data-autoeda 2.0.0

Automatic Exploratory Data Analysis, Cleaning, Validation, Visualization, and Smart Insights on ANY dataset.

AutoEDA

A production-ready Python package that performs automatic Exploratory Data Analysis (EDA), Data Cleaning, Data Validation, Visualization, and Smart Insights on ANY dataset.

Features

• Multi-Format Support: Analyze .csv, .xlsx, .xls, and .json effortlessly.
• Dataset Health Score: Instantly see a 0-100 score with category (Excellent/Good/Fair/Poor).
• Chat With Dataset: Ask natural language questions about your data interactively.
• Executive Summary: Get a concise, manager-friendly overview of your dataset.
• Streamlit Dashboard: Launch an interactive web interface with autoeda dashboard.
• PDF Report Generation: Professional PDF report with tables, charts, and insights.
• Smart Insights Engine: Generates 10+ business-style data observations automatically.
• Cleaning Recommendations: Actionable, numbered cleaning steps for your specific data.
• Large Dataset Mode: Smartly samples and optimizes memory for files >100MB or >100k rows.
• Cleaned Data Export: Export cleaned data straight from the CLI.
• Performance Optimized: Optional Polars backend for lightning-fast loading of large datasets.
• Visualizations: Automatically generates relevant charts using Matplotlib and Seaborn.
• Rich Terminal UI: Beautiful, organized CLI reports.

Installation

# Basic installation
pip install .

# Installation with Polars backend for large datasets
pip install .[fast]

Usage

CLI

# Complete analysis
autoeda data.csv
autoeda data.xlsx
autoeda data.json

# Clean data and export to cleaned_data.csv
autoeda data.csv --clean

# Generate a PDF report
autoeda data.csv --report

# Executive summary
autoeda data.csv --summary

# Chat with your dataset
autoeda data.csv --ask

# Only visualizations
autoeda data.csv --visualize

# Run everything
autoeda data.csv --all

# Launch Interactive Dashboard
autoeda dashboard

Python API

from autoeda.cli import analyze

# Complete analysis
df = analyze("data.csv")

# Executive summary only
analyze("data.csv", summary=True)

# Clean data
df_clean = analyze("data.csv", clean=True)