如何在Keras中创建自定义目标函数？

2 回答

• 17

  这是我的小片段，用于编写新的损失函数并在使用之前测试它们：

  import numpy as np
  
  from keras import backend as K
  
  _EPSILON = K.epsilon()
  
  def _loss_tensor(y_true, y_pred):
      y_pred = K.clip(y_pred, _EPSILON, 1.0-_EPSILON)
      out = -(y_true * K.log(y_pred) + (1.0 - y_true) * K.log(1.0 - y_pred))
      return K.mean(out, axis=-1)
  
  def _loss_np(y_true, y_pred):
      y_pred = np.clip(y_pred, _EPSILON, 1.0-_EPSILON)
      out = -(y_true * np.log(y_pred) + (1.0 - y_true) * np.log(1.0 - y_pred))
      return np.mean(out, axis=-1)
  
  def check_loss(_shape):
      if _shape == '2d':
          shape = (6, 7)
      elif _shape == '3d':
          shape = (5, 6, 7)
      elif _shape == '4d':
          shape = (8, 5, 6, 7)
      elif _shape == '5d':
          shape = (9, 8, 5, 6, 7)
  
      y_a = np.random.random(shape)
      y_b = np.random.random(shape)
  
      out1 = K.eval(_loss_tensor(K.variable(y_a), K.variable(y_b)))
      out2 = _loss_np(y_a, y_b)
  
      assert out1.shape == out2.shape
      assert out1.shape == shape[:-1]
      print np.linalg.norm(out1)
      print np.linalg.norm(out2)
      print np.linalg.norm(out1-out2)
  
  
  def test_loss():
      shape_list = ['2d', '3d', '4d', '5d']
      for _shape in shape_list:
          check_loss(_shape)
          print '======================'
  
  if __name__ == '__main__':
      test_loss()
  

  在这里你可以看到我正在测试binary_crossentropy损失，并且定义了2个单独的损失，一个numpy版本（_loss_np）另一个张量版本（_loss_tensor）[注意：如果你只是使用keras函数那么它将适用于Theano和Tensorflow ...但如果您依赖其中一个，您也可以通过K.theano.tensor.function或K.tf.function引用它们]

  后来我比较输出形状和输出的L2范数（应该几乎相等）和差分的L2范数（应该朝向0）

  一旦您对损失功能正常工作感到满意，您可以将其用作：

  model.compile(loss=_loss_tensor, optimizer=sgd)
  

  回复于 2024-05-12T08:35:29+08:00
• 3

  （答案已修复）一个简单的方法是调用Keras后端：

  import keras.backend as K
  
  def custom_loss(y_true,y_pred):
      return K.mean((y_true - y_pred)**2)
  

  然后：

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

  等于

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

  回复于 2024-05-12T08:35:29+08:00