I have a question here…Why we didn’t specify image size as (224,224,3) instead of
“IMAGE_SIZE + (3,)”
Is there a reason for this format?
Image size in tf hub
Course Q&A
TensorFlow: Data and Deployment Specialization
Advanced Deployment Scenarios with TensorFlow
Its probably because if you want to feed diffetent sizes in case you change your application, of course if this model allows it.
Or perhaps placing a variable in the model so its not changed accidentally
…