Hi
I'm trying to create repeatable results. I found the sources of randomization here. I made the bayes flag False which removes the dropout layer. I used common weights to start off which removes randomization due to weight initialization. Still my results are varying. Can someone help me with this? I've tried using both SGD and ADAM for optimizers.
Hi
I'm trying to create repeatable results. I found the sources of randomization here. I made the bayes flag False which removes the dropout layer. I used common weights to start off which removes randomization due to weight initialization. Still my results are varying. Can someone help me with this? I've tried using both SGD and ADAM for optimizers.