Cross-validation: Leave One Out

Leave One Out is a special type of K-Fold split where the number of folds (k) is equal to the number of observations(rows) in your dataset.

Note: K-Fold divides the data/rows into k-parts also known as folds. The variance(standard deviation) for LeaveOneOut is higher than regular K-Fold split.

Cross-validation can be done with the cross_val_score() helper function on the estimator(LogisticRegression), dataset and split technique(LeaveOneOut)

  • cross_val_score() returns scores of the estimator for each fold

This recipe includes the following topics:

  • Load data/file from github
  • Split columns into the usual feature columns(X) and target column(Y)
  • Split data using LeaveOneOut() class
  • Instantiate a classification model (LogisticRegression)
  • Call cross_val_score() to run cross validation
  • Calculate mean and standard deviation from scores returned by cross_val_score()

# import modules
import pandas as pd
import numpy as np
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import LeaveOneOut
from sklearn.model_selection import cross_val_score

# read data file from github
# dataframe: pimaDf
gitFileURL = ''
cols = ['preg', 'plas', 'pres', 'skin', 'test', 'mass', 'pedi', 'age', 'class']
pimaDf = pd.read_csv(gitFileURL, names = cols)

# convert into numpy array for scikit-learn
pimaArr = pimaDf.values

# Let's split columns into the usual feature columns(X) and target column(Y)
# Y represents the target 'class' column whose value is either '0' or '1'
X = pimaArr[:, 0:8]
Y = pimaArr[:, 8]

# split data using LeaveOneOut
cvloo = LeaveOneOut()

# instantiate a classification model
model = LogisticRegression()

# call cross_val_score() to run cross validation
resultArr = cross_val_score(model, X, Y, cv=cvloo)

# calculate mean of scores for all folds
meanAccuracy = resultArr.mean() * 100

# calculate standard deviation of scores for all folds
stdAccuracy = resultArr.std() * 100

# display accuracy
print("Mean accuracy: %.3f%%, Standard deviation: %.3f%%" % (meanAccuracy, stdAccuracy))
Mean accuracy: 76.823%, Standard deviation: 42.196%

Leave a Reply

Your email address will not be published. Required fields are marked *