Supervised Contrastive Learning for Product Matching

Overview

Contrastive Product Matching

This repository contains the code and data download links to reproduce the experiments of the paper "Supervised Contrastive Learning for Product Matching" by Ralph Peeters and Christian Bizer. ArXiv link. A comparison of the results to other systems using different benchmark datasets is found at Papers with Code - Entity Resolution.

  • Requirements

    Anaconda3

    Please keep in mind that the code is not optimized for portable or even non-workstation devices. Some of the scripts may require large amounts of RAM (64GB+) and GPUs. It is advised to use a powerful workstation or server when experimenting with some of the larger files.

    The code has only been used and tested on Linux (CentOS) servers.

  • Building the conda environment

    To build the exact conda environment used for the experiments, navigate to the project root folder where the file contrastive-product-matching.yml is located and run conda env create -f contrastive-product-matching.yml

    Furthermore you need to install the project as a package. To do this, activate the environment with conda activate contrastive-product-matching, navigate to the root folder of the project, and run pip install -e .

  • Downloading the raw data files

    Navigate to the src/data/ folder and run python download_datasets.py to automatically download the files into the correct locations. You can find the data at data/raw/

    If you are only interested in the separate datasets, you can download the WDC LSPC datasets and the deepmatcher splits for the abt-buy and amazon-google datasets on the respective websites.

  • Processing the data

    To prepare the data for the experiments, run the following scripts in that order. Make sure to navigate to the respective folders first.

    1. src/processing/preprocess/preprocess_corpus.py
    2. src/processing/preprocess/preprocess_ts_gs.py
    3. src/processing/preprocess/preprocess_deepmatcher_datasets.py
    4. src/processing/contrastive/prepare_data.py
    5. src/processing/contrastive/prepare_data_deepmatcher.py
  • Running the Contrastive Pre-training and Cross-entropy Fine-tuning

    Navigate to src/contrastive/

    You can find respective scripts for running the experiments of the paper in the subfolders lspc/ abtbuy/ and amazongoogle/. Note that you need to adjust the file path in these scripts for your system (replace your_path with path/to/repo).

    • Contrastive Pre-training

      To run contrastive pre-training for the abtbuy dataset for example use

      bash abtbuy/run_pretraining_clean_roberta.sh BATCH_SIZE LEARNING_RATE TEMPERATURE (AUG)

      You need to specify batch site, learning rate and temperature as arguments here. Optionally you can also apply data augmentation by passing an augmentation method as last argument (use all- for the augmentation used in the paper).

      For the WDC Computers data you need to also supply the size of the training set, e.g.

      bash lspc/run_pretraining_roberta.sh BATCH_SIZE LEARNING_RATE TEMPERATURE TRAIN_SIZE (AUG)

    • Cross-entropy Fine-tuning

      Finally, to use the pre-trained models for fine-tuning, run any of the fine-tuning scripts in the respective folders, e.g.

      bash abtbuy/run_finetune_siamese_frozen_roberta.sh BATCH_SIZE LEARNING_RATE TEMPERATURE (AUG)

      Please note, that BATCH_SIZE refers to the batch size used in pre-training. The fine-tuning batch size is locked to 64 but can be adjusted in the bash scripts if needed.

      Analogously for fine-tuning WDC computers, add the train size:

      bash lspc/run_finetune_siamese_frozen_roberta.sh BATCH_SIZE LEARNING_RATE TEMPERATURE TRAIN_SIZE (AUG)


Project based on the cookiecutter data science project template. #cookiecutterdatascience

Owner
Web-based Systems Group @ University of Mannheim
We explore technical and empirical questions concerning the development of global, decentralized information environments.
Web-based Systems Group @ University of Mannheim
paper: Hyperspectral Remote Sensing Image Classification Using Deep Convolutional Capsule Network

DC-CapsNet This is a tensorflow and keras based implementation of DC-CapsNet for HSI in the Remote Sensing Letters R. Lei et al., "Hyperspectral Remot

LEI 7 Nov 29, 2022
A map update dataset and benchmark

MUNO21 MUNO21 is a dataset and benchmark for machine learning methods that automatically update and maintain digital street map datasets. Previous dat

16 Nov 30, 2022
Deep Learning GPU Training System

DIGITS DIGITS (the Deep Learning GPU Training System) is a webapp for training deep learning models. The currently supported frameworks are: Caffe, To

NVIDIA Corporation 4.1k Jan 03, 2023
[CoRL 2021] A robotics benchmark for cross-embodiment imitation.

x-magical x-magical is a benchmark extension of MAGICAL specifically geared towards cross-embodiment imitation. The tasks still provide the Demo/Test

Kevin Zakka 36 Nov 26, 2022
Image Restoration Using Swin Transformer for VapourSynth

SwinIR SwinIR function for VapourSynth, based on https://github.com/JingyunLiang/SwinIR. Dependencies NumPy PyTorch, preferably with CUDA. Note that t

Holy Wu 11 Jun 19, 2022
[CVPR 2022] Thin-Plate Spline Motion Model for Image Animation.

[CVPR2022] Thin-Plate Spline Motion Model for Image Animation Source code of the CVPR'2022 paper "Thin-Plate Spline Motion Model for Image Animation"

yoyo-nb 1.4k Dec 30, 2022
Official code for: A Probabilistic Hard Attention Model For Sequentially Observed Scenes

"A Probabilistic Hard Attention Model For Sequentially Observed Scenes" Authors: Samrudhdhi Rangrej, James Clark Accepted to: BMVC'21 A recurrent atte

5 Nov 19, 2022
Code artifacts for the submission "Mind the Gap! A Study on the Transferability of Virtual vs Physical-world Testing of Autonomous Driving Systems"

Code Artifacts Code artifacts for the submission "Mind the Gap! A Study on the Transferability of Virtual vs Physical-world Testing of Autonomous Driv

Andrea Stocco 2 Aug 24, 2022
Seasonal Contrast: Unsupervised Pre-Training from Uncurated Remote Sensing Data

Seasonal Contrast: Unsupervised Pre-Training from Uncurated Remote Sensing Data This is the official PyTorch implementation of the SeCo paper: @articl

ElementAI 101 Dec 12, 2022
用opencv的dnn模块做yolov5目标检测,包含C++和Python两个版本的程序

yolov5-dnn-cpp-py yolov5s,yolov5l,yolov5m,yolov5x的onnx文件在百度云盘下载, 链接:https://pan.baidu.com/s/1d67LUlOoPFQy0MV39gpJiw 提取码:bayj python版本的主程序是main_yolov5.

365 Jan 04, 2023
Implemenets the Contourlet-CNN as described in C-CNN: Contourlet Convolutional Neural Networks, using PyTorch

C-CNN: Contourlet Convolutional Neural Networks This repo implemenets the Contourlet-CNN as described in C-CNN: Contourlet Convolutional Neural Networ

Goh Kun Shun (KHUN) 10 Nov 03, 2022
Chainer Implementation of Fully Convolutional Networks. (Training code to reproduce the original result is available.)

fcn - Fully Convolutional Networks Chainer implementation of Fully Convolutional Networks. Installation pip install fcn Inference Inference is done as

Kentaro Wada 218 Oct 27, 2022
PipeTransformer: Automated Elastic Pipelining for Distributed Training of Large-scale Models

PipeTransformer: Automated Elastic Pipelining for Distributed Training of Large-scale Models This repository is the official implementation of the fol

DistributedML 41 Dec 06, 2022
Code for Understanding Pooling in Graph Neural Networks

Select, Reduce, Connect This repository contains the code used for the experiments of: "Understanding Pooling in Graph Neural Networks" Setup Install

Daniele Grattarola 37 Dec 13, 2022
Keras implementation of the GNM model in paper ’Graph-Based Semi-Supervised Learning with Nonignorable Nonresponses‘

Graph-based joint model with Nonignorable Missingness (GNM) This is a Keras implementation of the GNM model in paper ’Graph-Based Semi-Supervised Lear

Fan Zhou 2 Apr 17, 2022
Efficient Conformer: Progressive Downsampling and Grouped Attention for Automatic Speech Recognition

Efficient Conformer: Progressive Downsampling and Grouped Attention for Automatic Speech Recognition Official implementation of the Efficient Conforme

Maxime Burchi 145 Dec 30, 2022
CLIP: Connecting Text and Image (Learning Transferable Visual Models From Natural Language Supervision)

CLIP (Contrastive Language–Image Pre-training) Experiments (Evaluation) Model Dataset Acc (%) ViT-B/32 (Paper) CIFAR100 65.1 ViT-B/32 (Our) CIFAR100 6

Myeongjun Kim 52 Jan 07, 2023
Automatically creates genre collections for your Plex media

Plex Auto Genres Plex Auto Genres is a simple script that will add genre collection tags to your media making it much easier to search for genre speci

Shane Israel 63 Dec 31, 2022
Source code for Fixed-Point GAN for Cloud Detection

FCD: Fixed-Point GAN for Cloud Detection PyTorch source code of Nyborg & Assent (2020). Abstract The detection of clouds in satellite images is an ess

Joachim Nyborg 8 Dec 22, 2022
Code for "Causal autoregressive flows" - AISTATS, 2021

Code for "Causal Autoregressive Flow" This repository contains code to run and reproduce experiments presented in Causal Autoregressive Flows, present

Ricardo Pio Monti 35 Dec 16, 2022