Why is the reshape working for my train data but not the test data?

Asked Oct 19 '20 at 04:48

Active Oct 19 '20 at 05:39

Viewed 211 times

Here I am using Fahion-mnist data set. Its already divided into test and train. I want each img to be a 28x28x3 shape.

X_train shape:  (60000, 784)

X_test shape:  (10000, 784)

# Reshaping
X_train = X_train.values.reshape(-1,28,28,3) 

print("X_train shape: ",X_train.shape) 

**output:** (20000, 28, 28, 3)

Works ok BUT !!

X_test = X_test.values.reshape(-1,28,28,3)
print("X_test shape: ",X_test.shape)

ValueError                                Traceback (most recent call last)
<ipython-input-110-9f1d9f52d152> in <module>()
1 X_test = X_test.values.reshape(-1,28,28,3)
2 print("X_test shape: ",X_test.shape)

 **ValueError: cannot reshape array of size 7840000 into shape (28,28,3)**

pls pls help me

I am following this, but i have removed the gray Scale part. https://www.kaggle.com/orhansertkaya/convolutional-neural-network-cnn-fashion-mnist

edited Oct 19 '20 at 05:39

asked Oct 19 '20 at 04:48

Exotic Me

Does this answer your question? [Cannot reshape array of size into shape](https://stackoverflow.com/questions/60028853/cannot-reshape-array-of-size-into-shape) – Ravi Saroch Oct 19 '20 at 04:53
No, as a tuple it gives similar error – Exotic Me Oct 19 '20 at 05:05
ValueError: cannot reshape array of size 7840000 into shape (3,28,28) – Exotic Me Oct 19 '20 at 05:05
is the total number of elements in your array a multiple of 3 x 28 x 28? – Paul H Oct 19 '20 at 05:54
original shape is: X_test shape: (10000, 784) – Exotic Me Oct 19 '20 at 14:06

Why is the reshape working for my train data but not the test data?

0 Answers0