How does UpSampling2D layer work in Keras?

Question

How does the UpSampling2D layer work in Keras? According to official documentation:

Repeats the rows and columns of the data by size[0] and size[1] respectively.

So, if size=(2, 2), how does it repeat the rows and columns of the input matrix? Could you please explain the procedure with an example?

Answer 1

If

Repeats the rows and columns of the data by size[0] and size[1] respectively.

does not help, then maybe an example would be helpful:

>>> import numpy as np
>>> from keras.layers import UpSampling2D
>>> from keras.models import Sequential
>>> model = Sequential()
>>> model.add(UpSampling2D(size=(2,2), input_shape=(3,3,1)))

>>> x = np.arange(9).reshape(1,3,3,1)
>>> x[0,:,:,0]  # this is what x looks like initially
array([[0, 1, 2],
       [3, 4, 5],
       [6, 7, 8]])
>>> y = model.predict(x)
>>> y[0,:,:,0] # this is what it looks like after upsampling
array([[0., 0., 1., 1., 2., 2.],
       [0., 0., 1., 1., 2., 2.],
       [3., 3., 4., 4., 5., 5.],
       [3., 3., 4., 4., 5., 5.],
       [6., 6., 7., 7., 8., 8.],
       [6., 6., 7., 7., 8., 8.]], dtype=float32)