from keras import backend as K
from keras.engine.topology import Layer
 
class MyLayer(Layer):
 
  def __init__(self, output_dim, **kwargs):
    self.output_dim = output_dim
    super(MyLayer, self).__init__(**kwargs)
 
  def build(self, input_shape):
    # 為該層創(chuàng)建一個可訓練的權重
    self.kernel = self.add_weight(name='kernel', 
                   shape=(input_shape[1], self.output_dim),
                   initializer='uniform',
                   trainable=True)
    super(MyLayer, self).build(input_shape) # 一定要在最后調用它
 
  def call(self, x):
    return K.dot(x, self.kernel)
 
  def compute_output_shape(self, input_shape):
    return (input_shape[0], self.output_dim)

from keras import backend as K
from keras.layers import Layer
 
class MyLayer(Layer):
 
  def __init__(self, output_dim, **kwargs):
    self.output_dim = output_dim
    super(MyLayer, self).__init__(**kwargs)
 
  def build(self, input_shape):
    # Create a trainable weight variable for this layer.
    self.kernel = self.add_weight(name='kernel', 
                   shape=(input_shape[1], self.output_dim),
                   initializer='uniform',
                   trainable=True)
    super(MyLayer, self).build(input_shape) # Be sure to call this at the end
 
  def call(self, x):
    return K.dot(x, self.kernel)
 
  def compute_output_shape(self, input_shape):
    return (input_shape[0], self.output_dim)

#losses.py
#y_true是分類的標簽，y_pred是分類中預測值（這里指，模型最后一層為softmax層，輸出的是每個類別的預測值）
def mean_squared_error(y_true, y_pred):
  return K.mean(K.square(y_pred - y_true), axis=-1)
def mean_absolute_error(y_true, y_pred):
  return K.mean(K.abs(y_pred - y_true), axis=-1)
def mean_absolute_percentage_error(y_true, y_pred):
  diff = K.abs((y_true - y_pred) / K.clip(K.abs(y_true),K.epsilon(),None))
  return 100. * K.mean(diff, axis=-1)
def mean_squared_logarithmic_error(y_true, y_pred):
  first_log = K.log(K.clip(y_pred, K.epsilon(), None) + 1.)
  second_log = K.log(K.clip(y_true, K.epsilon(), None) + 1.)
  return K.mean(K.square(first_log - second_log), axis=-1)
def squared_hinge(y_true, y_pred):
  return K.mean(K.square(K.maximum(1. - y_true * y_pred, 0.)), axis=-1)

def get_loss(labels,features, alpha,lambda_c,lambda_g,num_classes):
  #由于涉及研究內容，詳細代碼不做公開
  return loss
#total_loss(y_true,y_pred)，y_true代表標簽（類別），y_pred代表模型的輸出
#（ 如果是模型中間層輸出，即代表特征，如果模型輸出是經(jīng)過softmax就是代表分類預測值）
#其他有需要的參數(shù)也可以寫在里面
def total_loss(y_true,y_pred):
    git_loss=get_loss(y_true,y_pred,alpha=0.5,lambda_c=0.001,lambda_g=0.001,num_classes=45)
    return git_loss 

#這里使用vgg16模型
model = VGG16(input_tensor=image_input, include_top=True,weights='imagenet')
model.summary()
#fc2層輸出為特征
last_layer = model.get_layer('fc2').output
#獲取特征
feature = last_layer
#softmax層輸出為各類的預測值
out = Dense(num_classes,activation = 'softmax',name='predictions')(last_layer)
#該模型有一個輸入image_input,兩個輸出out，feature
custom_vgg_model = Model(inputs = image_input, outputs = [feature,out])
custom_vgg_model.summary()
#優(yōu)化器，梯度下降
sgd = optimizers.SGD(lr=learn_Rate,decay=decay_Rate,momentum=0.9,nesterov=True)
#這里面，剛才有兩個輸出，這里面使用兩個損失函數(shù)，total_loss對應的是fc2層輸出的特征
#categorical_crossentropy對應softmax層的損失函數(shù)
#loss_weights兩個損失函數(shù)的權重
custom_vgg_model.compile(loss={'fc2': 'total_loss','predictions': "categorical_crossentropy"},
             loss_weights={'fc2': 1, 'predictions':1},optimizer= sgd,
                   metrics={'predictions': 'accuracy'})
#這里使用dummy1，dummy2做演示，為0
dummy1 = np.zeros((y_train.shape[0],4096))
dummy2 = np.zeros((y_test.shape[0],4096))
#模型的輸入輸出必須和model.fit()中x，y兩個參數(shù)維度相同
#dummy1的維度和fc2層輸出的feature維度相同，y_train和softmax層輸出的預測值維度相同
#validation_data驗證數(shù)據(jù)集也是如此，需要和輸出層的維度相同
hist = custom_vgg_model.fit(x = X_train,y = {'fc2':dummy1,'predictions':y_train},batch_size=batch_Sizes,
                epochs=epoch_Times, verbose=1,validation_data=(X_test, {'fc2':dummy2,'predictions':y_test}))