About Reshape Usage in keras

from  import Reshape
layer_1 = Reshape((height, width, chns))( layer1)

from keras import backend as K
( layer1,(-1,2,4,8) ) 

from  import Sequential
from  import Reshape

model = Sequential()
# Change data shape to 3 rows and 4 columns
# Layer 1 of the model must specify the dimensions of the inputs, note that the size of the batch does not need to be specified
(Reshape((3, 4), input_shape=(12, )))
# Change data shape to 6 rows and 2 columns
(Reshape((6, 2)))
# Change the shape of the data to (2,2) in the 2nd and 3rd dimensions, automatically determining the size of the 1st dimension to be 3 based on the number of data elements.
(Reshape((-1, 2, 2)))
# Change the shape of the data to (2,2) for the 1st and 2nd dimensions and automatically determine the size of the 3rd dimension to be 3 based on the number of data elements.
(Reshape((2, 2, -1)))
()

context_shape = K.int_shape(context) #(None, 7, 7, 32)
    #Change the second dimension to be 32 and automatically size the 1st dimension to none based on the number of data elements?
    context = ((-1, context_shape[-1]))(context) #Tensor shape=(None, None, 32), dtype=float32