My accuracy for MNIST classification is coming out to be too high

#1

digits = load_digits()
X = digits.data
Y = digits.target
svc = svm.SVC()
cross_val_score(svc,X,Y,scoring=“accuracy”,cv=5).mean()
this is giving 96% accuracy

#2

Hey @shreyanarula3, Can you please share your code ?

#3

Capture
Capture760×547 45 KB
Capture
Capture760×547 45 KB

#4

Capture
Capture799×526 66.4 KB

#5

Hey @shreyanarula3, create a new notebook and just run the following code :

from sklearn.datasets import load_digits
from sklearn.model_selection import cross_val_score
from sklearn import svm
digits = load digits()
X = digits.data
Y = digits.target
svc = svm.SVC()
cross_val_score(svc,X,Y,scoring=‘accuracy’,cv=5).mean()

Output : 0.4487

I guess there was some error in the order in which you must have run the cells of your notebook.

Hope this helps :slight_smile: