NNCase producing padding layer for keras model irrespective of padding=valid or same

I am trying to create a model using keras 2 which uses tensor flow as back end.

Its a very simple code to test nncase conversion, nncase conversion graph is including a padding operation before the first convolution layer - when I use padding=valid or same, what needs to be done to avoid this padding operation

# Initializing the CNN
classifier = Sequential()

#lossbranch_1 = Sequential()
# First convolution layer C1
classifier.add(Convolution2D(16, (3, 3),strides=(2,2),padding='valid',dilation_rate=(1, 1),use_bias=True, input_shape=(240, 320, 3)))
classifier.add(Activation('relu'))
# Second convolution layer C2
c2 = Convolution2D(16, (3, 3),strides=(1,1),padding='same',use_bias=True, activation='relu')
classifier.add(c2)

kkfirstmodel.zip (10.6 KB) firstmodel5.pdf (16.4 KB)


here it is