SVM-iris.py

# This is a sample model to demonstrate how a Machine Learning model 
# can be implemented in Production as a REST API and how it can be consumed

# Import libraries and packages
from sklearn import svm, datasets
import pickle 
import numpy as np

# Load Sample data
iris = datasets.load_iris()

# Split loaded data into independent and target features
X = iris.data  
y = iris.target

# Train Support Vector Machine (SVM) model with all data 
svmModel = svm.SVC(kernel='poly', degree=3, C=1.0).fit(X, y)

# Persist model so that it can be used by different consumers
svmFile = open('SVMModel.pckl', 'wb')
pickle.dump(svmModel, svmFile)
svmFile.close()