This is the official source code of "BiCAT: Bi-Chronological Augmentation of Transformer for Sequential Recommendation".

Last update: Dec 06, 2022

Overview

BiCAT

This is our TensorFlow implementation for the paper: "BiCAT: Sequential Recommendation with Bidirectional Chronological Augmentation of Transformer". Our code is implemented based on Tensorflow version of SASRec and ASReP.

Environment

TensorFlow 1.12
Python 3.6.*

Datasets Prepare

Benchmarks: Amazon Review datasets Beauty, Movie Lens and Cell_Phones_and_Accessories. The data split is done in the leave-one-out setting. Make sure you download the datasets from the link. Please, use the DataProcessing.py under the data/, and make sure you change the DATASET variable value to your dataset name, then you run:

python DataProcessing.py

You will find the processed dataset in the directory with the name of your input dataset.

Beauty

1. Reversely Pre-training and Short Sequence Augmentation

Pre-train the model and output 20 items for sequences with length <= 20.

python main.py \
       --dataset=Beauty \
       --train_dir=default \
       --lr=0.001 \
       --hidden_units=128 \
       --maxlen=100 \
       --dropout_rate=0.7 \
       --num_blocks=2 \
       --l2_emb=0.0 \
       --num_heads=4 \
       --evalnegsample 100 \
       --reversed 1 \
       --reversed_gen_num 20 \
       --M 20

2. Next-Item Prediction with Reversed-Pre-Trained Model and Augmented dataset

python main.py \
       --dataset=Beauty \
       --train_dir=default \
       --lr=0.001 \
       --hidden_units=128 \
       --maxlen=100 \
       --dropout_rate=0.7 \
       --num_blocks=2 \
       --l2_emb=0.0 \
       --num_heads=4 \
       --evalnegsample 100 \
       --reversed_pretrain 1 \
       --aug_traindata 15 \
       --M 18

Cell_Phones_and_Accessories

1. Reversely Pre-training and Short Sequence Augmentation

Pre-train the model and output 20 items for sequences with length <= 20.

python main.py \
       --dataset=Cell_Phones_and_Accessories \
       --train_dir=default \
       --lr=0.001 \
       --hidden_units=32 \
       --maxlen=100 \
       --dropout_rate=0.5 \
       --num_blocks=2 \
       --l2_emb=0.0 \
       --num_heads=2 \
       --evalnegsample 100 \
       --reversed 1 \
       --reversed_gen_num 20 \
       --M 20

2. Next-Item Prediction with Reversed-Pre-Trained Model and Augmented dataset

python main.py \
       --dataset=Cell_Phones_and_Accessories \
       --train_dir=default \
       --lr=0.001 \
       --hidden_units=32 \
       --maxlen=100 \
       --dropout_rate=0.5 \
       --num_blocks=2 \
       --l2_emb=0.0 \
       --num_heads=2 \
       --evalnegsample 100 \
       --reversed_pretrain 1 \ 
       --aug_traindata 17 \
       --M 18

Citation

@misc{jiang2021sequential,
      title={Sequential Recommendation with Bidirectional Chronological Augmentation of Transformer}, 
      author={Juyong Jiang and Yingtao Luo and Jae Boum Kim and Kai Zhang and Sunghun Kim},
      year={2021},
      eprint={2112.06460},
      archivePrefix={arXiv},
      primaryClass={cs.IR}
}

This is the official source code of "BiCAT: Bi-Chronological Augmentation of Transformer for Sequential Recommendation".

Related tags

Overview

BiCAT

Environment

Datasets Prepare

Beauty

Cell_Phones_and_Accessories

Citation

Owner

John

RITA is a family of autoregressive protein models, developed by LightOn in collaboration with the OATML group at Oxford and the Debora Marks Lab at Harvard.

Sequential model-based optimization with a `scipy.optimize` interface

Finding all things on-prem Microsoft for password spraying and enumeration.

Repository for "Exploring Sparsity in Image Super-Resolution for Efficient Inference", CVPR 2021

SNE-RoadSeg in PyTorch, ECCV 2020

PyTorch code for the paper "FIERY: Future Instance Segmentation in Bird's-Eye view from Surround Monocular Cameras"

The fastest way to visualize GradCAM with your Keras models.

Single-step adversarial training (AT) has received wide attention as it proved to be both efficient and robust.

Code to reproduce the results for Compositional Attention

Implementation of self-attention mechanisms for general purpose. Focused on computer vision modules. Ongoing repository.

CUda Matrix Multiply library.

Data reduction pipeline for KOALA on the AAT.

SHRIMP: Sparser Random Feature Models via Iterative Magnitude Pruning

Simple and Robust Loss Design for Multi-Label Learning with Missing Labels

RMTD: Robust Moving Target Defence Against False Data Injection Attacks in Power Grids

A more easy-to-use implementation of KPConv

Molecular AutoEncoder in PyTorch

Pytorch implementation of the DeepDream computer vision algorithm

SCI-AIDE : High-fidelity Few-shot Histopathology Image Synthesis for Rare Cancer Diagnosis

Optimized primitives for collective multi-GPU communication