CODEWITHSUNDEEP
pythonarraysnumpyscikit-learn
Amaury Leproux
Amaury Leproux

Reputation: 229

Using the score method from Sklearn and get ValueError: multiclass-multioutput is not supported

This is my code:

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.datasets import fetch_mldata
from sklearn import neighbors
from sklearn.model_selection import train_test_split
mnist = fetch_mldata('MNIST original')

sample = np.random.randint(70000, size=5000)
data = mnist.data[sample]
target = mnist.data[sample]

xtrain, xtest, ytrain, ytest = train_test_split(data, target, train_size=0.8)

knn = neighbors.KNeighborsClassifier(n_neighbors=3)
knn.fit(xtrain, ytrain)

error = 1 - knn.score(xtest, ytest)
print('Erreur: %f' % error)

when I run "python numb.py" I got this message error:

File "/anaconda/lib/python2.7/site-packages/sklearn/metrics/classification.py", line 88, in _check_targets
    raise ValueError("{0} is not supported".format(y_type))
ValueError: multiclass-multioutput is not supported

Upvotes: 3

Views: 1673

Answers (1)

P. Camilleri
P. Camilleri

Reputation: 13218

This is a simple typo. ytest has the wrong shape, because you should write

target = mnist.target[sample]

Correcting this, the script runs fine.

Also, the way you build sample, you may have duplicates in it, which means some images may be both in test and train set. Consider using np.random.permutation to shuffle the order of your samples. And consider using a seed before calling np.random, to get reproducible results (or better, use check_random_state from sklearn)

Upvotes: 4

Related Questions