cv.py

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
%matplotlib inline
import seaborn as sns


loan_data = pd.read_csv("loansData.csv")
loan_data.head()

# Clean Interest.Rate column
loan_data['Interest.Rate'] = loan_data['Interest.Rate'].map(lambda x: float(x.rstrip('%')))

# Clean Loan.Length column
loan_data['Loan.Length'] = loan_data['Loan.Length'].map(lambda x: x.strip(' months'))

# Clean and convert FICO.Range
loan_data['FICO.Score'] = loan_data['FICO.Range'].map(lambda x: int(x[:3]))

# Review data to make sure cleaning and conversions look correct
plt.figure()
p = loan_data['FICO.Score'].hist()
plt.show()

from sklearn.cross_validation import KFold

kf = KFold(len(loan_data), n_folds=10)