AttributeError: ‘numpy.ndarray’ object has no attribute ‘score’ error

I have tried to look for a problem but there is nothing Im seeing wrong here. What could it be? This is for trying binary classification in SVM for the fashion MNIST data set but only classifying 5 and 7.

import pandas as pd
import numpy as np
import seaborn as sns
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC
from sklearn import svm
from sklearn.preprocessing import MinMaxScaler
from sklearn.svm import SVR
from sklearn.model_selection import KFold
import matplotlib.pyplot as plt

trainset = 'mnist_train.xlsx'
trs = pd.read_excel(trainset)
testset = 'mnist_test.xlsx'
tes = pd.read_excel(testset)
xtrain = trs.iloc[:, [1, 783]]
ytrain = trs.iloc[:, 0]
xtest = tes.iloc[:, [1, 783]]
ytest = tes.iloc[:, 0]


##Linear SVC

svclassifier = SVC(kernel='linear', C=1)
svclassifier.fit(xtest, ytest)
ypred = svclassifier.predict(xtest)
print(ypred.score(xtrain, ytrain))
print(ypred.score(xtest, ytest))

##Gaussian SVC

svclassifier = SVC(kernel='rbf', C=1)
svclassifier.fit(xtrain, ytrain)
ypred = svclassifier.predict(xtest)
print(ypred.score(xtrain, ytrain))
print(ypred.score(xtest, ytest))

Answer

ypred is an array of predicted class labels, so the exception makes sense.

What you should do is use the classifier’s score method:

svclassifier = SVC(kernel='rbf', C=1)
svclassifier.fit(xtrain, ytrain)
# ypred = svclassifier.predict(xtest)  # We don’t actually use this. 
print(svclassifier.score(xtrain, ytrain))
print(svclassifier.score(xtest, ytest))

Advertisement

Answer