easyclean 0.1.0

Fast, simple, chainable data cleaning for pandas DataFrames

easyclean

Fast, simple, chainable data cleaning for pandas DataFrames — built for data analysts who want clean data in a few lines instead of a wall of boilerplate.

Install

pip install easyclean

Quick start

import pandas as pd
from easyclean import clean

df = pd.read_csv("messy_data.csv")
df_clean = clean(df)

clean() standardizes column names, strips whitespace, drops duplicate rows, infers proper dtypes, and fills missing values — in one call.

Chainable API (for more control)

from easyclean import Cleaner

cleaner = Cleaner(df)
result = (
    cleaner
    .standardize_columns()
    .strip_strings()
    .drop_duplicates()
    .handle_missing(strategy="median")
    .infer_dtypes()
    .remove_outliers()
    .df
)

print(cleaner.report())

Example report output:

EasyClean Report
=================
Original shape: (1000, 8)
Final shape:    (974, 8)
Missing values: 42 -> 0

Steps applied:
  - Standardized column names (3 renamed)
  - Stripped whitespace from text columns
  - Dropped 12 duplicate row(s)
  - Handled missing values (strategy='median'): 42 value(s) filled/removed
  - Inferred dtypes (2 column(s) converted)
  - Removed 14 outlier row(s) (IQR factor=1.5)

Functions

Function	What it does
clean(df)	One-liner full pipeline
standardize_column_names(df)	Lowercase, strip, snake_case columns
strip_whitespace(df)	Trim whitespace in text columns
drop_duplicate_rows(df)	Remove exact duplicate rows
handle_missing(df, strategy=...)	Fill/drop missing values ("auto", "mean", "median", "mode", "drop", "value")
infer_dtypes(df)	Convert numeric-looking text columns to numbers
detect_outliers_iqr(df)	Boolean mask of IQR-based outliers
remove_outliers_iqr(df)	Drop IQR-based outlier rows

License

MIT