{"id":21291541,"url":"https://github.com/smartdataanalytics/transformers_dialogue_evaluators","last_synced_at":"2025-03-15T16:34:12.129Z","repository":{"id":145429163,"uuid":"308364698","full_name":"SmartDataAnalytics/transformers_dialogue_evaluators","owner":"SmartDataAnalytics","description":"Resources to reproduce the results reported in the paper: \"Language Model Transformers as Evaluators for Open-domain Dialogues\".","archived":false,"fork":false,"pushed_at":"2020-10-29T15:13:09.000Z","size":42498,"stargazers_count":2,"open_issues_count":0,"forks_count":1,"subscribers_count":10,"default_branch":"master","last_synced_at":"2025-01-22T06:22:38.736Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/SmartDataAnalytics.png","metadata":{"files":{"readme":"readme.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":"2020-10-29T15:07:19.000Z","updated_at":"2022-03-07T09:51:42.000Z","dependencies_parsed_at":"2023-09-13T06:17:09.919Z","dependency_job_id":null,"html_url":"https://github.com/SmartDataAnalytics/transformers_dialogue_evaluators","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SmartDataAnalytics%2Ftransformers_dialogue_evaluators","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SmartDataAnalytics%2Ftransformers_dialogue_evaluators/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SmartDataAnalytics%2Ftransformers_dialogue_evaluators/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SmartDataAnalytics%2Ftransformers_dialogue_evaluators/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/SmartDataAnalytics","download_url":"https://codeload.github.com/SmartDataAnalytics/transformers_dialogue_evaluators/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":243760425,"owners_count":20343629,"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","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":"2024-11-21T13:35:06.841Z","updated_at":"2025-03-15T16:34:12.108Z","avatar_url":"https://github.com/SmartDataAnalytics.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"## Language Model Transformers as Evaluators for Open-domain Dialogues\n\nThis repository provides the resources to reproduce the results reported in the paper: \"Language Model Transformers as Evaluators for Open-domain Dialogues\" ([link](http://jens-lehmann.org/files/2020/coling_lm_dialogue_eval.pdf)).\n\nThese are the instructions for reproducing the results. We provide the following scripts and resources:\n\n\n### Details about the contents\n\n- `transformers_dialogue_evaluators.py`\n    - the scripts compute probability scores for the ConvAI1 and ConvAI2 datasets using BERT, XLNet and GPT2\n    - Depending on the available hardware the script can take a day or even longer to execute and compute the results.\n    - just execute the script to obtain the results:\n        - `python -u transformers_dialogue_evaluators.py`\n- `convai(1|2)_results.pickle.bz2` - we provide the already computed probability scores as a shortcut for the correlation analysis\n- `convai(1|2)_corr.ipynb` - Jupyter notebooks that:\n    - calculate the various aggregated scores for dialogues\n    - compute the correlation scores\n    - visualize them in an interactive spreadsheet\n\n\n### Instructions\n\nPython 3.6 is used to run the scripts. We recommend using a virtual environment like (Ana|Mini)conda. Steps:\n\n1. Install dependencies\n    - `pip install jupyter requests numpy scipy scikit-learn seaborn tqdm torch==1.3.1 transformers==2.2.1 pandas qgrid`\n2. Activate qgrid Jupyter extension\n    - `jupyter nbextension enable --py --sys-prefix qgrid`\n    - Skipping this step would prevent Jupyter from rendering an interactive spreadsheet with the correlation scores\n3. Start Jupyter:\n    - `jupyter notebook`\n4. Open and run all the cells in the notebooks\n    - the correlation scores should be computed and visualized\n    - sample dialogues used in the paper are shown\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsmartdataanalytics%2Ftransformers_dialogue_evaluators","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsmartdataanalytics%2Ftransformers_dialogue_evaluators","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsmartdataanalytics%2Ftransformers_dialogue_evaluators/lists"}