DeepLearning.ai深度学习课程笔记
  • Introduction
  • 第一门课 神经网络和深度学习(Neural-Networks-and-Deep-Learning)
    • 第一周：深度学习引言(Introduction to Deep Learning)
      • 1.1 神经网络的监督学习(Supervised Learning with Neural Networks)
      • 1.2 为什么神经网络会流行？(Why is Deep Learning taking off?)
    • 第二周：神经网络的编程基础(Basics of Neural Network programming)
      • 2.1 二分类(Binary Classification)
      • 2.2 逻辑回归(Logistic Regression)
      • 2.3 逻辑回归的代价函数（Logistic Regression Cost Function）
      • 2.4 逻辑回归的梯度下降（Logistic Regression Gradient Descent）
      • 2.5 梯度下降的例子(Gradient Descent on m Examples)
      • 2.6 向量化 logistic 回归的梯度输出（Vectorizing Logistic Regression’s Gradient Output）
      • 2.7 （选修）logistic 损失函数的解释（Explanation of logistic regression cost function ）
      • Logistic Regression with a Neural Network mindset 代码
      • lr_utils.py
    • 第三周：浅层神经网络(Shallow neural networks)
      • 3.1 神经网络概述（Neural Network Overview）
      • 3.2 神经网络的表示（Neural Network Representation ）
      • 3.3 计算一个神经网络的输出（Computing a Neural Network's output ）
      • 3.4 多样本向量化（Vectorizing across multiple examples ）
      • 3.5 激活函数（Activation functions）
      • 3.6 为什么需要（ 非线性激活函数？（why need a nonlinear activation function?）
      • 3.7 激活函数的导数（Derivatives of activation functions ）
      • 3.8 神经网络的梯度下降（Gradient descent for neural networks）
      • 3.9 （选修）直观理解反向传播（Backpropagation intuition ）
      • 3.10 随机初始化（Random+Initialization）
      • Planar data classification with one hidden layer
      • planar_utils.py
      • testCases.py
    • 第四周：深层神经网络(Deep Neural Networks)
      • 4.1 深层神经网络（Deep L-layer neural network）
      • 4.2 前向传播和反向传播（Forward and backward propagation）
      • 4.3 深层网络中的前向传播（Forward propagation in a Deep Network ）
      • 4.4 为什么使用深层表示？（Why deep representations?）
      • 4.5 搭建神经网络块（Building blocks of deep neural networks）
      • 4.6 参数 VS 超参数（Parameters vs Hyperparameters）
      • Building your Deep Neural Network Step by Step
      • dnn_utils.py
      • testCases.py
      • Deep Neural Network Application
      • dnn_app_utils.py
  • 第二门课 改善深层神经网络：超参数调试、 正 则 化 以 及 优 化 (Improving Deep Neural Networks:Hyperparameter tuning, Regulariza
    • 第二门课 改善深层神经网络：超参数调试、正则化以及优化(Improving Deep Neural Networks:Hyperparameter tuning, Regularization and
      • 第一周：深度学习的实用层面(Practical aspects of Deep Learning)
        • 1.1 训练，验证，测试集（Train / Dev / Test sets）
        • 1.2 偏差，方差（Bias /Variance）
        • 1.3 机器学习基础（Basic Recipe for Machine Learning）
        • 1.4 正则化（Regularization）
        • 1.5 为什么正则化有利于预防过拟合呢？（Why regularization reduces overfitting?）
        • 1.6 dropout 正则化（Dropout Regularization）
        • 1.7 理解 dropout（Understanding Dropout）
        • 1.8 其他正则化方法（Other regularization methods）
        • 1.9 归一化输入（Normalizing inputs）
        • 1.10 梯度消失/梯度爆炸（Vanishing / Exploding gradients）
        • 1.11 神经网络的权重初始化（Weight Initialization for Deep Networks）
        • 1.12 梯度的数值逼近（Numerical approximation of gradients）
        • 1.13 梯度检验（Gradient checking）
        • 1.14 梯度检验应用的注意事项（Gradient Checking Implementation Notes）
        • Initialization
        • Gradient Checking
        • Regularization
        • reg_utils.py
        • testCases.py
      • 第二周：优化算法 (Optimization algorithms)
        • 2.1 Mini-batch 梯度下降（Mini-batch gradient descent）
        • 2.2 理解 mini-batch 梯度下降法（Understanding mini-batch gradient descent）
        • 2.3 指数加权平均数（Exponentially weighted averages）
        • 2.4 理解指数加权平均数（Understanding exponentially weighted averages ）
        • 2.5 指 数 加 权 平 均 的 偏 差 修 正 （ Bias correction in exponentially weighted averages ）
        • 2.6 动量梯度下降法（Gradient descent with Momentum ）
        • 2.7 RMSprop( root mean square prop)
        • 2.8 Adam 优化算法(Adam optimization algorithm)
        • 2.9 学习率衰减(Learning rate decay)
        • 2.10 局部最优的问题(The problem of local optima)
        • Optimization
        • opt_utils.py
        • testCases.py
      • 第 三 周 超 参 数 调 试 、 Batch 正 则 化 和 程 序 框 架 （Hyperparameter tuning）
        • 3.1 调试处理（Tuning process）
        • 3.2 为超参数选择合适的范围（Using an appropriate scale to pick hyperparameters）
        • 3.3 超参数训练的实践： Pandas VS Caviar（Hyperparameters tuning in practice: Pandas vs. Caviar）
        • 3.4 归一化网络的激活函数（ Normalizing activations in a network）
        • 3.5 将 Batch Norm 拟合进神经网络（Fitting Batch Norm into a neural network）
        • 3.6 Batch Norm 为什么奏效？（Why does Batch Norm work?）
        • 3.7 测试时的 Batch Norm（Batch Norm at test time）
        • 3.8 Softmax 回归（Softmax regression）
        • 3.9 训练一个 Softmax 分类器（Training a Softmax classifier）
        • tensorflow tutorial
        • improv_utils.py
        • tf_utils.py
  • 第三门课 结构化机器学习项目（Structuring Machine Learning Projects）
    • 第三门课 结构化机器学习项目（Structuring Machine Learning Projects）
      • 第一周 机器学习（ML）策略（1）（ML strategy（1））
        • 1.1 为什么是 ML 策略？（Why ML Strategy?）
        • 1.2 正交化（Orthogonalization）
        • 1.3 单一数字评估指标（Single number evaluation metric）
        • 1.4 满足和优化指标（Satisficing and optimizing metrics）
        • 1.5 训练/开发/测试集划分（Train/dev/test distributions）
        • 1.6 开发集和测试集的大小（Size of dev and test sets）
        • 1.7 什么时候该改变开发/测试集和指标？（When to change dev/test sets and metrics）
        • 1.8 为什么是人的表现？（ Why human-level performance?）
        • 1.9 可避免偏差（Avoidable bias）
        • 1.10 理解人的表现（Understanding human-level performance）
        • 1.11 超过人的表现（Surpassing human- level performance）
        • 1.12 改善你的模型的表现（Improving your model performance）
      • 第二周：机器学习策略（2）(ML Strategy (2))
        • 2.1 进行误差分析（Carrying out error analysis）
        • 2.2 清楚标注错误的数据（Cleaning up Incorrectly labeled data）
        • 2.3 快速搭建你的第一个系统，并进行迭代（Build your first system quickly, then iterate）
        • 2.4 在不同的划分上进行训练并测试（Training and testing on different distributions）
        • 2.5 不匹配数据划分的偏差和方差（Bias and Variance with mismatched data distributions）
        • 2.6 定位数据不匹配（Addressing data mismatch）
        • 2.7 迁移学习（Transfer learning）
        • 2.8 多任务学习（Multi-task learning）
        • 2.9 什么是端到端的深度学习？（What is end-to-end deep learning?）
        • 2.10 是否要使用端到端的深度学习？（Whether to use end-to-end learning?）
  • 第四门课 卷积神经网络（Convolutional Neural Networks）
    • 第四门课 卷积神经网络（Convolutional Neural Networks）
      • 第一周 卷积神经网络（Foundations of Convolutional Neural Networks）
        • 1.1 计算机视觉（Computer vision）
        • 1.2 边缘检测示例（Edge detection example）
        • 1.3 更多边缘检测内容（More edge detection）
        • 1.4 Padding
        • 1.5 卷积步长（Strided convolutions）
        • 1.6 三维卷积（Convolutions over volumes）
        • 1.7 单层卷积网络（One layer of a convolutional network）
        • 1.8 简单卷积网络示例（A simple convolution network example）
        • 1.9 池化层（Pooling layers）
        • 1.10 卷积神经网络示例（Convolutional neural network example）
        • 1.11 为什么使用卷积？（Why convolutions?）
        • Convolution model Step by Step
        • Convolutional Neural Networks: Application
        • cnn_utils
      • 第二周 深度卷积网络：实例探究（Deep convolutional models: case studies）
        • 2.1 经典网络（Classic networks）
        • 2.2 残差网络（Residual Networks (ResNets)）
        • 2.3 残差网络为什么有用？（Why ResNets work?）
        • 2.4 网络中的网络以及 1×1 卷积（Network in Network and 1×1 convolutions）
        • 2.5 谷歌 Inception 网络简介（Inception network motivation）
        • 2.6 Inception 网络（Inception network）
        • 2.7 迁移学习（Transfer Learning）
        • 2.8 数据扩充（Data augmentation）
        • 2.9 计算机视觉现状（The state of computer vision）
        • Residual Networks
        • Keras tutorial - the Happy House
        • kt_utils.py
      • 第三周 目标检测（Object detection）
        • 3.1 目标定位（Object localization）
        • 3.2 特征点检测（Landmark detection）
        • 3.3 目标检测（Object detection）
        • 3.4 卷积的滑动窗口实现（Convolutional implementation of sliding windows）
        • 3.5 Bounding Box预测（Bounding box predictions）
        • 3.6 交并比（Intersection over union）
        • 3.7 非极大值抑制（Non-max suppression）
        • 3.8 Anchor Boxes
        • 3.9 YOLO 算法（Putting it together: YOLO algorithm）
        • 3.10 候选区域（选修）（Region proposals (Optional)）
        • Autonomous driving application - Car detection
        • yolo_utils.py
      • 第四周 特殊应用：人脸识别和神经风格转换（Special applications: Face recognition &Neural style transfer）
        • 4.1 什么是人脸识别？（What is face recognition?）
        • 4.2 One-Shot学习（One-shot learning）
        • 4.3 Siamese 网络（Siamese network）
        • 4.4 Triplet 损失（Triplet 损失）
        • 4.5 面部验证与二分类（Face verification and binary classification）
        • 4.6 什么是深度卷积网络？（What are deep ConvNets learning?）
        • 4.7 代价函数（Cost function）
        • 4.8 内容代价函数（Content cost function）
        • 4.9 风格代价函数（Style cost function）
        • 4.10 一维到三维推广（1D and 3D generalizations of models）
        • Art Generation with Neural Style Transfer
        • nst_utils.py
        • Face Recognition for the Happy House
        • fr_utils.py
        • inception_blocks.py
  • 第五门课 序列模型(Sequence Models)
    • 第五门课 序列模型(Sequence Models)
      • 第一周 循环序列模型（Recurrent Neural Networks）
        • 1.1 为什么选择序列模型？（Why Sequence Models?）
        • 1.2 数学符号（Notation）
        • 1.3 循环神经网络模型（Recurrent Neural Network Model）
        • 1.4 通过时间的反向传播（Backpropagation through time）
        • 1.5 不同类型的循环神经网络（Different types of RNNs）
        • 1.6 语言模型和序列生成（Language model and sequence generation）
        • 1.7 对新序列采样（Sampling novel sequences）
        • 1.8 循环神经网络的梯度消失（Vanishing gradients with RNNs）
        • 1.9 GRU单元（Gated Recurrent Unit（GRU））
        • 1.10 长短期记忆（LSTM（long short term memory）unit）
        • 1.11 双向循环神经网络（Bidirectional RNN）
        • 1.12 深层循环神经网络（Deep RNNs）
        • Building your Recurrent Neural Network
        • rnn_utils.py
        • Dinosaurus Island -- Character level language model final
        • utils.py
        • shakespeare_utils.py
        • Improvise a Jazz Solo with an LSTM Network
      • 第二周 自然语言处理与词嵌入（Natural Language Processing and Word Embeddings）
        • 2.1 词汇表征（Word Representation）
        • 2.2 使用词嵌入（Using Word Embeddings）
        • 2.3 词嵌入的特性（Properties of Word Embeddings）
        • 2.4 嵌入矩阵（Embedding Matrix）
        • 2.5 学习词嵌入（Learning Word Embeddings）
        • 2.6 Word2Vec
        • 2.7 负采样（Negative Sampling）
        • 2.8 GloVe 词向量（GloVe Word Vectors）
        • 2.9 情感分类（Sentiment Classification）
        • 2.10 词嵌入除偏（Debiasing Word Embeddings）
        • Operations on word vectors
        • w2v_utils.py
        • Emojify
        • emo_utils.py
      • 第三周 序列模型和注意力机制（Sequence models & Attention mechanism）
        • 3.1 基础模型（Basic Models）
        • 3.2 选择最可能的句子（Picking the most likely sentence）
        • 3.3 集束搜索（Beam Search）
        • 3.4 改进集束搜索（Refinements to Beam Search）
        • 3.5 集束搜索的误差分析（Error analysis in beam search）
        • 3.6 Bleu 得分（选修）（Bleu Score (optional)）
        • 3.7 注意力模型直观理解（Attention Model Intuition）
        • 3.8注意力模型（Attention Model）
        • 3.9语音识别（Speech recognition）
        • 3.10触发字检测（Trigger Word Detection）
        • Neural machine translation with attention
        • nmt_utils.py
        • Trigger word detection
        • td_utils.py
Powered by GitBook
On this page
  • 1 - The problem of very deep neural networks
  • 2 - Building a Residual Network
  • 2.1 - The identity block
  • 2.2 - The convolutional block
  • 3 - Building your first ResNet model (50 layers)
  • 4 - Test on your own image (Optional/Ungraded)
  • References

Was this helpful?

  1. 第四门课 卷积神经网络（Convolutional Neural Networks）
  2. 第四门课 卷积神经网络（Convolutional Neural Networks）
  3. 第二周 深度卷积网络：实例探究（Deep convolutional models: case studies）

Residual Networks

Previous2.9 计算机视觉现状（The state of computer vision）NextKeras tutorial - the Happy House

Last updated 6 years ago

Was this helpful?

Welcome to the second assignment of this week! You will learn how to build very deep convolutional networks, using Residual Networks (ResNets). In theory, very deep networks can represent very complex functions; but in practice, they are hard to train. Residual Networks, introduced by , allow you to train much deeper networks than were previously practically feasible.

In this assignment, you will:

  • Implement the basic building blocks of ResNets.

  • Put together these building blocks to implement and train a state-of-the-art neural network for image classification.

This assignment will be done in Keras.

Before jumping into the problem, let's run the cell below to load the required packages.

import numpy as np
from keras import layers
from keras.layers import Input, Add, Dense, Activation, ZeroPadding2D, BatchNormalization, Flatten, Conv2D, AveragePooling2D, MaxPooling2D, GlobalMaxPooling2D
from keras.models import Model, load_model
from keras.preprocessing import image
from keras.utils import layer_utils
from keras.utils.data_utils import get_file
from keras.applications.imagenet_utils import preprocess_input
import pydot
from IPython.display import SVG
from keras.utils.vis_utils import model_to_dot
from keras.utils import plot_model
from resnets_utils import *
from keras.initializers import glorot_uniform
import scipy.misc
from matplotlib.pyplot import imshow
%matplotlib inline


import keras.backend as K
K.set_image_data_format('channels_last')
K.set_learning_phase(1)

Using TensorFlow backend.

1 - The problem of very deep neural networks

Last week, you built your first convolutional neural network. In recent years, neural networks have become deeper, with state-of-the-art networks going from just a few layers (e.g., AlexNet) to over a hundred layers.

The main benefit of a very deep network is that it can represent very complex functions. It can also learn features at many different levels of abstraction, from edges (at the lower layers) to very complex features (at the deeper layers). However, using a deeper network doesn't always help. A huge barrier to training them is vanishing gradients: very deep networks often have a gradient signal that goes to zero quickly, thus making gradient descent unbearably slow. More specifically, during gradient descent, as you backprop from the final layer back to the first layer, you are multiplying by the weight matrix on each step, and thus the gradient can decrease exponentially quickly to zero (or, in rare cases, grow exponentially quickly and "explode" to take very large values).

During training, you might therefore see the magnitude (or norm) of the gradient for the earlier layers descrease to zero very rapidly as training proceeds:

Figure 1 : Vanishing gradient The speed of learning decreases very rapidly for the early layers as the network trains

You are now going to solve this problem by building a Residual Network!

2 - Building a Residual Network

In ResNets, a "shortcut" or a "skip connection" allows the gradient to be directly backpropagated to earlier layers:

Figure 2 : A ResNet block showing a skip-connection

The image on the left shows the "main path" through the network. The image on the right adds a shortcut to the main path. By stacking these ResNet blocks on top of each other, you can form a very deep network.

We also saw in lecture that having ResNet blocks with the shortcut also makes it very easy for one of the blocks to learn an identity function. This means that you can stack on additional ResNet blocks with little risk of harming training set performance. (There is also some evidence that the ease of learning an identity function--even more than skip connections helping with vanishing gradients--accounts for ResNets' remarkable performance.)

Two main types of blocks are used in a ResNet, depending mainly on whether the input/output dimensions are same or different. You are going to implement both of them.

2.1 - The identity block

Figure 3 : Identity block. Skip connection "skips over" 2 layers.

The upper path is the "shortcut path." The lower path is the "main path." In this diagram, we have also made explicit the CONV2D and ReLU steps in each layer. To speed up training we have also added a BatchNorm step. Don't worry about this being complicated to implement--you'll see that BatchNorm is just one line of code in Keras!

In this exercise, you'll actually implement a slightly more powerful version of this identity block, in which the skip connection "skips over" 3 hidden layers rather than 2 layers. It looks like this:

Figure 4 : Identity block. Skip connection "skips over" 3 layers

Here're the individual steps.

First component of main path:

  • The first BatchNorm is normalizing the channels axis. Its name should be bn_name_base + '2a'.

  • Then apply the ReLU activation function. This has no name and no hyperparameters.

Second component of main path:

  • The second BatchNorm is normalizing the channels axis. Its name should be bn_name_base + '2b'.

  • Then apply the ReLU activation function. This has no name and no hyperparameters.

Third component of main path:

  • The third BatchNorm is normalizing the channels axis. Its name should be bn_name_base + '2c'. Note that there is no ReLU activation function in this component.

Final step:

  • The shortcut and the input are added together.

  • Then apply the ReLU activation function. This has no name and no hyperparameters.

Exercise: Implement the ResNet identity block. We have implemented the first component of the main path. Please read over this carefully to make sure you understand what it is doing. You should implement the rest.

  • For the activation, use: Activation('relu')(X)

# GRADED FUNCTION: identity_block

def identity_block(X, f, filters, stage, block):
    """
    Implementation of the identity block as defined in Figure 3

    Arguments:
    X -- input tensor of shape (m, n_H_prev, n_W_prev, n_C_prev)
    f -- integer, specifying the shape of the middle CONV's window for the main path
    filters -- python list of integers, defining the number of filters in the CONV layers of the main path
    stage -- integer, used to name the layers, depending on their position in the network
    block -- string/character, used to name the layers, depending on their position in the network

    Returns:
    X -- output of the identity block, tensor of shape (n_H, n_W, n_C)
    """

    # defining name basis
    conv_name_base = 'res' + str(stage) + block + '_branch'
    bn_name_base = 'bn' + str(stage) + block + '_branch'

    # Retrieve Filters
    F1, F2, F3 = filters

    # Save the input value. You'll need this later to add back to the main path. 
    X_shortcut = X

    # First component of main path
    X = Conv2D(filters = F1, kernel_size = (1, 1), strides = (1,1), padding = 'valid', name = conv_name_base + '2a', kernel_initializer = glorot_uniform(seed=0))(X)
    X = BatchNormalization(axis = 3, name = bn_name_base + '2a')(X)
    X = Activation('relu')(X)

    ### START CODE HERE ###

    # Second component of main path (≈3 lines)
    X = Conv2D(filters = F2, kernel_size = (f, f), strides = (1, 1), padding = 'same', name = conv_name_base + '2b', kernel_initializer = glorot_uniform(seed = 0))(X)
    X = BatchNormalization(axis = 3, name = bn_name_base + '2b')(X)
    X = Activation('relu')(X)

    # Third component of main path (≈2 lines)
    X = Conv2D(filters = F3, kernel_size = (1, 1), strides = (1, 1), padding = 'valid', name = conv_name_base + '3c', kernel_initializer = glorot_uniform(seed = 0))(X)
    X = BatchNormalization(axis = 3, name = bn_name_base + '2c')(X)

    # Final step: Add shortcut value to main path, and pass it through a RELU activation (≈2 lines)
    X = Add()([X_shortcut, X])
    X = Activation('relu')(X)

    ### END CODE HERE ###

    return X
tf.reset_default_graph()

with tf.Session() as test:
    np.random.seed(1)
    A_prev = tf.placeholder("float", [3, 4, 4, 6])
    X = np.random.randn(3, 4, 4, 6)
    A = identity_block(A_prev, f = 2, filters = [2, 4, 6], stage = 1, block = 'a')
    test.run(tf.global_variables_initializer())
    out = test.run([A], feed_dict={A_prev: X, K.learning_phase(): 0})
    print("out = " + str(out[0][1][1][0]))

"""
out = [ 0.94822985 0. 1.16101444 2.747859 0. 1.36677003]
"""

2.2 - The convolutional block

You've implemented the ResNet identity block. Next, the ResNet "convolutional block" is the other type of block. You can use this type of block when the input and output dimensions don't match up. The difference with the identity block is that there is a CONV2D layer in the shortcut path:

Figure 4 : Convolutional block

The details of the convolutional block are as follows.

First component of main path:

  • The first BatchNorm is normalizing the channels axis. Its name should be bn_name_base + '2a'.

  • Then apply the ReLU activation function. This has no name and no hyperparameters.

Second component of main path:

  • The second BatchNorm is normalizing the channels axis. Its name should be bn_name_base + '2b'.

  • Then apply the ReLU activation function. This has no name and no hyperparameters.

Third component of main path:

  • The third BatchNorm is normalizing the channels axis. Its name should be bn_name_base + '2c'. Note that there is no ReLU activation function in this component.

Shortcut path:

  • The BatchNorm is normalizing the channels axis. Its name should be bn_name_base + '1'.

Final step:

  • The shortcut and the main path values are added together.

  • Then apply the ReLU activation function. This has no name and no hyperparameters.

    Exercise: Implement the convolutional block. We have implemented the first component of the main path; you should implement the rest. As before, always use 0 as the seed for the random initialization, to ensure consistency with our grader.

  • For the activation, use: Activation('relu')(X)

# GRADED FUNCTION: convolutional_block

def convolutional_block(X, f, filters, stage, block, s = 2):
    """
    Implementation of the convolutional block as defined in Figure 4

    Arguments:
    X -- input tensor of shape (m, n_H_prev, n_W_prev, n_C_prev)
    f -- integer, specifying the shape of the middle CONV's window for the main path
    filters -- python list of integers, defining the number of filters in the CONV layers of the main path
    stage -- integer, used to name the layers, depending on their position in the network
    block -- string/character, used to name the layers, depending on their position in the network
    s -- Integer, specifying the stride to be used

    Returns:
    X -- output of the convolutional block, tensor of shape (n_H, n_W, n_C)
    """

    # defining name basis
    conv_name_base = 'res' + str(stage) + block + '_branch'
    bn_name_base = 'bn' + str(stage) + block + '_branch'

    # Retrieve Filters
    F1, F2, F3 = filters

    # Save the input value
    X_shortcut = X


    ##### MAIN PATH #####
    # First component of main path 
    X = Conv2D(F1, (1, 1), strides = (s,s), name = conv_name_base + '2a', kernel_initializer = glorot_uniform(seed=0))(X)
    X = BatchNormalization(axis = 3, name = bn_name_base + '2a')(X)
    X = Activation('relu')(X)

    ### START CODE HERE ###

    # Second component of main path (≈3 lines)
    X = Conv2D(F2, (f, f), strides = (1, 1), padding = 'same', name = conv_name_base + '2b', kernel_initializer = glorot_uniform(seed = 0))(X)
    X = BatchNormalization(axis = 3, name = bn_name_base + '2b')(X)
    X = Activation('relu')(X)

    # Third component of main path (≈2 lines)
    X = Conv2D(F3, (1, 1), strides = (1, 1), name = conv_name_base + '2c', kernel_initializer = glorot_uniform(seed = 0))(X)
    X = BatchNormalization(axis = 3, name = bn_name_base + '2c')(X)

    ##### SHORTCUT PATH #### (≈2 lines)
    X_shortcut = Conv2D(F3, (1, 1), strides = (s, s), name = conv_name_base + '1', kernel_initializer = glorot_uniform(seed = 0))(X_shortcut)
    X_shortcut = BatchNormalization(axis = 3, name = bn_name_base + '1')(X_shortcut)

    # Final step: Add shortcut value to main path, and pass it through a RELU activation (≈2 lines)
    X = Add()([X, X_shortcut])
    X = Activation('relu')(X)

    ### END CODE HERE ###

    return X
tf.reset_default_graph()

with tf.Session() as test:
    np.random.seed(1)
    A_prev = tf.placeholder("float", [3, 4, 4, 6])
    X = np.random.randn(3, 4, 4, 6)
    A = convolutional_block(A_prev, f = 2, filters = [2, 4, 6], stage = 1, block = 'a')
    test.run(tf.global_variables_initializer())
    out = test.run([A], feed_dict={A_prev: X, K.learning_phase(): 0})
    print("out = " + str(out[0][1][1][0]))

"""
out = [ 0.09018463 1.23489773 0.46822017 0.0367176 0. 0.65516603]
"""

3 - Building your first ResNet model (50 layers)

You now have the necessary blocks to build a very deep ResNet. The following figure describes in detail the architecture of this neural network. "ID BLOCK" in the diagram stands for "Identity block," and "ID BLOCK x3" means you should stack 3 identity blocks together.

Figure 5 : ResNet-50 model

The details of this ResNet-50 model are:

  • Zero-padding pads the input with a pad of (3,3)

  • Stage 1:

  • The 2D Convolution has 64 filters of shape (7,7) and uses a stride of (2,2). Its name is "conv1".

  • BatchNorm is applied to the channels axis of the input.

  • MaxPooling uses a (3,3) window and a (2,2) stride.

  • Stage 2:

  • The convolutional block uses three set of filters of size [64,64,256], "f" is 3, "s" is 1 and the block is "a".

  • The 2 identity blocks use three set of filters of size [64,64,256], "f" is 3 and the blocks are "b" and "c".

  • Stage 3:

  • The convolutional block uses three set of filters of size [128,128,512], "f" is 3, "s" is 2 and the block is "a".

  • The 3 identity blocks use three set of filters of size [128,128,512], "f" is 3 and the blocks are "b", "c" and "d".

  • Stage 4:

  • The convolutional block uses three set of filters of size [256, 256, 1024], "f" is 3, "s" is 2 and the block is "a".

  • The 5 identity blocks use three set of filters of size [256, 256, 1024], "f" is 3 and the blocks are "b", "c", "d", "e" and "f".

  • Stage 5:

  • The convolutional block uses three set of filters of size [512, 512, 2048], "f" is 3, "s" is 2 and the block is "a".

  • The 2 identity blocks use three set of filters of size [512, 512, 2048], "f" is 3 and the blocks are "b" and "c".

  • The 2D Average Pooling uses a window of shape (2,2) and its name is "avg_pool".

  • The flatten doesn't have any hyperparameters or name.

  • The Fully Connected (Dense) layer reduces its input to the number of classes using a softmax activation. Its name should be 'fc' + str(classes).

Exercise: Implement the ResNet with 50 layers described in the figure above. We have implemented Stages 1 and 2. Please implement the rest. (The syntax for implementing Stages 3-5 should be quite similar to that of Stage 2.) Make sure you follow the naming convention in the text above.

You'll need to use this function:

Here're some other functions we used in the code below:

# GRADED FUNCTION: ResNet50

def ResNet50(input_shape = (64, 64, 3), classes = 6):
    """
    Implementation of the popular ResNet50 the following architecture:
    CONV2D -> BATCHNORM -> RELU -> MAXPOOL -> CONVBLOCK -> IDBLOCK*2 -> CONVBLOCK -> IDBLOCK*3
    -> CONVBLOCK -> IDBLOCK*5 -> CONVBLOCK -> IDBLOCK*2 -> AVGPOOL -> TOPLAYER

    Arguments:
    input_shape -- shape of the images of the dataset
    classes -- integer, number of classes

    Returns:
    model -- a Model() instance in Keras
    """

    # Define the input as a tensor with shape input_shape
    X_input = Input(input_shape)


    # Zero-Padding
    X = ZeroPadding2D((3, 3))(X_input)

    # Stage 1
    X = Conv2D(64, (7, 7), strides = (2, 2), name = 'conv1', kernel_initializer = glorot_uniform(seed=0))(X)
    X = BatchNormalization(axis = 3, name = 'bn_conv1')(X)
    X = Activation('relu')(X)
    X = MaxPooling2D((3, 3), strides=(2, 2))(X)

    # Stage 2
    X = convolutional_block(X, f = 3, filters = [64, 64, 256], stage = 2, block='a', s = 1)
    X = identity_block(X, 3, [64, 64, 256], stage=2, block='b')
    X = identity_block(X, 3, [64, 64, 256], stage=2, block='c')

    ### START CODE HERE ###

    # Stage 3 (≈4 lines)
    X = convolutional_block(X, f = 3, filters = [128, 128, 512], stage = 3, block = 'a', s = 2)
    X = identity_block(X, 3, [128, 128, 512], stage = 3, block = 'b')
    X = identity_block(X, 3, [128, 128, 512], stage = 3, block = 'c')
    X = identity_block(X, 3, [128, 128, 512], stage = 3, block = 'd')

    # Stage 4 (≈6 lines)
    X = convolutional_block(X, f = 3, filters = [256, 256, 1024], stage = 4, block = 'a', s = 2)
    X = identity_block(X, 3, [256, 256, 1024], stage = 4, block = 'b')
    X = identity_block(X, 3, [256, 256, 1024], stage = 4, block = 'c')
    X = identity_block(X, 3, [256, 256, 1024], stage = 4, block = 'd')
    X = identity_block(X, 3, [256, 256, 1024], stage = 4, block = 'e')
    X = identity_block(X, 3, [256, 256, 1024], stage = 4, block = 'f')

    # Stage 5 (≈3 lines)
    X = convolutional_block(X, f = 3, filters = [512, 512, 2048], stage = 5, block = 'a', s = 2)
    X = identity_block(X, 3, [512, 512, 2048], stage = 5, block = 'b')
    X = identity_block(X, 3, [512, 512, 2048], stage = 5, block = 'c')

    # AVGPOOL (≈1 line). Use "X = AveragePooling2D(...)(X)"
    X = AveragePooling2D(pool_size = (2, 2), name = 'avg_pool')(X)

    ### END CODE HERE ###

    # output layer
    X = Flatten()(X)
    X = Dense(classes, activation='softmax', name='fc' + str(classes), kernel_initializer = glorot_uniform(seed=0))(X)


    # Create model
    model = Model(inputs = X_input, outputs = X, name='ResNet50')

    return model

Run the following code to build the model's graph. If your implementation is not correct you will know it by checking your accuracy when running model.fit(...) below.

model = ResNet50(input_shape = (64, 64, 3), classes = 6)

As seen in the Keras Tutorial Notebook, prior training a model, you need to configure the learning process by compiling the model.

model.compile(optimizer='sgd', loss='categorical_crossentropy', metrics=['accuracy'])

The model is now ready to be trained. The only thing you need is a dataset.

Let's load the SIGNS Dataset.

Figure 6 : SIGNS dataset

X_train_orig, Y_train_orig, X_test_orig, Y_test_orig, classes = load_dataset()


# Normalize image vectors
X_train = X_train_orig/255.
X_test = X_test_orig/255.


# Convert training and test labels to one hot matrices
Y_train = convert_to_one_hot(Y_train_orig, 6).T
Y_test = convert_to_one_hot(Y_test_orig, 6).T


print ("number of training examples = " + str(X_train.shape[0]))
print ("number of test examples = " + str(X_test.shape[0]))
print ("X_train shape: " + str(X_train.shape))
print ("Y_train shape: " + str(Y_train.shape))
print ("X_test shape: " + str(X_test.shape))
print ("Y_test shape: " + str(Y_test.shape))

number of training examples = 1080 number of test examples = 120 X_train shape: (1080, 64, 64, 3) Y_train shape: (1080, 6) X_test shape: (120, 64, 64, 3) Y_test shape: (120, 6)

Run the following cell to train your model on 2 epochs with a batch size of 32. On a CPU it should take you around 5min per epoch.

model.fit(X_train, Y_train, epochs = 2, batch_size = 8)


Epoch 1/2
1080/1080 [==============================] - 261s - loss: 2.9455 - acc: 0.5241
Epoch 2/2
1080/1080 [==============================] - 258s - loss: 2.8257 - acc: 0.5120

<keras.callbacks.History at 0x7f8554e4e6a0>

Let's see how this model (trained on only two epochs) performs on the test set.

preds = model.evaluate(X_test, Y_test)
print ("Loss = " + str(preds[0]))
print ("Test Accuracy = " + str(preds[1]))
120/120 [==============================] - 9s
Loss = 13.4317459742
Test Accuracy = 0.166666667163

For the purpose of this assignment, we've asked you to train the model only for two epochs. You can see that it achieves poor performances. Please go ahead and submit your assignment; to check correctness, the online grader will run your code only for a small number of epochs as well.

After you have finished this official (graded) part of this assignment, you can also optionally train the ResNet for more iterations, if you want. We get a lot better performance when we train for ~20 epochs, but this will take more than an hour when training on a CPU.

Using a GPU, we've trained our own ResNet50 model's weights on the SIGNS dataset. You can load and run our trained model on the test set in the cells below. It may take ≈1min to load the model.

model = load_model('ResNet50.h5')
preds = model.evaluate(X_test, Y_test)
print ("Loss = " + str(preds[0]))
print ("Test Accuracy = " + str(preds[1]))
120/120 [==============================] - 10s
Loss = 0.530178320408
Test Accuracy = 0.866666662693

ResNet50 is a powerful model for image classification when it is trained for an adequate number of iterations. We hope you can use what you've learnt and apply it to your own classification problem to perform state-of-the-art accuracy.

Congratulations on finishing this assignment! You've now implemented a state-of-the-art image classification system!

4 - Test on your own image (Optional/Ungraded)

If you wish, you can also take a picture of your own hand and see the output of the model. To do this: 1. Click on "File" in the upper bar of this notebook, then click "Open" to go on your Coursera Hub. 2. Add your image to this Jupyter Notebook's directory, in the "images" folder 3. Write your image's name in the following code 4. Run the code and check if the algorithm is right!

img_path = 'images/my_image.jpg'
img = image.load_img(img_path, target_size=(64, 64))
x = image.img_to_array(img)/255.
x = np.expand_dims(x, axis=0)
x = preprocess_input(x)
print('Input image shape:', x.shape)
my_image = scipy.misc.imread(img_path)
imshow(my_image)
print("class prediction vector [p(0), p(1), p(2), p(3), p(4), p(5)] = ")
print(model.predict(x))
Input image shape: (1, 64, 64, 3)
class prediction vector [p(0), p(1), p(2), p(3), p(4), p(5)] = [[ 1. 0. 0. 0. 0. 0.]]

You can also print a summary of your model by running the following code.

model.summary()
____________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to
====================================================================================================
input_1 (InputLayer) (None, 64, 64, 3) 0
____________________________________________________________________________________________________
zero_padding2d_1 (ZeroPadding2D) (None, 70, 70, 3) 0 input_1[0][0]
____________________________________________________________________________________________________
conv1 (Conv2D) (None, 32, 32, 64) 9472 zero_padding2d_1[0][0]
____________________________________________________________________________________________________
bn_conv1 (BatchNormalization) (None, 32, 32, 64) 256 conv1[0][0]
____________________________________________________________________________________________________
activation_4 (Activation) (None, 32, 32, 64) 0 bn_conv1[0][0]
____________________________________________________________________________________________________
max_pooling2d_1 (MaxPooling2D) (None, 15, 15, 64) 0 activation_4[0][0]
____________________________________________________________________________________________________
res2a_branch2a (Conv2D) (None, 15, 15, 64) 4160 max_pooling2d_1[0][0]
____________________________________________________________________________________________________
bn2a_branch2a (BatchNormalizatio (None, 15, 15, 64) 256 res2a_branch2a[0][0]
____________________________________________________________________________________________________
activation_5 (Activation) (None, 15, 15, 64) 0 bn2a_branch2a[0][0]
____________________________________________________________________________________________________
res2a_branch2b (Conv2D) (None, 15, 15, 64) 36928 activation_5[0][0]
____________________________________________________________________________________________________
bn2a_branch2b (BatchNormalizatio (None, 15, 15, 64) 256 res2a_branch2b[0][0]
____________________________________________________________________________________________________
activation_6 (Activation) (None, 15, 15, 64) 0 bn2a_branch2b[0][0]
____________________________________________________________________________________________________
res2a_branch2c (Conv2D) (None, 15, 15, 256) 16640 activation_6[0][0]
____________________________________________________________________________________________________
res2a_branch1 (Conv2D) (None, 15, 15, 256) 16640 max_pooling2d_1[0][0]
____________________________________________________________________________________________________
bn2a_branch2c (BatchNormalizatio (None, 15, 15, 256) 1024 res2a_branch2c[0][0]
____________________________________________________________________________________________________
bn2a_branch1 (BatchNormalization (None, 15, 15, 256) 1024 res2a_branch1[0][0]
____________________________________________________________________________________________________
add_2 (Add) (None, 15, 15, 256) 0 bn2a_branch2c[0][0]
bn2a_branch1[0][0]
____________________________________________________________________________________________________
activation_7 (Activation) (None, 15, 15, 256) 0 add_2[0][0]
____________________________________________________________________________________________________
res2b_branch2a (Conv2D) (None, 15, 15, 64) 16448 activation_7[0][0]
____________________________________________________________________________________________________
bn2b_branch2a (BatchNormalizatio (None, 15, 15, 64) 256 res2b_branch2a[0][0]
____________________________________________________________________________________________________
activation_8 (Activation) (None, 15, 15, 64) 0 bn2b_branch2a[0][0]
____________________________________________________________________________________________________
res2b_branch2b (Conv2D) (None, 15, 15, 64) 36928 activation_8[0][0]
____________________________________________________________________________________________________
bn2b_branch2b (BatchNormalizatio (None, 15, 15, 64) 256 res2b_branch2b[0][0]
____________________________________________________________________________________________________
activation_9 (Activation) (None, 15, 15, 64) 0 bn2b_branch2b[0][0]
____________________________________________________________________________________________________
res2b_branch2c (Conv2D) (None, 15, 15, 256) 16640 activation_9[0][0]
____________________________________________________________________________________________________
bn2b_branch2c (BatchNormalizatio (None, 15, 15, 256) 1024 res2b_branch2c[0][0]
____________________________________________________________________________________________________
add_3 (Add) (None, 15, 15, 256) 0 bn2b_branch2c[0][0]
activation_7[0][0]
____________________________________________________________________________________________________
activation_10 (Activation) (None, 15, 15, 256) 0 add_3[0][0]
____________________________________________________________________________________________________
res2c_branch2a (Conv2D) (None, 15, 15, 64) 16448 activation_10[0][0]
____________________________________________________________________________________________________
bn2c_branch2a (BatchNormalizatio (None, 15, 15, 64) 256 res2c_branch2a[0][0]
____________________________________________________________________________________________________
activation_11 (Activation) (None, 15, 15, 64) 0 bn2c_branch2a[0][0]
____________________________________________________________________________________________________
res2c_branch2b (Conv2D) (None, 15, 15, 64) 36928 activation_11[0][0]
____________________________________________________________________________________________________
bn2c_branch2b (BatchNormalizatio (None, 15, 15, 64) 256 res2c_branch2b[0][0]
____________________________________________________________________________________________________
activation_12 (Activation) (None, 15, 15, 64) 0 bn2c_branch2b[0][0]
____________________________________________________________________________________________________
res2c_branch2c (Conv2D) (None, 15, 15, 256) 16640 activation_12[0][0]
____________________________________________________________________________________________________
bn2c_branch2c (BatchNormalizatio (None, 15, 15, 256) 1024 res2c_branch2c[0][0]
____________________________________________________________________________________________________
add_4 (Add) (None, 15, 15, 256) 0 bn2c_branch2c[0][0]
activation_10[0][0]
____________________________________________________________________________________________________
activation_13 (Activation) (None, 15, 15, 256) 0 add_4[0][0]
____________________________________________________________________________________________________
res3a_branch2a (Conv2D) (None, 8, 8, 128) 32896 activation_13[0][0]
____________________________________________________________________________________________________
bn3a_branch2a (BatchNormalizatio (None, 8, 8, 128) 512 res3a_branch2a[0][0]
____________________________________________________________________________________________________
activation_14 (Activation) (None, 8, 8, 128) 0 bn3a_branch2a[0][0]
____________________________________________________________________________________________________
res3a_branch2b (Conv2D) (None, 8, 8, 128) 147584 activation_14[0][0]
____________________________________________________________________________________________________
bn3a_branch2b (BatchNormalizatio (None, 8, 8, 128) 512 res3a_branch2b[0][0]
____________________________________________________________________________________________________
activation_15 (Activation) (None, 8, 8, 128) 0 bn3a_branch2b[0][0]
____________________________________________________________________________________________________
res3a_branch2c (Conv2D) (None, 8, 8, 512) 66048 activation_15[0][0]
____________________________________________________________________________________________________
res3a_branch1 (Conv2D) (None, 8, 8, 512) 131584 activation_13[0][0]
____________________________________________________________________________________________________
bn3a_branch2c (BatchNormalizatio (None, 8, 8, 512) 2048 res3a_branch2c[0][0]
____________________________________________________________________________________________________
bn3a_branch1 (BatchNormalization (None, 8, 8, 512) 2048 res3a_branch1[0][0]
____________________________________________________________________________________________________
add_5 (Add) (None, 8, 8, 512) 0 bn3a_branch2c[0][0]
bn3a_branch1[0][0]
____________________________________________________________________________________________________
activation_16 (Activation) (None, 8, 8, 512) 0 add_5[0][0]
____________________________________________________________________________________________________
res3b_branch2a (Conv2D) (None, 8, 8, 128) 65664 activation_16[0][0]
____________________________________________________________________________________________________
bn3b_branch2a (BatchNormalizatio (None, 8, 8, 128) 512 res3b_branch2a[0][0]
____________________________________________________________________________________________________
activation_17 (Activation) (None, 8, 8, 128) 0 bn3b_branch2a[0][0]
____________________________________________________________________________________________________
res3b_branch2b (Conv2D) (None, 8, 8, 128) 147584 activation_17[0][0]
____________________________________________________________________________________________________
bn3b_branch2b (BatchNormalizatio (None, 8, 8, 128) 512 res3b_branch2b[0][0]
____________________________________________________________________________________________________
activation_18 (Activation) (None, 8, 8, 128) 0 bn3b_branch2b[0][0]
____________________________________________________________________________________________________
res3b_branch2c (Conv2D) (None, 8, 8, 512) 66048 activation_18[0][0]
____________________________________________________________________________________________________
bn3b_branch2c (BatchNormalizatio (None, 8, 8, 512) 2048 res3b_branch2c[0][0]
____________________________________________________________________________________________________
add_6 (Add) (None, 8, 8, 512) 0 bn3b_branch2c[0][0]
activation_16[0][0]
____________________________________________________________________________________________________
activation_19 (Activation) (None, 8, 8, 512) 0 add_6[0][0]
____________________________________________________________________________________________________
res3c_branch2a (Conv2D) (None, 8, 8, 128) 65664 activation_19[0][0]
____________________________________________________________________________________________________
bn3c_branch2a (BatchNormalizatio (None, 8, 8, 128) 512 res3c_branch2a[0][0]
____________________________________________________________________________________________________
activation_20 (Activation) (None, 8, 8, 128) 0 bn3c_branch2a[0][0]
____________________________________________________________________________________________________
res3c_branch2b (Conv2D) (None, 8, 8, 128) 147584 activation_20[0][0]
____________________________________________________________________________________________________
bn3c_branch2b (BatchNormalizatio (None, 8, 8, 128) 512 res3c_branch2b[0][0]
____________________________________________________________________________________________________
activation_21 (Activation) (None, 8, 8, 128) 0 bn3c_branch2b[0][0]
____________________________________________________________________________________________________
res3c_branch2c (Conv2D) (None, 8, 8, 512) 66048 activation_21[0][0]
____________________________________________________________________________________________________
bn3c_branch2c (BatchNormalizatio (None, 8, 8, 512) 2048 res3c_branch2c[0][0]
____________________________________________________________________________________________________
add_7 (Add) (None, 8, 8, 512) 0 bn3c_branch2c[0][0]
activation_19[0][0]
____________________________________________________________________________________________________
activation_22 (Activation) (None, 8, 8, 512) 0 add_7[0][0]
____________________________________________________________________________________________________
res3d_branch2a (Conv2D) (None, 8, 8, 128) 65664 activation_22[0][0]
____________________________________________________________________________________________________
bn3d_branch2a (BatchNormalizatio (None, 8, 8, 128) 512 res3d_branch2a[0][0]
____________________________________________________________________________________________________
activation_23 (Activation) (None, 8, 8, 128) 0 bn3d_branch2a[0][0]
____________________________________________________________________________________________________
res3d_branch2b (Conv2D) (None, 8, 8, 128) 147584 activation_23[0][0]
____________________________________________________________________________________________________
bn3d_branch2b (BatchNormalizatio (None, 8, 8, 128) 512 res3d_branch2b[0][0]
____________________________________________________________________________________________________
activation_24 (Activation) (None, 8, 8, 128) 0 bn3d_branch2b[0][0]
____________________________________________________________________________________________________
res3d_branch2c (Conv2D) (None, 8, 8, 512) 66048 activation_24[0][0]
____________________________________________________________________________________________________
bn3d_branch2c (BatchNormalizatio (None, 8, 8, 512) 2048 res3d_branch2c[0][0]
____________________________________________________________________________________________________
add_8 (Add) (None, 8, 8, 512) 0 bn3d_branch2c[0][0]
activation_22[0][0]
____________________________________________________________________________________________________
activation_25 (Activation) (None, 8, 8, 512) 0 add_8[0][0]
____________________________________________________________________________________________________
res4a_branch2a (Conv2D) (None, 4, 4, 256) 131328 activation_25[0][0]
____________________________________________________________________________________________________
bn4a_branch2a (BatchNormalizatio (None, 4, 4, 256) 1024 res4a_branch2a[0][0]
____________________________________________________________________________________________________
activation_26 (Activation) (None, 4, 4, 256) 0 bn4a_branch2a[0][0]
____________________________________________________________________________________________________
res4a_branch2b (Conv2D) (None, 4, 4, 256) 590080 activation_26[0][0]
____________________________________________________________________________________________________
bn4a_branch2b (BatchNormalizatio (None, 4, 4, 256) 1024 res4a_branch2b[0][0]
____________________________________________________________________________________________________
activation_27 (Activation) (None, 4, 4, 256) 0 bn4a_branch2b[0][0]
____________________________________________________________________________________________________
res4a_branch2c (Conv2D) (None, 4, 4, 1024) 263168 activation_27[0][0]
____________________________________________________________________________________________________
res4a_branch1 (Conv2D) (None, 4, 4, 1024) 525312 activation_25[0][0]
____________________________________________________________________________________________________
bn4a_branch2c (BatchNormalizatio (None, 4, 4, 1024) 4096 res4a_branch2c[0][0]
____________________________________________________________________________________________________
bn4a_branch1 (BatchNormalization (None, 4, 4, 1024) 4096 res4a_branch1[0][0]
____________________________________________________________________________________________________
add_9 (Add) (None, 4, 4, 1024) 0 bn4a_branch2c[0][0]
bn4a_branch1[0][0]
____________________________________________________________________________________________________
activation_28 (Activation) (None, 4, 4, 1024) 0 add_9[0][0]
____________________________________________________________________________________________________
res4b_branch2a (Conv2D) (None, 4, 4, 256) 262400 activation_28[0][0]
____________________________________________________________________________________________________
bn4b_branch2a (BatchNormalizatio (None, 4, 4, 256) 1024 res4b_branch2a[0][0]
____________________________________________________________________________________________________
activation_29 (Activation) (None, 4, 4, 256) 0 bn4b_branch2a[0][0]
____________________________________________________________________________________________________
res4b_branch2b (Conv2D) (None, 4, 4, 256) 590080 activation_29[0][0]
____________________________________________________________________________________________________
bn4b_branch2b (BatchNormalizatio (None, 4, 4, 256) 1024 res4b_branch2b[0][0]
____________________________________________________________________________________________________
activation_30 (Activation) (None, 4, 4, 256) 0 bn4b_branch2b[0][0]
____________________________________________________________________________________________________
res4b_branch2c (Conv2D) (None, 4, 4, 1024) 263168 activation_30[0][0]
____________________________________________________________________________________________________
bn4b_branch2c (BatchNormalizatio (None, 4, 4, 1024) 4096 res4b_branch2c[0][0]
____________________________________________________________________________________________________
add_10 (Add) (None, 4, 4, 1024) 0 bn4b_branch2c[0][0]
activation_28[0][0]
____________________________________________________________________________________________________
activation_31 (Activation) (None, 4, 4, 1024) 0 add_10[0][0]
____________________________________________________________________________________________________
res4c_branch2a (Conv2D) (None, 4, 4, 256) 262400 activation_31[0][0]
____________________________________________________________________________________________________
bn4c_branch2a (BatchNormalizatio (None, 4, 4, 256) 1024 res4c_branch2a[0][0]
____________________________________________________________________________________________________
activation_32 (Activation) (None, 4, 4, 256) 0 bn4c_branch2a[0][0]
____________________________________________________________________________________________________
res4c_branch2b (Conv2D) (None, 4, 4, 256) 590080 activation_32[0][0]
____________________________________________________________________________________________________
bn4c_branch2b (BatchNormalizatio (None, 4, 4, 256) 1024 res4c_branch2b[0][0]
____________________________________________________________________________________________________
activation_33 (Activation) (None, 4, 4, 256) 0 bn4c_branch2b[0][0]
____________________________________________________________________________________________________
res4c_branch2c (Conv2D) (None, 4, 4, 1024) 263168 activation_33[0][0]
____________________________________________________________________________________________________
bn4c_branch2c (BatchNormalizatio (None, 4, 4, 1024) 4096 res4c_branch2c[0][0]
____________________________________________________________________________________________________
add_11 (Add) (None, 4, 4, 1024) 0 bn4c_branch2c[0][0]
activation_31[0][0]
____________________________________________________________________________________________________
activation_34 (Activation) (None, 4, 4, 1024) 0 add_11[0][0]
____________________________________________________________________________________________________
res4d_branch2a (Conv2D) (None, 4, 4, 256) 262400 activation_34[0][0]
____________________________________________________________________________________________________
bn4d_branch2a (BatchNormalizatio (None, 4, 4, 256) 1024 res4d_branch2a[0][0]
____________________________________________________________________________________________________
activation_35 (Activation) (None, 4, 4, 256) 0 bn4d_branch2a[0][0]
____________________________________________________________________________________________________
res4d_branch2b (Conv2D) (None, 4, 4, 256) 590080 activation_35[0][0]
____________________________________________________________________________________________________
bn4d_branch2b (BatchNormalizatio (None, 4, 4, 256) 1024 res4d_branch2b[0][0]
____________________________________________________________________________________________________
activation_36 (Activation) (None, 4, 4, 256) 0 bn4d_branch2b[0][0]
____________________________________________________________________________________________________
res4d_branch2c (Conv2D) (None, 4, 4, 1024) 263168 activation_36[0][0]
____________________________________________________________________________________________________
bn4d_branch2c (BatchNormalizatio (None, 4, 4, 1024) 4096 res4d_branch2c[0][0]
____________________________________________________________________________________________________
add_12 (Add) (None, 4, 4, 1024) 0 bn4d_branch2c[0][0]
activation_34[0][0]
____________________________________________________________________________________________________
activation_37 (Activation) (None, 4, 4, 1024) 0 add_12[0][0]
____________________________________________________________________________________________________
res4e_branch2a (Conv2D) (None, 4, 4, 256) 262400 activation_37[0][0]
____________________________________________________________________________________________________
bn4e_branch2a (BatchNormalizatio (None, 4, 4, 256) 1024 res4e_branch2a[0][0]
____________________________________________________________________________________________________
activation_38 (Activation) (None, 4, 4, 256) 0 bn4e_branch2a[0][0]
____________________________________________________________________________________________________
res4e_branch2b (Conv2D) (None, 4, 4, 256) 590080 activation_38[0][0]
____________________________________________________________________________________________________
bn4e_branch2b (BatchNormalizatio (None, 4, 4, 256) 1024 res4e_branch2b[0][0]
____________________________________________________________________________________________________
activation_39 (Activation) (None, 4, 4, 256) 0 bn4e_branch2b[0][0]
____________________________________________________________________________________________________
res4e_branch2c (Conv2D) (None, 4, 4, 1024) 263168 activation_39[0][0]
____________________________________________________________________________________________________
bn4e_branch2c (BatchNormalizatio (None, 4, 4, 1024) 4096 res4e_branch2c[0][0]
____________________________________________________________________________________________________
add_13 (Add) (None, 4, 4, 1024) 0 bn4e_branch2c[0][0]
activation_37[0][0]
____________________________________________________________________________________________________
activation_40 (Activation) (None, 4, 4, 1024) 0 add_13[0][0]
____________________________________________________________________________________________________
res4f_branch2a (Conv2D) (None, 4, 4, 256) 262400 activation_40[0][0]
____________________________________________________________________________________________________
bn4f_branch2a (BatchNormalizatio (None, 4, 4, 256) 1024 res4f_branch2a[0][0]
____________________________________________________________________________________________________
activation_41 (Activation) (None, 4, 4, 256) 0 bn4f_branch2a[0][0]
____________________________________________________________________________________________________
res4f_branch2b (Conv2D) (None, 4, 4, 256) 590080 activation_41[0][0]
____________________________________________________________________________________________________
bn4f_branch2b (BatchNormalizatio (None, 4, 4, 256) 1024 res4f_branch2b[0][0]
____________________________________________________________________________________________________
activation_42 (Activation) (None, 4, 4, 256) 0 bn4f_branch2b[0][0]
____________________________________________________________________________________________________
res4f_branch2c (Conv2D) (None, 4, 4, 1024) 263168 activation_42[0][0]
____________________________________________________________________________________________________
bn4f_branch2c (BatchNormalizatio (None, 4, 4, 1024) 4096 res4f_branch2c[0][0]
____________________________________________________________________________________________________
add_14 (Add) (None, 4, 4, 1024) 0 bn4f_branch2c[0][0]
activation_40[0][0]
____________________________________________________________________________________________________
activation_43 (Activation) (None, 4, 4, 1024) 0 add_14[0][0]
____________________________________________________________________________________________________
res5a_branch2a (Conv2D) (None, 2, 2, 512) 524800 activation_43[0][0]
____________________________________________________________________________________________________
bn5a_branch2a (BatchNormalizatio (None, 2, 2, 512) 2048 res5a_branch2a[0][0]
____________________________________________________________________________________________________
activation_44 (Activation) (None, 2, 2, 512) 0 bn5a_branch2a[0][0]
____________________________________________________________________________________________________
res5a_branch2b (Conv2D) (None, 2, 2, 512) 2359808 activation_44[0][0]
____________________________________________________________________________________________________
bn5a_branch2b (BatchNormalizatio (None, 2, 2, 512) 2048 res5a_branch2b[0][0]
____________________________________________________________________________________________________
activation_45 (Activation) (None, 2, 2, 512) 0 bn5a_branch2b[0][0]
____________________________________________________________________________________________________
res5a_branch2c (Conv2D) (None, 2, 2, 2048) 1050624 activation_45[0][0]
____________________________________________________________________________________________________
res5a_branch1 (Conv2D) (None, 2, 2, 2048) 2099200 activation_43[0][0]
____________________________________________________________________________________________________
bn5a_branch2c (BatchNormalizatio (None, 2, 2, 2048) 8192 res5a_branch2c[0][0]
____________________________________________________________________________________________________
bn5a_branch1 (BatchNormalization (None, 2, 2, 2048) 8192 res5a_branch1[0][0]
____________________________________________________________________________________________________
add_15 (Add) (None, 2, 2, 2048) 0 bn5a_branch2c[0][0]
bn5a_branch1[0][0]
____________________________________________________________________________________________________
activation_46 (Activation) (None, 2, 2, 2048) 0 add_15[0][0]
____________________________________________________________________________________________________
res5b_branch2a (Conv2D) (None, 2, 2, 512) 1049088 activation_46[0][0]
____________________________________________________________________________________________________
bn5b_branch2a (BatchNormalizatio (None, 2, 2, 512) 2048 res5b_branch2a[0][0]
____________________________________________________________________________________________________
activation_47 (Activation) (None, 2, 2, 512) 0 bn5b_branch2a[0][0]
____________________________________________________________________________________________________
res5b_branch2b (Conv2D) (None, 2, 2, 512) 2359808 activation_47[0][0]
____________________________________________________________________________________________________
bn5b_branch2b (BatchNormalizatio (None, 2, 2, 512) 2048 res5b_branch2b[0][0]
____________________________________________________________________________________________________
activation_48 (Activation) (None, 2, 2, 512) 0 bn5b_branch2b[0][0]
____________________________________________________________________________________________________
res5b_branch2c (Conv2D) (None, 2, 2, 2048) 1050624 activation_48[0][0]
____________________________________________________________________________________________________
bn5b_branch2c (BatchNormalizatio (None, 2, 2, 2048) 8192 res5b_branch2c[0][0]
____________________________________________________________________________________________________
add_16 (Add) (None, 2, 2, 2048) 0 bn5b_branch2c[0][0]
activation_46[0][0]
____________________________________________________________________________________________________
activation_49 (Activation) (None, 2, 2, 2048) 0 add_16[0][0]
____________________________________________________________________________________________________
res5c_branch2a (Conv2D) (None, 2, 2, 512) 1049088 activation_49[0][0]
____________________________________________________________________________________________________
bn5c_branch2a (BatchNormalizatio (None, 2, 2, 512) 2048 res5c_branch2a[0][0]
____________________________________________________________________________________________________
activation_50 (Activation) (None, 2, 2, 512) 0 bn5c_branch2a[0][0]
____________________________________________________________________________________________________
res5c_branch2b (Conv2D) (None, 2, 2, 512) 2359808 activation_50[0][0]
____________________________________________________________________________________________________
bn5c_branch2b (BatchNormalizatio (None, 2, 2, 512) 2048 res5c_branch2b[0][0]
____________________________________________________________________________________________________
activation_51 (Activation) (None, 2, 2, 512) 0 bn5c_branch2b[0][0]
____________________________________________________________________________________________________
res5c_branch2c (Conv2D) (None, 2, 2, 2048) 1050624 activation_51[0][0]
____________________________________________________________________________________________________
bn5c_branch2c (BatchNormalizatio (None, 2, 2, 2048) 8192 res5c_branch2c[0][0]
____________________________________________________________________________________________________
add_17 (Add) (None, 2, 2, 2048) 0 bn5c_branch2c[0][0]
activation_49[0][0]
____________________________________________________________________________________________________
activation_52 (Activation) (None, 2, 2, 2048) 0 add_17[0][0]
____________________________________________________________________________________________________
avg_pool (AveragePooling2D) (None, 1, 1, 2048) 0 activation_52[0][0]
____________________________________________________________________________________________________
flatten_1 (Flatten) (None, 2048) 0 avg_pool[0][0]
____________________________________________________________________________________________________
fc6 (Dense) (None, 6) 12294 flatten_1[0][0]
====================================================================================================
Total params: 23,600,006
Trainable params: 23,546,886
Non-trainable params: 53,120
____________________________________________________________________________________________________

Finally, run the code below to visualize your ResNet50. You can also download a .png picture of your model by going to "File -> Open...-> model.png".

plot_model(model, to_file='model.png')
SVG(model_to_dot(model).create(prog='dot', format='svg'))

What you should remember:

  • Very deep "plain" networks don't work in practice because they are hard to train due to vanishing gradients.

  • The skip-connections help to address the Vanishing Gradient problem. They also make it easy for a ResNet block to learn an identity function.

  • There are two main type of blocks: The identity block and the convolutional block.

  • Very deep Residual Networks are built by stacking these blocks together.

References

This notebook presents the ResNet algorithm due to He et al. (2015). The implementation here also took significant inspiration and follows the structure given in the github repository of Francois Chollet:

The identity block is the standard block used in ResNets, and corresponds to the case where the input activation (say a[l]a^{[l]}a[l]) has the same dimension as the output activation (say a[l+2]a^{[l+2]}a[l+2]). To flesh out the different steps of what happens in a ResNet's identity block, here is an alternative diagram showing the individual steps:

The first CONV2D has F1F_1F1​ filters of shape (1,1) and a stride of (1,1). Its padding is "valid" and its name should be conv_name_base + '2a'. Use 0 as the seed for the random initialization.

The second CONV2D has F2F_2F2​ filters of shape (f,f)(f,f)(f,f) and a stride of (1,1). Its padding is "same" and its name should be conv_name_base + '2b'. Use 0 as the seed for the random initialization.

The third CONV2D has F3F_3F3​ filters of shape (1,1) and a stride of (1,1). Its padding is "valid" and its name should be conv_name_base + '2c'. Use 0 as the seed for the random initialization.

To implement the Conv2D step:

To implement BatchNorm: (axis: Integer, the axis that should be normalized (typically the channels axis))

To add the value passed forward by the shortcut:

The CONV2D layer in the shortcut path is used to resize the input xxx to a different dimension, so that the dimensions match up in the final addition needed to add the shortcut value back to the main path. (This plays a similar role as the matrix WsW_sWs​ discussed in lecture.) For example, to reduce the activation dimensions's height and width by a factor of 2, you can use a 1x1 convolution with a stride of 2. The CONV2D layer on the shortcut path does not use any non-linear activation function. Its main role is to just apply a (learned) linear function that reduces the dimension of the input, so that the dimensions match up for the later addition step.

The first CONV2D has F1F_1F1​ filters of shape (1,1) and a stride of (s,s). Its padding is "valid" and its name should be conv_name_base + '2a'.

The second CONV2D has F2F_2F2​ filters of (f,f) and a stride of (1,1). Its padding is "same" and it's name should be conv_name_base + '2b'.

The third CONV2D has F3F_3F3​ filters of (1,1) and a stride of (1,1). Its padding is "valid" and it's name should be conv_name_base + '2c'.

The CONV2D has F3F_3F3​ filters of shape (1,1) and a stride of (s,s). Its padding is "valid" and its name should be conv_name_base + '1'.

(axis: Integer, the axis that should be normalized (typically the features axis))

Average pooling

Conv2D:

BatchNorm: (axis: Integer, the axis that should be normalized (typically the features axis))

Zero padding:

Max pooling:

Fully conected layer:

Addition:

Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun -

Francois Chollet's github repository:

He et al.
See reference
See reference
See reference
Conv Hint
BatchNorm Hint
Addition Hint
see reference
See reference
See reference
See reference
See reference
See reference
See reference
Deep Residual Learning for Image Recognition (2015)
https://github.com/fchollet/deep-learning-models/blob/master/resnet50.py