featurelens-ai 2.0.3

FeatureLens - AI-powered feature analysis library

FeatureLens

FeatureLens is a Python library for analyzing feature importance using statistical methods, machine learning, and optional AI-based explanations. It helps developers and data scientists understand which features contribute most to a target variable before model training.

Overview

Feature selection and understanding feature importance are critical steps in building effective machine learning models. FeatureLens simplifies this process by combining:

• Statistical correlation analysis
• Machine learning-based importance scoring
• Automated ranking of features
• Optional AI-generated explanations
• Visual insights for better interpretation

This allows users to quickly identify which features are most relevant for prediction tasks.

Features

• Correlation analysis between features and the target variable.
• Machine learning-based feature importance (e.g., Random Forest).
• Combined scoring system for ranking features.
• Clean and structured output report.
• Summary of the most and least important features.
• Optional visualization:
  • Correlation heatmap
  • Feature importance bar chart
• AI-based explanations for feature relevance (optional mode).

Installation

Install the latest version via pip:

pip install featurelens-ai==2.0.3

Usage

Basic Analysis

import pandas as pd
from feature_insight import analyze

# Load Data
df = pd.read_csv("data.csv")

# Run Analysis
analyze(df, df["target"])

Analysis With Graphs

To generate visual insights, simply set show_graphs=True.

import pandas as pd
from feature_insight import analyze

df = pd.read_csv("student_data.csv")

analyze(df, df["pass_exam"], show_graphs=True)

Expected Output

Feature Report

Feature Analysis Report

Feature: hours_studied
Score: 0.72
Insight: If a student studies more, they understand better and are more likely to pass.

Summary

Summary:
- hours_studied is the most important feature.
- practice_tests has moderate importance.
- sleep_hours has lower contribution.

Graphs (Optional)

When show_graphs=True, the library generates:

• Correlation Heatmap
• Feature Importance Chart
• Feature Distribution (top features)
• Feature vs Target Relationship

Input Requirements

• Data (data): Must be a pandas DataFrame or a valid CSV file path.
• Target (target): Must be a pandas Series.

Correct Usage:

analyze(df, df["buy"])

Incorrect Usage:

analyze(df, "buy")  # Passing a string for the target is not currently supported

Parameters

Parameter	Type	Description
data	DataFrame / str	The input dataset to analyze.
target	Series	The target column (pandas Series) to predict.
show_graphs	bool	Set to True to show visual charts and graphs.