The Iris dataset is a labeled collection of 150 flower records covering three iris species, with 50 samples per species. Each record contains flower properties such as petal length and sepal length; the species label is the prediction target. section 3.5.2
A practical workflow inspects and transforms the data before modeling:
- Load the CSV with pandas and check sample rows, shape, descriptive statistics, and variable spread. section 3.5.2
- Separate the four flower-property columns from the species target, then standardize the independent variables. section 3.5.2
- Use covariance analysis, eigenvalue decomposition, or scikit-learn PCA to identify principal components and their explained variance. section 3.5.2
Two-dimensional PCA view
Fit PCA to the standardized flower properties, retain two components, transform the observations, and plot the points using encoded species labels. The result keeps all 150 rows while reducing four variables to two. section 3.5.2
import numpy as np import pandas as pd import seaborn as sns import matplotlib.pyplot as plt from sklearn.decomposition import PCA from sklearn.preprocessing import StandardScaler
Use the supplied Python setup and notebook snippets to load, inspect, standardize, decompose, and visualize the dataset with pandas, NumPy, and scikit-learn. section 3.5.2
