In the first phase (representation learning), the CNN (
) is trained with a large dataset, 
, that trains the parameters, 
and 
, of the network. After this, these parameters, 
, are fixed and shared across later phases. The activations ...
Representational learning
Get Hands-On One-shot Learning with Python now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
