Hello,
Thank you for this great work and well-written paper.
When training the linear classifier (e.g. Alexnet with ImageNet), how do you perform the train/test split? 1. Do you use all of the classes in your test set?
2. How do you choose views for each sample?
Many thanks,
Jonny
Hello,
Thank you for this great work and well-written paper.
When training the linear classifier (e.g. Alexnet with ImageNet), how do you perform the train/test split? 1. Do you use all of the classes in your test set?
2. How do you choose views for each sample?
Many thanks,
Jonny