HAPZero

• HAPZero: Hierarchical Attribute Prompting Based Zero-Shot Learning for Pest Recognition This repository contains the reference code for the paper "HAPZero: Hierarchical Attribute Prompting Based Zero-Shot Learning for Pest Recognition".

🌈 Model Architecture

📚 Dependencies

• Python 3.6.7
• PyTorch = 1.7.0
• All experiments are performed with one RTX 4090 GPU.

⚡ Prerequisites

• Dataset: please download the dataset, i.e., IP102 to the dataset root path on your machine, Datasets can be download from Xian et al. (CVPR2017) and take them into dir ./datasets/.
• Data split: dataset split files for the three groups are placed at ./data/xlsa19/split
• Attribute w2v: download from link IP102 Att and place it in ./data/xlsa/w2v.
• Download pretranined vision Transformer as the vision encoder and place it in ./pretrain_model_vit.

🚀 Train & Eval

Before running commands, you can set the hyperparameters in config on different datasets:

config/ip102.yaml       #IP102

Train:

 python train.py

Eval:

 python test.py

You can test our trained model: GroupA, [GroupB](please wait), [GroupC](please wait).

📕 Ackowledgement

We thank the following repos providing helpful components in our work. PSVMA