loss_blog_2

second.py

class model:
    
    def __init__(self):
        xavier=tf.keras.initializers.GlorotUniform()
        self.l1=tf.keras.layers.Dense(64,kernel_initializer=xavier,activation=tf.nn.relu,input_shape=[1])
        self.l2=tf.keras.layers.Dense(64,kernel_initializer=xavier,activation=tf.nn.relu)
        self.out=tf.keras.layers.Dense(1,kernel_initializer=xavier)
        self.train_op = tf.keras.optimizers.Adagrad(learning_rate=0.1)
        
    # Running the model
    def run(self,X):
        boom=self.l1(X)
        boom1=self.l2(boom)
        boom2=self.out(boom1)
        return boom2
      
    #Custom loss fucntion
    def get_loss(self,X,Y):
        boom=self.l1(X)
        boom1=self.l2(boom)
        boom2=self.out(boom1)
        return tf.math.square(boom2-Y)
      
    # get gradients
    def get_grad(self,X,Y):
        with tf.GradientTape() as tape:
            tape.watch(self.l1.variables)
            tape.watch(self.l2.variables)
            tape.watch(self.out.variables)
            L = self.get_loss(X,Y)
            g = tape.gradient(L, [self.l1.variables[0],self.l1.variables[1],self.l2.variables[0],self.l2.variables[1],self.out.variables[0],self.out.variables[1]])
        return g
      
    # perform gradient descent
    def network_learn(self,X,Y):
        g = self.get_grad(X,Y)
        self.train_op.apply_gradients(zip(g, [self.l1.variables[0],self.l1.variables[1],self.l2.variables[0],self.l2.variables[1],self.out.variables[0],self.out.variables[1]]))