how to use lambda layer rebulid the input？

[1183710222]

base_model = models.att_unet_2d((128, 128, 3), filter_num=[64, 128, 256, 512], n_labels=1,
stack_num_down=2, stack_num_up=2, activation='ReLU',
atten_activation='ReLU', attention='add', output_activation='Sigmoid',
batch_norm=True, pool=False, unpool=False,
backbone='VGG16', weights='imagenet',
freeze_backbone=True, freeze_batch_norm=True,
name='attunet')
print(base_model.summary()) out put params =10M
but I used the lambda layer and tried to rebuild this shape
like this
inp = Input(shape=(128, 128, N))
l1 = Conv2D(3, (1, 1))(inp) # map N channels data to 3 channels
out = base_model(l1)
model = Model(inp, out, name=base_model.name)
print(model.summary()) out put params = 6
why？I sincerely seek for your assistance .