F# Applications in Data Science

A collection of projects demonstrating functional programming approaches to data science using F#. This repository showcases how F#'s expressive syntax, type safety, and functional paradigms make it an excellent choice for data analysis, transformation, and visualization workflows.

🔗 View Live Demo → — Interactive website showcasing the project, visualizations, and code examples.

Why F# for Data Science?

F# brings unique advantages to data science that complement traditional tools like Python and R:

Feature	Benefit
Type Inference	Catch data type errors at compile time, not runtime
Pipeline Operators	Chain transformations naturally with |>
Immutability by Default	Safer data transformations without side effects
Pattern Matching	Elegant handling of missing data and edge cases
REPL Support	Interactive exploration via F# Interactive (FSI)
.NET Ecosystem	Access to mature libraries and enterprise integration

---

Projects

🐧 Penguin Data Analysis

A complete data analysis pipeline exploring morphological measurements of penguins from the Palmer Archipelago. This project demonstrates the full data science workflow in F#.

What it demonstrates:

• Loading and exploring CSV datasets
• Data cleaning and preprocessing
• Handling missing values functionally
• Outlier detection and removal
• Feature engineering
• Interactive data visualization

→ View Project Details

---

Core Concepts Demonstrated

1. Data Loading with Deedle

Deedle is F#'s answer to pandas—a powerful data frame library for structured data manipulation.

open Deedle

// Load CSV into a typed data frame
let df = Frame.ReadCsv("penguins.csv", hasHeaders = true)
printfn "Rows: %d, Columns: %d" df.RowCount df.ColumnCount

2. Functional Data Transformation

F#'s pipeline operator (|>) enables readable, chainable data transformations:

let cleanedData = 
    rawData
    |> Frame.filterRows (fun _ row -> hasValidMeasurements row)
    |> Frame.mapCols (fun name col -> handleMissingValues name col)
    |> Frame.filterRows (fun _ row -> isWithinValidRange row)

3. Safe Missing Value Handling

Pattern matching and OptionalValue<T> provide compile-time safe handling of missing data:

let billLength = row.TryGetAs<float>("bill_length_mm")
let billDepth = row.TryGetAs<float>("bill_depth_mm")

// Count valid measurements using functional composition
let validCount = 
    [billLength.HasValue; billDepth.HasValue; flipperLength.HasValue; bodyMass.HasValue]
    |> List.filter id
    |> List.length

4. Interactive Visualization with Plotly.NET

Plotly.NET brings interactive, publication-quality visualizations to F#:

open Plotly.NET

Chart.Scatter(x = flipperData, y = massData, mode = StyleParam.Mode.Markers)
|> Chart.withTitle "Flipper Length vs Body Mass"
|> Chart.withXAxisStyle "Flipper Length (mm)"
|> Chart.withYAxisStyle "Body Mass (g)"
|> Chart.saveHtml("plots/flipper_vs_mass.html")

5. Feature Engineering

Create derived features using Deedle's expressive column operations:

// Create bill area as a derived feature
let billAreaSeries = df?bill_length_mm * df?bill_depth_mm
df.AddColumn("bill_area_mm2", billAreaSeries)

---

Getting Started

Prerequisites

• .NET 9.0 SDK or later
• F# compiler (included with .NET SDK)

Quick Start

# Clone the repository
git clone https://github.com/SiD-array/F-applications.git
cd F-applications

# Run the Penguin Data Analysis project
cd PenguinDataDemo
dotnet restore
dotnet run

---

Project Structure

F-applications/
├── PenguinDataDemo/              # Penguin data analysis project
│   ├── src/
│   │   ├── DataLoad.fs           # Data loading utilities
│   │   ├── DataClean.fs          # Data cleaning & preprocessing
│   │   └── Visualization.fs      # Plotting & visualization
│   ├── plots/                    # Generated HTML visualizations
│   ├── Program.fs                # Main entry point
│   ├── penguins.csv              # Palmer Penguins dataset
│   └── PenguinDataDemo.fsproj    # Project configuration
├── docs/                         # Project showcase website
│   ├── index.html                # Main webpage
│   ├── style.css                 # Styling
│   ├── script.js                 # Interactions
│   └── plots/                    # Embedded visualizations
├── README.md                     # This file
└── .gitignore

---

Technology Stack

Library	Version	Purpose
F#	.NET 9.0	Functional-first programming language
Deedle	3.0.0	Data frames & series manipulation
FSharp.Data	6.6.0	Type providers for data access
Plotly.NET	5.1.0	Interactive visualizations
Plotly.NET.Interactive	5.0.0	Notebook integration

---

Learning Resources

F# for Data Science

• FsLab - F# data science packages
• F# for Fun and Profit - Comprehensive F# tutorials
• Deedle Documentation - Data frame operations

Visualization

• Plotly.NET Documentation - Chart types and customization
• XPlot - Alternative plotting library

---

Roadmap

Future projects planned for this repository:

• Time Series Analysis - Stock price prediction with ML.NET
• Statistical Analysis - Hypothesis testing and regression
• Machine Learning - Classification with ML.NET and F#
• Data Pipeline - ETL workflows with type providers

---

Contributing

Contributions are welcome! Whether it's:

• 🐛 Bug fixes
• 📝 Documentation improvements
• ✨ New data science examples
• 🎨 Visualization enhancements

Please feel free to submit a Pull Request.

---

License

This project is open source and available for educational purposes.

---

Author

SiD-array - GitHub Profile

---

Functional programming meets data science 📊