search

chenghao-ch94 / AGNN

[MIPR 2022 & TMM 2023] "Attentive Graph Neural Networks for Few-shot Learning" with its extension version
15 stars 5 forks source link

How can reproduce the results from the 2023 TMM paper？ #1

Open Zlh111111 opened 1 month ago

Zlh111111 commented 1 month ago

Under the 5-way 1-shot setting on the mini dataset, the source code can only reproduce the results of the inductive setting and is far from achieving the results of the transductive setting. The code should be under the transductive settings, or how should I set it up as a transductive setting？ text

chenghao-ch94 commented 1 month ago

Thank you for your interest in our work. You can follow the script in the README file in the folder 'Extension'.