The iris dataset contains 150 iris plants divided equally among three species. Each sample records sepal length, sepal width, petal length, and petal width, making it a compact classification problem. section 7.6
Its CSV representation separates four numeric parameters from a species label. Swift code shuffles the rows, converts labels into classification vectors, and prepares the data for a neural network. section 7.6
Classifying held-out flowers
A network with layer structure [4, 6, 3] trains on 140 samples for 20 passes, then validates the remaining 10. The demonstrated run identifies 9 of 10 correctly, or 90%. section 7.6
5.1,3.5,1.4,0.2,Iris-setosa 4.9,3.0,1.4,0.2,Iris-setosa 4.7,3.2,1.3,0.2,Iris-setosa 4.6,3.1,1.5,0.2,Iris-setosa 5.0,3.6,1.4,0.2,Iris-setosa
Shows the dataset's four measurements followed by its labeled species, making the tabular classification format concrete. section 7.6
