Custom loss function in Keras

by

All you have to do is define a function for that, using keras backend functions for calculations. The function must take the true values and the model predicted values.

Now, since I’m not sure about what are g, q, x an y in your function, I’ll just create a basic example here without caring about what it means or whether it’s an actual useful function:

import keras.backend as K

def customLoss(yTrue,yPred):
    return K.sum(K.log(yTrue) - K.log(yPred))

All backend functions can be seen here.

After that, compile your model using that function instead of a regular one:

model.compile(loss=customLoss, optimizer = .....)

More Related Contents:

  1. Keras, How to get the output of each layer?
  2. How to apply gradient clipping in TensorFlow?
  3. Keras replacing input layer
  4. Keras custom loss function: Accessing current input pattern
  5. How can I use a pre-trained neural network with grayscale images?
  6. Dimension of shape in conv1D
  7. Removing then Inserting a New Middle Layer in a Keras Model
  8. How do I get the weights of a layer in Keras?
  9. Multiple outputs in Keras
  10. ValueError: Input 0 of layer sequential is incompatible with the layer: : expected min_ndim=4, found ndim=3. Full shape received: [8, 28, 28]
  11. Keras LSTM input dimension setting
  12. How to implement Grad-CAM on a trained network
  13. How to fix “ResourceExhaustedError: OOM when allocating tensor”
  14. How do I plot a Keras/Tensorflow subclassing API model?
  15. Strange behaviour of the loss function in keras model, with pretrained convolutional base
  16. Multi-input Multi-output Model with Keras Functional API
  17. How to export Keras .h5 to tensorflow .pb?
  18. Keras + Tensorflow and Multiprocessing in Python
  19. Does Any one got “AttributeError: ‘str’ object has no attribute ‘decode’ ” , while Loading a Keras Saved Model
  20. How does Keras 1d convolution layer work with word embeddings – text classification problem? (Filters, kernel size, and all hyperparameter)
  21. ValueError at /image/ Tensor Tensor(“activation_5/Softmax:0”, shape=(?, 4), dtype=float32) is not an element of this graph
  22. What do I need K.clear_session() and del model for (Keras with Tensorflow-gpu)?
  23. PyTorch NotImplementedError in forward
  24. Why can’t I get reproducible results in Keras even though I set the random seeds?
  25. AttributeError: module ‘tensorflow.python.keras.utils.generic_utils’ has no attribute ‘populate_dict_with_module_objects’
  26. What is the role of TimeDistributed layer in Keras?
  27. NotImplementedError: Cannot convert a symbolic Tensor (2nd_target:0) to a numpy array
  28. How to use TensorFlow metrics in Keras
  29. How to visualize RNN/LSTM gradients in Keras/TensorFlow?
  30. How is the Keras Conv1D input specified? I seem to be lacking a dimension

Leave a Comment