'Change backpropagation scoring method in SKlearn

Is it possible to change the way backpropagation in SKlearn is implemented for MLP? It is implemented by default using the Mean Squared Error to update the network. However, I would like to use the Mean Percentage Error instead. Is that possible, or would I have to use Keras? Any help would be appreciated!

scikit-learndeep-learningneural-networkbackpropagationmlp


Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source

Related Questions