Hi, I notice in your original paper's experimental results part, you have written that the RELATION NET's result is
57.02 for 1-shot
71.07 for 5-shot
however, in the original paper "Learning to compare: Relation network for few-shot learning", the result is 50.44 and 65.32 respectively. So I wonder how you get the better results? Could you share this part of code? I will be very appreciate for your generous. Thanks a lot!
(my email is 729406307@qq.com)
Hi, I notice in your original paper's experimental results part, you have written that the RELATION NET's result is 57.02 for 1-shot 71.07 for 5-shot however, in the original paper "Learning to compare: Relation network for few-shot learning", the result is 50.44 and 65.32 respectively. So I wonder how you get the better results? Could you share this part of code? I will be very appreciate for your generous. Thanks a lot! (my email is 729406307@qq.com)