{"id":24686965,"url":"https://github.com/query/pymeant","last_synced_at":"2025-10-14T13:32:30.356Z","repository":{"id":18201302,"uuid":"21331508","full_name":"query/pymeant","owner":"query","description":"A proof-of-concept Python implementation of a simplified version of the MEANT machine translation evaluation metric.","archived":false,"fork":false,"pushed_at":"2014-06-29T20:21:33.000Z","size":108,"stargazers_count":1,"open_issues_count":0,"forks_count":1,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-07-15T07:40:36.502Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"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/query.png","metadata":{"files":{"readme":"README.markdown","changelog":null,"contributing":null,"funding":null,"license":"COPYING","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2014-06-29T20:21:17.000Z","updated_at":"2018-09-27T09:17:58.000Z","dependencies_parsed_at":"2022-07-26T21:32:04.279Z","dependency_job_id":null,"html_url":"https://github.com/query/pymeant","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/query/pymeant","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/query%2Fpymeant","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/query%2Fpymeant/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/query%2Fpymeant/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/query%2Fpymeant/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/query","download_url":"https://codeload.github.com/query/pymeant/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/query%2Fpymeant/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":271780454,"owners_count":24819292,"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":"2025-08-24T02:00:11.135Z","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":[],"created_at":"2025-01-26T16:17:25.235Z","updated_at":"2025-10-14T13:32:25.295Z","avatar_url":"https://github.com/query.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# PyMEANT\n\nPyMEANT is a proof-of-concept Python implementation of a simplified\nversion of the MEANT machine translation evaluation metric presented\nin Lo et al. (2012).\nIt was originally submitted as a final course project for [the Machine\nTranslation class][mt-class] at Johns Hopkins University in spring 2014.\nYou may wish to read [the project writeup][writeup] for more details.\n\n[mt-class]: http://mt-class.org/jhu/\n[writeup]: https://github.com/query/mt-submissions/raw/master/project/writeup.pdf\n\n\n## Caveats\n\nBefore using PyMEANT, please note the following:\n\n* PyMEANT is an unoptimized pure-Python implementation, and as a result\n  can be very slow even on modest data sets.\n\n* Predicate and argument weighting are not implemented.\n  Thus, PyMEANT's results cannot be directly compared with MEANT's.\n\n* Jaccard similarity is used as the lexical similarity measurement, as\n  described in Tumuluru et al. (2012), instead of the MinMax-MI metric\n  outlined in the original paper.\n\n\n## Usage\n\nTo install, use `setup.py`:\n\n    $ python setup.py install\n\nBefore scoring translation hypotheses, you will need to train a lexical\nsimilarity model using `python -m pymeant train`.\nA parser for Gigaword corpus files is included for convenience:\n\n    $ python -m pymeant.formats.gigaword nyt199504.gz | python -m pymeant train - lexsim.pkl\n\nTo perform the actual scoring, use `python -m pymeant score`, passing in\nthe hypotheses and reference sentences as both plain text (one per line)\nand [ASSERT][assert]-tagged parse files:\n\n    $ python -m pymeant score lexsim.pkl hypotheses.{txt,parse} reference.{txt,parse}\n\nFor further information, pass the `--help` option.\n\n[assert]: http://cemantix.org/software/assert.html\n\n\n## References\n\n* Chi-kiu Lo, Anand Karthik Tumuluru, and Dekai Wu.\n  2012.\n  [Fully automatic semantic MT evaluation][W12-3129].\n  In _Proceedings of the 7th Workshop of Statistical Machine\n  Translation_, pages 243–252.\n  Association for Computational Linguistics.\n\n* Anand Karthik Tumuluru, Chi-kiu Lo, and Dekai Wu.\n  2012.\n  Accuracy and robustness in measuring the lexical similarity of\n  semantic role fillers for automatic semantic MT evaluation.\n  In _26th Pacific Asia Conference on Language, Information and\n  Computation_, pages 574–581.\n\n[W12-3129]: http://anthology.aclweb.org/W/W12/W12-3129.pdf\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fquery%2Fpymeant","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fquery%2Fpymeant","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fquery%2Fpymeant/lists"}