Deep Learning Dictionary - Lightweight Crash Course

CNN Learnable Parameters - Deep Learning Dictionary

We've already learned about learnable parameters as being the parameters within a network that are learned by the network during training. The bulk of these parameters are made up of the network's weights and biases.

The total number of learnable parameters influences the size of the network on disk and in-memory, and we can generally use the number of learnable parameters as a rough measure of model complexity, with more complex models having a larger number of learnable parameters.