Iris-Species-Classification

Implemented a Multi-class Logistic Regression model using PyTorch to classify Iris species. Achieved 96.67% test accuracy with a cross-entropy loss of 0.106. Features a full pipeline: Label Encoding, StandardScaler normalization, and the Adam optimizer. Optimized over 500 epochs.

Dataset Used: https://www.kaggle.com/datasets/uciml/iris

Input features

Sepal Length: The measurement of the outer leaf-like part of the flower from the base to the tip, expressed in centimeters.

Sepal Width: The measurement of the outer leaf-like part of the flower at its widest horizontal point, expressed in centimeters.

Petal Length: The measurement of the inner colorful leaf of the flower from the base to the tip, expressed in centimeters.

Petal Width: The measurement of the inner colorful leaf of the flower at its widest horizontal point, expressed in centimeters.

Target variable

The output feature of the Iris dataset is a categorical label representing the specific species of the flower: Iris Setosa, Iris Versicolor, or Iris Virginica.