Iris Flower Classification

CodeAlpha Data Science Internship — Task 1

Intern: Omokhoa Oshose Tosayoname
Intern ID: CA/DF1/71570
Duration: 20th May 2026 – 20th June 2026

---

Overview

This project trains and evaluates multiple machine learning classification models on the classic Iris dataset. The objective is to classify Iris flowers into three species based on four physical measurements.

Species:

• Iris setosa
• Iris versicolor
• Iris virginica

Features used:

• Sepal Length (cm)
• Sepal Width (cm)
• Petal Length (cm)
• Petal Width (cm)

---

Project Structure

CodeAlpha_IrisClassification/
├── data/
│   └── Iris.csv                   # Dataset
├── notebooks/
│   └── iris_classification.ipynb  # Main notebook
├── requirements.txt
└── README.md

---

Models Trained

Model	Description
K-Nearest Neighbours (KNN)	Instance-based learner; K=5 default, tuned up to K=20
Support Vector Machine (SVM)	RBF kernel; optimal for non-linear boundaries
Random Forest	Ensemble of 100 decision trees
Logistic Regression	Baseline linear classifier

---

Key Results

• All four models achieved very high accuracy (above 93%) on the test set.
• Petal Length and Petal Width are the most discriminative features.
• Iris setosa is perfectly linearly separable from the other two species.
• Iris versicolor and Iris virginica show slight overlap in sepal dimensions but are well separated by petal dimensions.

---

Visualisations Produced

• Class distribution bar chart
• Feature distributions by species (histograms)
• Pairplot of all feature combinations
• Correlation heatmap
• Box plots per feature per species
• Confusion matrices for all four models
• Model performance comparison chart
• Random Forest feature importance chart
• KNN accuracy vs. K tuning curve
• Decision boundary plots (KNN and SVM on petal features)

---

How to Run

1. Clone this repository:

   git clone https://github.com/Tosa9/CodeAlpha_IrisClassification.git
   cd CodeAlpha_IrisClassification

2. Install dependencies:

   pip install -r requirements.txt

3. Launch the notebook:

   jupyter notebook notebooks/iris_classification.ipynb

---

Dependencies

See requirements.txt for the full list.

---

Dataset Source

Iris Dataset — Kaggle

---

CodeAlpha Data Science Internship | Task 1
#CodeAlpha #DataScience #MachineLearning #IrisClassification