prepdatakit 1.5.8

A comprehensive toolkit for preprocessing datasets, including data reading, data summary generation, handling missing values, and categorical data encoding.

PrepDataKit

PrepDataKit is a Python package that provides a toolkit for preprocessing datasets. It offers various functions to assist in reading data from different file formats, summarizing datasets, handling missing values, and encoding categorical data.

Installation

You can install PrepDataKit using pip:

pip install prepdatakit

Sample Data

Category	Price	In Stock	Description
Fruit	2.50	True	Ripe and delicious
Animal	None	False	Needs more data
Color	1.99		Vivid and bright
Tool	9.99	True	Heavy duty and reliable (Maybe)

Download CSV

Usage

Here's an example of how to use PrepDataKit:

from prepdatakit import prepdatakit
import time
        
if __name__ == "__main__":
    
    data = prepdatakit.read_file("reviews.csv")

    # Reading the file
    print("Data Information:")
    print(prepdatakit.tabulate(data.head(), headers="keys", tablefmt="fancy_grid"))
    print("\nData Type:", type(data))
    print("Data Shape:", data.shape)
    print("=" * 50)

    # Generating summary
    summary = prepdatakit.get_summary(data)
    print("\nSummary Statistics:")
    for key, value in summary.items():
        print(key + ":")
        if isinstance(value, prepdatakit.pd.DataFrame):
            print(prepdatakit.tabulate(value, headers="keys", tablefmt="fancy_grid"))
        elif isinstance(value, dict):
            for k, v in value.items():
                print(f"  {k}: {v}")
        print("-" * 50)

    # Handling missing values
    clean_data = prepdatakit.handle_missing_values(data, strategy="remove")
    print("\nCleaned Data:")
    # print(tabulate(clean_data.head(), headers='keys', tablefmt='fancy_grid'))
    with open("clean_data.txt", "w", encoding="utf-8") as f:
        f.write(prepdatakit.tabulate(clean_data, headers="keys", tablefmt="fancy_grid"))
    print("\nData Type:", type(clean_data))

    # Encoding categorical data
    encoded_data = prepdatakit.one_hot_encode(clean_data)
    print("\nEncoded Data:")
    with open("encoded_data.txt", "w", encoding="utf-8") as f:
        f.write(prepdatakit.tabulate(encoded_data, headers="keys", tablefmt="psql"))
    # print(tabulate(encoded_data.head(), headers='keys', tablefmt='plain'))
    print("\nData Type:", type(encoded_data))
    print("Data Shape:", encoded_data.shape)
    print("=" * 50)