Where is "tf.keras.layers."?

Course Q&A Deep Learning Specialization Convolutional Neural Networks
1

Hi,

This is a very low-level question. I am in the hw of "
Residual_Networks". I found out the code is directly given like

    X = Conv2D(filters = F1, kernel_size = 1, strides = (1,1), padding = 'valid', kernel_initializer = initializer(seed=0))(X)

So, where is the “tf.keras.layers”. ? I tried

 X = tf.keras.layers.Conv2D(filters = F1, kernel_size = 1, strides = (1,1), padding = 'valid', kernel_initializer = initializer(seed=0))(X)

the code still works.

I assume this is something to do at the very beginning. Can someone show me which line made this possible? Thank you.

import tensorflow as tf
import numpy as np
import scipy.misc
from tensorflow.keras.applications.resnet_v2 import ResNet50V2
from tensorflow.keras.preprocessing import image
from tensorflow.keras.applications.resnet_v2 import preprocess_input, decode_predictions
from tensorflow.keras import layers
from tensorflow.keras.layers import Input, Add, Dense, Activation, ZeroPadding2D, BatchNormalization, Flatten, Conv2D, AveragePooling2D, MaxPooling2D, GlobalMaxPooling2D
from tensorflow.keras.models import Model, load_model
from resnets_utils import *
from tensorflow.keras.initializers import random_uniform, glorot_uniform, constant, identity
from tensorflow.python.framework.ops import EagerTensor
from matplotlib.pyplot import imshow

from test_utils import summary, comparator
import public_tests
2

Hi sunson29,

It’s this line:

from tensorflow.keras.layers import Input, Add, Dense, Activation, ZeroPadding2D, BatchNormalization, Flatten, Conv2D, AveragePooling2D, MaxPooling2D, GlobalMaxPooling2D

3

oh, all those “names” ( Input, Add, Dense, Activation, ZeroPadding2D, BatchNormalization, Flatten, Conv2D, AveragePooling2D, MaxPooling2D, GlobalMaxPooling2D
) after “import” can be directly called. Am I correct?

4

Yes, that’s right. Or you can for instance use tensorflow.keras.layers.Conv2D. It’s the same.

1 Like