I'm using the feedforward neural network with 100 inputs and 10 units in hidden layer and one output neural network. I train the network several times using the same input training data and the same network architecture/settings with random initialize but I understand that there will be differences in the weightings produced within the NN each time and that no two neural networks will be identical, but what can I try to produce networks that are more consistent across each train, given the identical data?
Try Kessler rule , it will propagate the 100 training example in one vector , which will guide you to the weights which classify all in right way .... then test with some of training examples
wish it helps
To obtain indentical ANN, you just have to fix the seed of the random function.
The right question is why you would like to obtain identical ANN ?
Hello Raphaël Feraud
the seed of random function is always [-0.01 0.01 ] but I also have problem
- Badges
- Science topic