Keras - Cnn Model Summary Diemension Interpretation

I am using Keras library to build this deep learning model: INPUT(depth=1, height=15, width=27) -> CONV[depth=8](height=4, width=27) -> POOL(height=2, width=1) -> (Regress

Solution 1:

Change border_mode='same' to border_mode='valid'. Border mode same adds zero padding to the input to make sure that the output of the convolutional layer has the same shape as its input. With border mode valid convolution is performed only where the input and the filter fully overlap.