kaggle-downloader-package 0.1.0

A package to download datasets from Kaggle.

KaggleDownloader

KaggleDownloader is a Python package designed to simplify the process of downloading and extracting datasets from Kaggle. With this package, you can search for datasets by theme, authenticate your Kaggle account, download any available dataset, and extract the contents of zip files, all with a simple interface.

Features

• Authenticate with Kaggle: Easily authenticate using your Kaggle API credentials.
• Search for Datasets: Search for datasets on Kaggle based on specific themes.
• Download Datasets: Download any dataset from Kaggle using its slug.
• Extract Zip Files: Extract the contents of zip files downloaded from Kaggle.
• Directory Management: Automatically create directories to store datasets if they don't exist.

Installation

First, install the package using Poetry. Make sure you have Poetry installed.

poetry install

Usage

You can use KaggleDownloader through the command line. Here’s an example:

python -m kaggle_downloader <dataset_slug> --path <download_path>

Replace <dataset_slug> with the identifier of the Kaggle dataset and <download_path> with the location where you want the dataset to be saved.

Example:

python -m kaggle_downloader zynicide/wine-reviews --path ./datasets

Methods

Authentication

The method authenticate_kaggle() allows you to authenticate with your Kaggle account by providing the path to your kaggle.json file, which contains your API credentials.

Searching for Datasets

The method search_datasets(dataset_theme) allows you to search for datasets on Kaggle related to a specific theme.

Downloading Datasets

The method download_dataset(dataset_slug, path) allows you to download a dataset by providing its Kaggle slug and specifying the location to save it.

Extracting Zip Files

The method extract_zip(zip_path, extract_to) enables you to extract the contents of a zip file. If the file is not a valid zip or doesn’t exist, an error is raised.

Contributing

Feel free to contribute to this project by submitting issues, feature requests, or pull requests on GitHub.

License

This project is licensed under the MIT License. See the LICENSE file for details.

Author

Mariano Gobea Alcoba
Email: [e-mail]