search

brjathu / iTAML

Official implementation of "iTAML : An Incremental Task-Agnostic Meta-learning Approach". CVPR 2020
96 stars 16 forks source link

A question about theta and task-specific phi #21

Open LeungWaiHo opened 3 years ago

LeungWaiHo commented 3 years ago

Dear author: I have read your paper and feel interest about your task-agnostic setting, I see PHI = {theta, phi} where phi =[phi_1^T phi_2^T ...]^T and PHI_i = {theta, phi_i} , but when I read your code I could not find out (maybe I don`t know) where to separate the parameter phi to phi_1 and phi_2 to create PHI_i for training task-specific model in the inner loop.

LeungWaiHo commented 3 years ago

https://github.com/brjathu/iTAML/issues/13#issuecomment-664729648

and I don`t understand why only calculate the loss for only that task (such as task 0) can only update the classfication fc head parameters.