle(source, file = NULL, dest = NULL, type = "dataset", ...)

Arguments

source: A Kaggle dataset slug (e.g., "user/dataset-name") or a Kaggle URL. A URL may point at any tab of the dataset or competition page (/data, /code, /versions/2) and may carry a query; a competition URL is read as a competition without type being set.
file: The specific file to read from the dataset, as a path within it; a file the CLI saved under its base name is found too. If NULL, the download is searched for files these readers handle (CSV, TSV, Excel, Parquet and JSON, with compressed CSV/TSV and .ndjson); with several, the newest is read and a message names it.
dest: Directory to keep the download in. The default is a fresh per-dataset directory under tempdir(). A supplied dest receives this download's files, replacing files of the same name; nothing else in it is read, unpacked or changed.
type: Either "dataset" (default) or "competition". Left unset, a competition URL in source sets it.
...: Additional arguments passed to the format-specific reader.

Value

A tidylearn_data object containing the downloaded data.

Examples

if (FALSE) { # \dontrun{
# Needs the Kaggle CLI and Kaggle credentials
tl_read_kaggle("zillow/zecon", file = "Zip_time_series.csv")
tl_read_kaggle("titanic", file = "train.csv", type = "competition")
tl_read_kaggle("https://www.kaggle.com/competitions/titanic/data",
               file = "train.csv")
} # }