How to reshape a numpy array into 3 dimensional for input to convolutional layer?

Question

I have x_train array which is of 235 elements and each element is a numpy array of rows 125 and 125 columns . I neeed this array to be in shape of (235,125,125) so as to feed to my convolutional neural network model ((((similar to case when we need to feed MNIST digit array of (6000,28,28) where we have 60000 samples each of which is 28x28 array)))

Can someone tell how to reshape the x_train array to (235,125,125) ?

P.S I initially converted x_train from pandas series to numpy array below is the snapshot

Josh Friedlander · Accepted Answer · 2020-02-16 13:04:49Z

1

You need to use Numpy's reshape command:

x_train = x_train.reshape(x_train.shape[0], 125, 125, 1)

Have a look at this part of Keras' MNIST CNN tutorial.

answered Feb 16, 2020 at 13:04

Josh Friedlander

11.8k7 gold badges42 silver badges89 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

George Pipis · Accepted Answer · 2021-05-11 14:06:39Z

1

You can use the np.newaxis as follows:

x_train [...,np.newaxis]

And you will the shape will be:

x_train [...,np.newaxis].shape
(235, 125, 125, 1)

answered May 11, 2021 at 14:06

George Pipis

1,89225 silver badges14 bronze badges

Collectives™ on Stack Overflow

How to reshape a numpy array into 3 dimensional for input to convolutional layer?

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related