吴恩达deeplearning Lesson4 Week2 keras入门 2个案例：happyhouse+resnet

Keras±+Tutorial±+Happy+House+v2

遇到两个问题。
model函数建立如下：

def HappyModel(input_shape):
    """
    Implementation of the HappyModel.
    
    Arguments:
    input_shape -- shape of the images of the dataset

    Returns:
    model -- a Model() instance in Keras
    """
    
    ### START CODE HERE ###
    # Feel free to use the suggested outline in the text above to get started, and run through the whole
    # exercise (including the later portions of this notebook) once. The come back also try out other
    # network architectures as well. 
    X_input = Input(input_shape)

    # Zero-Padding: pads the border of X_input with zeroes
    X = ZeroPadding2D((3, 3))(X_input)

    # CONV -> BN -> RELU Block applied to X
    X = Conv2D(32, (5, 5), strides = (1, 1), name = 'conv0')(X)
    X = BatchNormalization(axis = 3, name = 'bn0')(X)
    X = Activation('relu')(X)
    X = MaxPooling2D((2, 2), name='max_pool0')(X)
    
    X = Conv2D(16, (3, 3), strides = (1, 1), name = 'conv1')(X)
    X = BatchNormalization(axis = 3, name = 'bn1')(X)
    X = Activation('relu')(X)
    # MAXPOOL
    X = MaxPooling2D((2, 2), name='max_pool1')(X)

    # FLATTEN X (means convert it to a vector) + FULLYCONNECTED
    X = Flatten()(X)
    X = Dense(1, activation='sigmoid', name='fc')(X)

    # Create model. This creates your Keras model instance, you'll use this instance to train/test the model.
    model = Model(inputs = X_input, outputs = X, name='HappyModel')
    
    ### END CODE HERE ###
    
    return model

input

这里的 X_input = Input(input_shape) 的input_shape 代表的是每张图片的shape，在本例子中是（64,64,3）。并不是(m,64,64,3)。
input的意义是告诉model要处理的每一个样本是什么样的，再去做后续处理。

正确率50%

我将例子给的model搭建顺序直接放到code中，发现永远都是正确率50%
例子放上来，下面是错误示范：

def model(input_shape):
    # Define the input placeholder as a tensor with shape input_shape. Think of this as your input image!
    X_input = Input(input_shape)

    # Zero-Padding: pads the border of X_input with zeroes
    X = ZeroPadding2D((3, 3))(X_input)

    # CONV -> BN -> RELU Block applied to X
    X = Conv2D(32, (7, 7), strides = (1, 1), name = 'conv0')(X)
    X = BatchNormalization(axis = 3, name = 'bn0')(X)
    X = Activation('relu')(X)

    # MAXPOOL
    X = MaxPooling2D((2, 2), name='max_pool')(X)

    # FLATTEN X (means convert it to a vector) + FULLYCONNECTED
    X = Flatten()(X)
    X = Dense(1, activation='sigmoid', name='fc')(X)

    # Create model. This creates your Keras model instance, you'll use this instance to train/test the model.
    model = Model(inputs = X_input, outputs = X, name='HappyModel')

    return model

我在想是不是我的设置问题，去discuss论坛上看，被各路豪杰一顿误导。

最后看到其中一位大哥说 是因为例子的误导，如果照搬例子则会如此，我果断加了一层conv、bn、relu、pool就达到了正常的水平(代码在keras大标题下面)

compile中 loss 选择问题

一开始去keras官方文档的例子看，使用例子推荐的categorical_crossentropy：

happyModel.compile(optimizer = 'Adamax', loss='categorical_crossentropy', metrics = ["accuracy"])

报错。
因为本例子的标签只有0和1，所以更改为’binary_crossentropy’

代码

import numpy as np
from keras import layers
from keras.layers import Input, Dense, Activation, ZeroPadding2D, BatchNormalization, Flatten, Conv2D
from keras.layers import AveragePooling2D, MaxPooling2D, Dropout, GlobalMaxPooling2D, GlobalAveragePooling2D
from keras.models import 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 kt_utils import *

import keras.backend as K
K.set_image_data_format('channels_last')
import matplotlib.pyplot as plt
from matplotlib.pyplot import imshow

%matplotlib inline

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.

# Reshape
Y_train = Y_train_orig.T
Y_test = Y_test_orig.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))
def HappyModel(input_shape):
    """
    Implementation of the HappyModel.
    
    Arguments:
    input_shape -- shape of the images of the dataset

    Returns:
    model -- a Model() instance in Keras
    """
    
    ### START CODE HERE ###
    # Feel free to use the suggested outline in the text above to get started, and run through the whole
    # exercise (including the later portions of this notebook) once. The come back also try out other
    # network architectures as well. 
    X_input = Input(input_shape)

    # Zero-Padding: pads the border of X_input with zeroes
    X = ZeroPadding2D((3, 3))(X_input)

    # CONV -> BN -> RELU Block applied to X
    X = Conv2D(32, (5, 5), strides = (1, 1), name = 'conv0')(X)
    X = BatchNormalization(axis = 3, name = 'bn0')(X)
    X = Activation('relu')(X)
    X = MaxPooling2D((2, 2), name='max_pool0')(X)
    
    X = Conv2D(16, (3, 3), strides = (1, 1), name = 'conv1')(X)
    X = BatchNormalization(axis = 3, name = 'bn1')(X)
    X = Activation('relu')(X)
    # MAXPOOL
    X = MaxPooling2D((2, 2), name='max_pool1')(X)

    # FLATTEN X (means convert it to a vector) + FULLYCONNECTED
    X = Flatten()(X)
    X = Dense(1, activation='sigmoid', name='fc')(X)

    # Create model. This creates your Keras model instance, you'll use this instance to train/test the model.
    model = Model(inputs = X_input, outputs = X, name='HappyModel')
    
    ### END CODE HERE ###
    
    return model
happyModel = HappyModel((X_train.shape[1],X_train.shape[2],X_train.shape[3]))
happyModel.compile(optimizer = 'Adamax', loss='binary_crossentropy', metrics = ["accuracy"])
happyModel.fit(x =X_train, y = Y_train, epochs = 100, batch_size = 64)
### START CODE HERE ### (1 line)
preds = happyModel.evaluate(x = X_test, y = Y_test)
### END CODE HERE ###
print()
print ("Loss = " + str(preds[0]))
print ("Test Accuracy = " + str(preds[1]))

150/150 [==============================] - 1s 5ms/step

Loss = 0.0807561850548
Test Accuracy = 0.960000003974

Residual+Networks±+v2

找不到resnets_utils

找不到因为本地没有，文件缺省了。
去Coursera的课程notebook上下载对应文件。
我的chrome上Coursera的notebook经常上不去，反而用edge效果不错。而且国内连接时断时续，点不进去的时候，点之前点开的notebook，file->open 可以向上退一级文件夹选择需要的文件。并放置到指定位置。

思路

通过构建
identity_block(X, f, filters, stage, block)
convolutional_block(X, f, filters, stage, block, s = 2)
两个模块，并将模块进行堆叠来实现下图的resnet-50

keras 细节

层函数后面的括号代表这层的输入，如果是resnet，记得更改。如下

X_shortcut = Conv2D(F3, (1, 1), strides = (s,s),padding = 'valid', 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)
#这里后面括号注意更改

注意tensor形状相同才行进行层相加操作。

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