Skip to content

Latest commit

 

History

History
209 lines (150 loc) · 4.65 KB

user-guide.md

File metadata and controls

209 lines (150 loc) · 4.65 KB

User Guide

Installation and usage guide for InsightfulPy v0.2.0.

Table of Contents

Installation

Requirements

  • Python 3.8 or higher
  • pip

Install from PyPI

pip install insightfulpy

Install from Source

git clone https://github.com/dhaneshbb/insightfulpy.git
cd insightfulpy
pip install .

For development installation, see Development Setup.

Dependencies

Core dependencies include pandas, numpy, matplotlib, seaborn, scipy, and others. See pyproject.toml for complete list.

Quick Start

Import and Load Data

import pandas as pd
import insightfulpy as ipy

# Load your data
df = pd.read_csv('your_data.csv')

Get Help

ipy.help()         # Overview with function categories
ipy.list_all()     # List all functions
ipy.quick_start()  # Quick start guide
ipy.examples()     # Usage examples

Basic Analysis

# Dataset overview
ipy.columns_info('Sales Data', df)

# Numerical summary
num_stats = ipy.num_summary(df)
print(num_stats)

# Categorical summary
cat_stats = ipy.cat_summary(df)
print(cat_stats)

Core Functions

Dataset Overview

ipy.columns_info('Sales Data', df)  # Dataset structure
ipy.analyze_data(df)  # General analysis

Summary Statistics

ipy.num_summary(df)  # Numerical statistics
ipy.cat_summary(df)  # Categorical statistics
ipy.grouped_summary(df, groupby='category')  # Grouped summary

Data Quality Checks

# Missing and infinite values
ipy.missing_inf_values(df)
ipy.missing_inf_values(df, missing=True, df_table=True)

# Outlier detection
ipy.detect_outliers(df)
ipy.detect_outliers(df, max_display=20)

# Data type issues
ipy.detect_mixed_data_types(df)
ipy.cat_high_cardinality(df, threshold=100)

Visualizations

# Missing data patterns
ipy.show_missing(df)

# Distribution plots (batched)
ipy.kde_batches(df, batch_num=1)
ipy.box_plot_batches(df, batch_num=1)
ipy.qq_plot_batches(df, batch_num=1)
ipy.plot_boxplots(df)  # All columns

# Categorical plots (batched)
ipy.cat_bar_batches(df, batch_num=1, show_percentage=True)
ipy.cat_pie_chart_batches(df, batch_num=1)

# Relationship plots (batched)
ipy.num_vs_num_scatterplot_pair_batch(df, pair_num=0, batch_num=1)
ipy.cat_vs_cat_pair_batch(df, pair_num=0, batch_num=1)
ipy.num_vs_cat_box_violin_pair_batch(df, pair_num=0, batch_num=1)

Statistical Analysis

# Individual column analysis
ipy.num_analysis_and_plot(df, 'price')
ipy.num_analysis_and_plot(df, 'price', target='category')
ipy.cat_analyze_and_plot(df, 'category')

# Statistical calculations
ipy.calc_stats(df['price'])
ipy.iqr_trimmed_mean(df['price'])
ipy.mad(df['price'])
ipy.calculate_skewness_kurtosis(df)

Multi-Dataset Analysis

# Compare datasets
dfs = {'sales_2022': df1, 'sales_2023': df2}
base, linked = ipy.compare_df_columns('sales_2022', dfs)
ipy.linked_key(dfs)
ipy.display_key_columns('sales_2022', dfs)

# Interconnected outliers
ipy.interconnected_outliers(df, ['price', 'quantity', 'discount'])

# Comparison analysis
ipy.comp_num_analysis(df)
ipy.comp_num_analysis(df, missing_df=True)
ipy.comp_cat_analysis(df, missing_df=True)

Working with Batches

Batch functions split visualizations into manageable groups. Call without batch_num to see available batches:

batches = ipy.kde_batches(df)  # Shows batch info
ipy.kde_batches(df, batch_num=1)  # Plots batch 1

Environment Compatibility

InsightfulPy works in Jupyter, IPython, terminal, and scripts. DataFrames display automatically in interactive environments.

For scripts, add plt.show() to display plots:

import matplotlib.pyplot as plt
ipy.kde_batches(df, batch_num=1)
plt.show()

Examples

Complete workflow example available at docs/examples/example.ipynb:

cd docs/examples/ && jupyter notebook example.ipynb

See Also

Version: 0.2.0 | Status: Beta | Python: 3.8-3.12

Copyright 2025 dhaneshbb | License: MIT | Homepage: https://github.com/dhaneshbb/insightfulpy