IRIS Data Classification
We will do the following steps to classfy iris data:
- load iris data
- create feature
- build model and fit data
- predict
Requiremetns
- pandas
- numpy
- sklearn
Loading IRIS dataset
import pandas as pd
df = pd.read_csv('iris.csv')
print(df.head())
print(df.describe())
Create Feature from Loaded Data
import numpy as np
X = []
y = []
for sl, sw, pl, pw in zip(df['sepal length'], df['sepal width'], df['petal length'], df['petal width']):
x = [sl, sw, pl, pw]
X.append(x)
for label in df['species']:
if label =='se' or label == 'setosa':
y.append(0)
elif label=='versicolor':
y.append(1)
else:
y.append(2)
X = np.array(X)
y = np.array(y)
Build Model and Fit Data
from sklearn.svm import SVC
model = SVC()
model.fit(X, y)
It’s done training. Let predict
Predict
model.predict([[1, 2, 3, 4]])
Comments