Skip to main content

A data preprocessing library in Python

Project description

Data Preprocessing Library

A comprehensive data preprocessing library in Python. This library provides various functions for data cleaning, transformation, and manipulation operations, designed to make your data preprocessing tasks easier and more efficient.

Features

  • Handling Missing Values: Mean, median, constant imputation, and deletion.
  • Outlier Detection and Correction: IQR method with threshold.
  • Data Standardization and Normalization: Min-Max and Standard normalization.
  • Text Cleaning and Manipulation: Remove stopwords, lowercase conversion, punctuation removal, lemmatization (requires NLTK).
  • Feature Engineering: Create new features from existing ones.
  • Data Type Conversion: Convert to numeric, categorical, and datetime.
  • Encoding Categorical Data: One-hot encoding and label encoding.
  • Date and Time Manipulation: Extract date components, calculate date differences.

Usage

Here's an example of how to use the library:

  • Load the data

    • data = pd.read_csv('synthetic_sample_data.csv')
  • Handling missing values

    • missing_handler = MissingValueHandler(strategy="mean")
    • data = missing_handler.fit_transform(data)
  • Outlier detection and correction

    • outlier_handler = OutlierHandler(method="iqr", threshold=1.5)
    • data = outlier_handler.fit_transform(data)
  • Data standardization

    • scaler = Scaler(method="standard")
    • data = scaler.fit_transform(data)
  • Text cleaning

    • text_cleaner = TextCleaner(remove_stopwords=True, lemmatize=True)
      • if 'Summary' in data.columns: data['Summary'] = data['Summary'].astype(str).apply(text_cleaner.clean)
  • Encoding categorical data

    • if 'Genre' in data.columns: data, _ = CategoricalEncoder.label_encode(data, 'Genre')

    • if 'Shooting Location' in data.columns: data, _ = CategoricalEncoder.one_hot_encode(data, 'Shooting Location')

    • Date and time manipulation

      • if 'Release Date' in data.columns: data = DateTimeHandler.convert_to_datetime(data, 'Release Date', format='%d/%m/%Y') data = DateTimeHandler.extract_date_component(data, 'Release Date', 'year')
    • Feature engineering

      • if 'Budget in USD' in data.columns and 'Awards' in data.columns: data = FeatureEngineer.add_difference(data, 'Budget in USD', 'Awards', 'Budget_Awards_Diff')
      • if 'Rating' in data.columns and 'Popular' in data.columns: data = FeatureEngineer.add_product(data, 'Rating', 'Popular', 'Rating_Times_Popular')
  • Data type conversion

    • if 'Rating' in data.columns: data = DataTypeConverter.to_numeric(data, 'Rating')
    • if 'Genre' in data.columns: data = DataTypeConverter.to_categorical(data, 'Genre')

    print(data.head())

Licence

  • This project is licensed under the MIT License.

Contact

Installation

  • To install the library, use the following command:
pip install data_preprocessing_lib_rbb

Veri Ön İşleme Kütüphanesi

Python'da bir veri ön işleme kütüphanesi. Bu kütüphane, veri temizleme, dönüştürme ve manipülasyon işlemleri için çeşitli fonksiyonlar sunar, veri ön işleme görevlerinizi daha kolay ve verimli hale getirmek için tasarlanmıştır.

Özellikler

  • Handling Missing Values: Ortalama, medyan, sabit deÄŸer ile doldurma ve silme.
  • Outlier Detection and Correction: IQR yöntemi ile eÅŸik deÄŸer kullanarak aykırı deÄŸer tespiti ve düzeltilmesi.
  • Data Standardization and Normalization: Min-Max ve Standart normalizasyon iÅŸlemleri.
  • Text Cleaning and Manipulation: Durdurma kelimelerini çıkarma, küçük harfe çevirme, noktalama iÅŸaretlerini çıkarma, kelimeyi köküne indirgeme (NLTK kütüphanesinin kullanımını gerektirir).
  • Feature Engineering: Mevcut özelliklerden yeni özellikler oluÅŸturma.
  • Data Type Conversion: Sayısal, kategorik ve gün/saat dönüşümleri.
  • Encoding Categorical Data: One-hot kodlama ve label kodlama.
  • Date and Time Manipulation: Tarih bileÅŸenlerini çıkarma, tarih farklarını hesaplama.

Kullanım

Kütüphaneyi nasıl kullanacağınıza dair bir örnek:

  • Veriyi yükle

    • data = pd.read_csv('synthetic_sample_data.csv')
  • Eksik deÄŸerlerin iÅŸlenmesi

    • missing_handler = MissingValueHandler(strategy="mean")
    • data = missing_handler.fit_transform(data)
  • Aykırı deÄŸerlerin tespiti ve düzeltilmesi

    • outlier_handler = OutlierHandler(method="iqr", threshold=1.5)
    • data = outlier_handler.fit_transform(data)
  • # Verinin standartlaÅŸtırılması

    • scaler = Scaler(method="standard")
    • data = scaler.fit_transform(data)
  • Metin temizleme

    • text_cleaner = TextCleaner(remove_stopwords=True, lemmatize=True)
      • if 'Summary' in data.columns: data['Summary'] = data['Summary'].astype(str).apply(text_cleaner.clean)
  • Kategorik verilerin kodlanması

    • if 'Genre' in data.columns: data, _ = CategoricalEncoder.label_encode(data, 'Genre')

    • if 'Shooting Location' in data.columns: data, _ = CategoricalEncoder.one_hot_encode(data, 'Shooting Location')

    • Tarih ve zaman manipülasyonu

      • if 'Release Date' in data.columns: data = DateTimeHandler.convert_to_datetime(data, 'Release Date', format='%d/%m/%Y') data = DateTimeHandler.extract_date_component(data, 'Release Date', 'year')
    • Özellik mühendisliÄŸi

      • if 'Budget in USD' in data.columns and 'Awards' in data.columns: data = FeatureEngineer.add_difference(data, 'Budget in USD', 'Awards', 'Budget_Awards_Diff')
      • if 'Rating' in data.columns and 'Popular' in data.columns: data = FeatureEngineer.add_product(data, 'Rating', 'Popular', 'Rating_Times_Popular')
    • Veri tipi dönüşümü

      • if 'Rating' in data.columns: data = DataTypeConverter.to_numeric(data, 'Rating')
      • if 'Genre' in data.columns: data = DataTypeConverter.to_categorical(data, 'Genre')

      print(data.head())

Lisans

  • Bu proje MIT Lisansı altında lisanslanmıştır.

İletişim

Kurulum

  • Kütüphaneyi kurmak için aÅŸağıdaki komutu kullanın:
pip install data_preprocessing_lib_rbb

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

data_preprocessing_lib_rbb-0.0.1b0.tar.gz (12.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

data_preprocessing_lib_rbb-0.0.1b0-py3-none-any.whl (16.0 kB view details)

Uploaded Python 3

File details

Details for the file data_preprocessing_lib_rbb-0.0.1b0.tar.gz.

File metadata

File hashes

Hashes for data_preprocessing_lib_rbb-0.0.1b0.tar.gz
Algorithm Hash digest
SHA256 9843abf78ae3dd0387382ec6e5ab03ec50498574124fdbc0c0bb32e326376bfe
MD5 9c83a115a33ba06f5e4ecb009a6a8d5a
BLAKE2b-256 54883dfff1dea96f5ec36c6c058b6fa337f86579759726144788bfd8fd937e21

See more details on using hashes here.

File details

Details for the file data_preprocessing_lib_rbb-0.0.1b0-py3-none-any.whl.

File metadata

File hashes

Hashes for data_preprocessing_lib_rbb-0.0.1b0-py3-none-any.whl
Algorithm Hash digest
SHA256 55c162d1adbd2ba5c913cb97b43046c44732b279288931febeca642244a5f2f1
MD5 749ea9a6979a3a9b10d07e3640e76fa1
BLAKE2b-256 5410f4054c42daa29e5975e447b39cc1086d3477803dbf0765a6488a9a75d424

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page