I've been working on optimizing a neural network. I cannot use the built in routines per se since the the ANN is embedded in a constrained optimization. It would be nice to have a pair of functions to take the ANN parameters (weights, biases) and flatten it into a vector, then to take a vector and unflatten it into a NN. I've done this manually myself for an ANN with 1 hidden layer. But I want to explore a deeper NN, and it'd be nice to have this function pair.
MATLAB: Flatten and unflatten a neural network
constrained optimizationneural networkoptimization
Related Question
- I need help with a neural network toolbox.
- How to generate the weights and bias values after training a network
- Normalize Inputs and Targets of neural network
- Neural network about weights
- When I am the training a elman neural network it is working only with 5 or lower no. of hidden layers, as I tried to increase no of layers its giving only 50% accuracy which means network is not trained proparly. what could probably go wraong
- How to get neural net weight matrix
Best Answer