Python-Deep-Learning
Python-Deep-Learning copied to clipboard
Hyperparameter tuning
One question that I keep getting from students is why do we choose certain hyperparameters. I keep explaining that it's an art form also there's a lot of research around it and sometimes we follow that's in the documentation for a particular model or function.
It would be helpful to add a comment explaining why we chose soft max or why we chose 64 neurons etc