{"id":22808789,"url":"https://github.com/dnlbauer/bh24de_ontogpt","last_synced_at":"2026-01-11T13:00:14.312Z","repository":{"id":267324238,"uuid":"900881307","full_name":"dnlbauer/bh24de_ontogpt","owner":"dnlbauer","description":null,"archived":false,"fork":false,"pushed_at":"2024-12-12T13:51:08.000Z","size":21,"stargazers_count":2,"open_issues_count":0,"forks_count":1,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-12-07T23:59:28.338Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"cc0-1.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/dnlbauer.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,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-12-09T16:35:47.000Z","updated_at":"2025-03-18T09:54:03.000Z","dependencies_parsed_at":"2024-12-09T18:20:14.619Z","dependency_job_id":"0ba187c0-208a-4721-8864-5397041af8ff","html_url":"https://github.com/dnlbauer/bh24de_ontogpt","commit_stats":null,"previous_names":["dnlbauer/bh24de_ontogpt"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/dnlbauer/bh24de_ontogpt","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dnlbauer%2Fbh24de_ontogpt","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dnlbauer%2Fbh24de_ontogpt/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dnlbauer%2Fbh24de_ontogpt/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dnlbauer%2Fbh24de_ontogpt/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/dnlbauer","download_url":"https://codeload.github.com/dnlbauer/bh24de_ontogpt/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dnlbauer%2Fbh24de_ontogpt/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28304263,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-11T11:18:18.743Z","status":"ssl_error","status_checked_at":"2026-01-11T11:07:56.842Z","response_time":60,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"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":"2024-12-12T11:12:00.666Z","updated_at":"2026-01-11T13:00:14.079Z","avatar_url":"https://github.com/dnlbauer.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Biohackathon Germany 2024\n\nInvestigated the application of [OntoGPT](https://github.com/monarch-initiative/ontogpt) for extracting terms and information from description of [Senckenbergs collection](https://search.senckenberg.de/).\n\n## Templates\n\n### Simple Schema: example to extract genus and species from text\n\n```bash\nontogpt extract -i examples/lathyrus.txt -t templates/simple_schema.yaml -m ollama/llama3\n\ninput_text: |\n  Lathyrus is a pea and a pea is a plant. Lathyrus vestitus is a Lathyrus.\nraw_completion_output: |-\n  Here are the extracted entities in the desired format:\n\n  genus: Lathyrus\n  species: vestitus\nprompt: |+\n  From the text below, extract the following entities in the following format:\n\n  genus: \u003cThe genus of an organism given by its common name or latin name\u003e\n  species: \u003cA species name corresponding to the genus.\u003e\n\n\n  Text:\n  Lathyrus is a pea and a pea is a plant. Lathyrus vestitus is a Lathyrus.\n\n\n  ===\n\nextracted_object:\n  genus: NCBITaxon:3853\n  species: AUTO:vestitus\nnamed_entities:\n  - id: NCBITaxon:3853\n    label: Lathyrus\n    original_spans:\n      - 0:7\n      - 40:47\n      - 63:70\n  - id: AUTO:vestitus\n    label: vestitus\n    original_spans:\n      - 49:56\n\n```\n\n### Habitat schema\n\nTries to extract genus, species and environmental terms independent of the input language.\nThere are 3 \"versions\":\n\n- [templates/habitat.yaml] was an initial attempt\n- [templates/habitat_v2.yaml] aimes to improve the extraction accuracy by giving more examples\nand more details about how to extract ENVO terms.\n- [templates/habitat_noprompt.yaml] has no custom prompts as a base line.\n\n## results\n\nAll templates were applied to 3 examples \"habitat\", \"frischwiese\" and \"pine forest\". See folder [results](results).\nAll results were calculated using ollama/mistral:latest (f974a74358d6):\n\n```bash\nontogpt extract -i \u003cinput\u003e -t \u003ctemplate\u003e --show-prompt -m ollama/mistral -o \u003coutput\u003e\n```\n\n### Lessons learned\n\n- Prompt engineering is hard: tiny changes to the prompt or even the order of extracted attributes can change the output.\n- The model choice is crucial and the template needs to be tailored to the model. Generally mistral outperformed llama **for me**.\n- The output is language specific. It works better when input is translated to english first.\n- Most LLMs are trained on common terms. It would be interesting to see if a model trained with more scientific datasets of a related topic would perform better.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdnlbauer%2Fbh24de_ontogpt","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fdnlbauer%2Fbh24de_ontogpt","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdnlbauer%2Fbh24de_ontogpt/lists"}