zhaijianyang / KP4SR

KP4SR

The source code for our paper: Knowledge Prompt-tuning for Sequential Recommendation.

Overview

Preparation

Our code runs on the Ascend 910 NPU of Huawei modelarts.

• Ascend 910ProA
• torch == 1.8.1+ascend.rc3.20221020
• torch-npu == 1.8.1rc3.post20221020

Usage

• train: run scripts/modelarts_train.sh
• test: run scripts/modelarts_test.sh