code for the paper in ACM Multimedia 2020: Rethinking Generative Zero-Shot Learning: An Ensemble Learning Perspective for Recognising Visual Patches
You can download the dataset CUBird and NABird put the uncompressed dta to the folder "data"
Raw wikipedia article data of CUBird and NABird, as well as detailed merging information of NABird, can be obtained here.
CUBird SCS mode && SCE mode
python run_trainer.py --splitmode easy --dataset CUB2011
python run_trainer.py --splitmode hard --dataset CUB2011
NABird SCS mode && SCE mode python run_trainer.py --splitmode easy --dataset NABird
python run_trainer.py --splitmode hard --dataset NABird