search

fabiodimarco / tf-levenberg-marquardt

Tensorflow implementation of Levenberg-Marquardt training algorithm
MIT License
93 stars 18 forks source link

Random results #14

Closed advanced-bencoding closed 1 year ago

advanced-bencoding commented 1 year ago

Hello, I get a random result everytime I fit the model with my data. Is there a way to specify seed so that i get the same result with respect to a seed? or does this algorithm just work differently?

fabiodimarco commented 1 year ago

Hi, yes you can initialise your model the same way you would initialise a keras model. The training process should be deterministic. Have you checked if the weights are actually initialised equal on each run?