search

renmengye / few-shot-ssl-public

Meta Learning for Semi-Supervised Few-Shot Classification
Other
553 stars 101 forks source link

A doubt regarding the test data. #10

Closed Hsankesara closed 5 years ago

Hsankesara commented 5 years ago

Hey @renmengye, I have a doubt regarding the paper. Suppose we trained the model by taking 5 classes in each episode. I have total of 40 classes. After training, I have test data where I have to classify each image to one of the 40 categories. Now how can I do that? The most logical way is to calculate a noormalized probability for each class and then assigned an image to a class with the highest probability. But this somehow looks not the right way as we trained the model for only 5 classes. Kindly help me here Regards

Duan-JM commented 5 years ago

During the test stage, if u select 5 way, it means that the unlabeled query samples are in this 5 way. For each episode in the test, we still random pick novel classes.

Hsankesara commented 5 years ago

Thanks, @VDeamoV .