{"id":16750609,"url":"https://github.com/mberr/ea-sota-comparison","last_synced_at":"2025-08-09T10:09:00.993Z","repository":{"id":55423255,"uuid":"308658939","full_name":"mberr/ea-sota-comparison","owner":"mberr","description":"Code for paper \"A Critical Assessment of State-of-the-Art in Entity Alignment\" (https://arxiv.org/abs/2010.16314)","archived":false,"fork":false,"pushed_at":"2023-04-01T10:06:19.000Z","size":110,"stargazers_count":16,"open_issues_count":0,"forks_count":2,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-04-10T14:35:34.189Z","etag":null,"topics":["entity-alignment","graph-neural-network","knowledge-graph","word-embedding-evaluation"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/mberr.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2020-10-30T14:44:54.000Z","updated_at":"2024-01-04T16:51:53.000Z","dependencies_parsed_at":"2022-08-15T00:00:49.948Z","dependency_job_id":"42afcd93-1de4-4d2f-9634-87bde8195575","html_url":"https://github.com/mberr/ea-sota-comparison","commit_stats":null,"previous_names":[],"tags_count":3,"template":false,"template_full_name":null,"purl":"pkg:github/mberr/ea-sota-comparison","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mberr%2Fea-sota-comparison","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mberr%2Fea-sota-comparison/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mberr%2Fea-sota-comparison/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mberr%2Fea-sota-comparison/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/mberr","download_url":"https://codeload.github.com/mberr/ea-sota-comparison/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/mberr%2Fea-sota-comparison/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":269561838,"owners_count":24438430,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-08-09T02:00:10.424Z","response_time":111,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["entity-alignment","graph-neural-network","knowledge-graph","word-embedding-evaluation"],"created_at":"2024-10-13T02:28:47.724Z","updated_at":"2025-08-09T10:09:00.907Z","avatar_url":"https://github.com/mberr.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# A Critical Assessment of State-of-the-Art in Entity Alignment\n\n[![Arxiv](https://img.shields.io/badge/arXiv-2010.16314-b31b1b)](https://arxiv.org/abs/2010.16314)\n[![Python 3.8](https://img.shields.io/badge/Python-3.8-2d618c?logo=python)](https://docs.python.org/3.8/)\n[![PyTorch](https://img.shields.io/badge/Made%20with-PyTorch-ee4c2c?logo=pytorch)](https://pytorch.org/docs/stable/index.html)\n[![License: MIT](https://img.shields.io/badge/License-MIT-green.svg)](https://opensource.org/licenses/MIT)\n\nThis repository contains the source code for the paper\n```\nA Critical Assessment of State-of-the-Art in Entity Alignment\nMax Berrendorf, Ludwig Wacker, and Evgeniy Faerman\nhttps://arxiv.org/abs/2010.16314\n```\n\n# Installation\nSetup and activate virtual environment:\n```shell script\npython3.8 -m venv ./venv\nsource ./venv/bin/activate\n```\n\nInstall requirements (in this virtual environment):\n```shell script\npip install -U pip\npip install -U -r requirements.txt\n```\n\nIn order to run the DGMC scripts, you additionally need to setup \nits requirements as described in the corresponding GitHub repository's \n[README](https://github.com/rusty1s/deep-graph-matching-consensus/blob/a25f89751f4a3a0d509baa6bbada8b4153c635f6/README.md).\nWe do not include them into [`requirements.txt`](./requirements.txt), \nsince their installation is a bit more involved, including non-Python dependencies. \n\n# Preparation\n\n## MLFlow\nIn order to track results to a MLFlow server, start it first by running\n```shell script\nmlflow server\n```\n_Note: When storing the result for many configurations, we recommend to setup a\ndatabase backend following the [instructions](https://mlflow.org/docs/latest/tracking.html)._\nFor the following examples, we assume that the server is running at\n```shell script\nTRACKING_URI=http://localhost:5000\n```\n\n## OpenEA RDGCN embeddings\n[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.6985518.svg)](https://doi.org/10.5281/zenodo.6985518)\n\nPlease download the RDGCN embeddings extracted with the [OpenEA codebase](https://github.com/nju-websoft/OpenEA/tree/2a6e0b03ec8cdcad4920704d1c38547a3ad72abe)\nfrom [here](https://doi.org/10.5281/zenodo.6985518)\nand place them in `~/.kgm/openea_rdgcn_embeddings`.\nThey have a file name matching the pattern `*_*_15K_V2.pt` and require in total around 160MiB storage.\n\n## BERT initialization\n[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.6985518.svg)](https://doi.org/10.5281/zenodo.6985518)\n\nTo generate data for the BERT-based initialization, run\n```shell script\n(venv) PYTHONPATH=./src python3 executables/prepare_bert.py\n```\n\nWe also provide preprocessed files at [this url](https://doi.org/10.5281/zenodo.6985518).\nIf you prefer to use those, please download and place them in `~/.kgm/bert_prepared`. \nThey have a file name matching `*_bert-base-multilingual-cased_*` and require in total around 6.1GiB storage. \n\n# Experiments\n\nFor all experiments the results are logged to the running MLFlow instance.\n\n_Note: The hyperparameter searches takes a significant amount of time (~multiple days),\n and requires access to GPU(s). You can abort the script at any time, and inspect the\n  current results via the web interface of MLFlow._\n\n\n## Zero-Shot\nFor the zero-shot evaluation run\n```shell script\n(venv) PYTHONPATH=./src python3 executables/zero_shot.py --tracking_uri=${TRACKING_URI} \n```\n\n## GCN-Align\nTo run the hyperparameter search run\n```shell script\n(venv) PYTHONPATH=./src python3 executables/tune_gcn_align.py --tracking_uri=${TRACKING_URI} \n```\n\n## RDGCN\nTo run the hyperparameter search run\n```shell script\n(venv) PYTHONPATH=./src python3 executables/tune_rdgcn.py --tracking_uri=${TRACKING_URI} \n```\n\n## DGMC\nTo run the hyperparameter search run\n```shell script\n(venv) PYTHONPATH=./src python3 executables/tune_dgmc.py  --tracking_uri=${TRACKING_URI} \n```\n\n# Evaluation\nTo summarize the dataset statistics run\n```shell script\n(venv) PYTHONPATH=./src python3 executables/summarize.py --target datasets --force\n```\n\nTo summarize all experiments run\n```shell script\n(venv) PYTHONPATH=./src python3 executables/summarize.py --target results --tracking_uri=${TRACKING_URI} --force\n```\n\nTo generate the ablation study table run\n```shell script\n(venv) PYTHONPATH=./src python3 executables/summarize.py --target ablation --tracking_uri=${TRACKING_URI} --force\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmberr%2Fea-sota-comparison","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmberr%2Fea-sota-comparison","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmberr%2Fea-sota-comparison/lists"}