lazyanalyst 1.0.0

Automated data analysis library with end-to-end analysis pipeline

LazyAnalyst v1.0.0

Automated data analysis library for Python

LazyAnalyst is an end-to-end data analysis library that automates everything you'd do manually with Pandas and NumPy. Load a dataset, run one line of code, and get a complete analysis with insights, visualizations, statistical tests, and a professional HTML report.

Quick Start

import lazyanalyst as dp

# Analyze any CSV or Excel file
result = dp.analyze("sales_data.csv")

# Open the interactive dashboard
result.dashboard()

# Or view the professional report
result.report()

That's it! LazyAnalyst handles:

• Automated data loading and type detection
• Data quality auditing and reporting
• Intelligent data cleaning
• Exploratory data analysis
• Statistical testing (Pearson, Spearman, ANOVA, Chi-Square)
• Feature engineering
• Interactive Plotly dashboard
• Professional HTML report generation
• Automated insights and interpretations

Installation

Install via pip:

pip install lazyanalyst

Requirements: Python 3.8+

Or install from source:

git clone https://github.com/Tenali-Rama/lazyanalyst.git
cd lazyanalyst
pip install -e .

Features

1. Automated Pipeline

No configuration needed. Just provide a CSV or Excel file and LazyAnalyst handles the rest.

2. Data Quality Auditing

Automatically detects:

• Missing values
• Duplicate rows
• Outliers
• Data type inconsistencies
• Quality score calculation

3. Intelligent Cleaning

• Auto-detects and fixes common issues
• Handles missing values intelligently
• Removes duplicates
• Converts data types automatically

4. Exploratory Data Analysis (EDA)

• Summary statistics (mean, median, std, min, max)
• Distribution analysis
• Correlation detection
• Categorical value counts

5. Statistical Testing

Runs appropriate tests automatically:

• Pearson/Spearman Correlation for numerical relationships
• Independent T-Test for 2-group comparisons
• ANOVA for 3+ group comparisons
• Chi-Square for categorical relationships

6. Feature Engineering

• Polynomial features
• Interaction terms
• Scaled/normalized versions
• Log transforms for skewed data

7. Visualizations

Generates:

• Distribution histograms
• Categorical bar charts
• Correlation heatmaps
• Scatter plots for relationships

8. Interactive Dashboard

Beautiful, self-contained HTML dashboard with all analyses and charts.

9. Professional Report

PDF-ready HTML report with executive summary, findings, and visualizations.

Example Usage

Basic Analysis

import lazyanalyst as dp

result = dp.analyze("data.csv")
result.dashboard()  # Open interactive dashboard
result.report()      # Open HTML report

With Options

result = dp.analyze("data.xlsx", dashboard=True, report=True)

# Access cleaned data
cleaned_df = result.cleaned_data()

Supported File Types

• CSV (auto-detects encoding and delimiter)
• XLSX (Excel workbooks)

Output Files

LazyAnalyst creates an outputs/ folder with:

• cleaned_data.csv — Your cleaned dataset
• report.html — Professional report
• dashboard.html — Interactive dashboard
• insights.txt — Text summary of insights
• plots/ — All generated visualizations

Architecture

LazyAnalyst consists of 11 integrated modules:

1. loader.py — File loading with auto type inference
2. schema.py — Column type detection
3. quality.py — Data quality auditing
4. cleaner.py — Automated data cleaning
5. eda.py — Exploratory data analysis
6. visualizer.py — Chart generation
7. features.py — Feature engineering
8. stats.py — Statistical testing
9. insights.py — Natural language insights
10. dashboard.py — Interactive dashboard generation
11. reporter.py — HTML report generation

Documentation

Full documentation available in the GitHub repository.

License

MIT License - See LICENSE file for details

Troubleshooting

"FileNotFoundError"

• Check file path is correct
• Use absolute path if relative path doesn't work

"ValueError: unsupported file type"

• Ensure file is .csv or .xlsx

Dashboard won't open

• Check plotly and dash are installed
• Try opening dashboard.html directly in browser

Support

For questions or issues, check this README or the GitHub repository.

---

Transform your data analysis workflow with one line of code.