{"id":20779434,"url":"https://github.com/freedomintelligence/dotagpt","last_synced_at":"2026-03-14T01:02:02.478Z","repository":{"id":199844085,"uuid":"667689290","full_name":"FreedomIntelligence/DotaGPT","owner":"FreedomIntelligence","description":"Chinese Medical instruction-tuning Dataset","archived":false,"fork":false,"pushed_at":"2024-08-07T16:53:47.000Z","size":18,"stargazers_count":5,"open_issues_count":0,"forks_count":0,"subscribers_count":7,"default_branch":"main","last_synced_at":"2025-06-14T14:04:04.366Z","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":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/FreedomIntelligence.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,"zenodo":null}},"created_at":"2023-07-18T05:03:04.000Z","updated_at":"2024-08-21T06:41:08.000Z","dependencies_parsed_at":null,"dependency_job_id":"f58cd7c9-cef3-4456-a00c-081f35fa3523","html_url":"https://github.com/FreedomIntelligence/DotaGPT","commit_stats":null,"previous_names":["freedomintelligence/dotagpt"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/FreedomIntelligence/DotaGPT","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/FreedomIntelligence%2FDotaGPT","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/FreedomIntelligence%2FDotaGPT/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/FreedomIntelligence%2FDotaGPT/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/FreedomIntelligence%2FDotaGPT/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/FreedomIntelligence","download_url":"https://codeload.github.com/FreedomIntelligence/DotaGPT/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/FreedomIntelligence%2FDotaGPT/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":259827506,"owners_count":22917709,"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":"2024-11-17T13:27:55.069Z","updated_at":"2026-03-14T01:02:02.446Z","avatar_url":"https://github.com/FreedomIntelligence.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"## Get Started\n\n```bash\ngit clone https://github.com/FreedomIntelligence/DotaGPT.git\npip install -r requirements.txt\n```\n\n## Evaluation\n\n### Step 1: Download and Prepare Data\n\nDownload the datasets from Hugging Face:\n- [FreedomIntelligence/DoctorFLAN](https://huggingface.co/datasets/FreedomIntelligence/DoctorFLAN)\n- [FreedomIntelligence/DotaBench](https://huggingface.co/datasets/FreedomIntelligence/DotaBench)\n\n### Step 2: Generate and Position the Data\n\n**Data Format**\n\nFor `DotaBench`, the data is structured as follows. Each entry is a JSON object representing a series of interaction turns with a reference answer:\n\n```json\n{\n  \"id\": 0,\n  \"turn_1_question\": \"example question 1\",\n  \"turn_1_answer\": \"[model-generated answer for turn 1]\",\n  \"turn_2_question\": \"example question 2\",\n  \"turn_2_answer\": \"[model-generated answer for turn 2]\",\n  \"turn_3_question\": \"example question 3\",\n  \"turn_3_answer\": \"[model-generated answer for turn 3]\",\n  \"reference\": \"example reference\"\n}\n```\nComplete the fields: `turn_1_answer`, `turn_2_answer`, `turn_3_answer`.\n\nFor `DoctorFLAN`, the data format is as follows, with each entry representing a single-turn interaction:\n\n```json\n{\n  \"id\": 0,\n  \"input\": \"example input\",\n  \"output\": \"[model-generated output]\",\n  \"reference\": \"example reference answer\"\n}\n```\nComplete the field: `output`.\n\nStore the generated model responses in the location: `data/{eval_set}/{model_name}.jsonl`. Ensure that all required fields are correctly filled.\n\n### Step 3: Configuration\n\nPrepare a YAML configuration file specifying model details, API keys, etc. Example (`configs/eval.yaml`):\n\n```yaml\napi_key: \"your-openai-api-key\"\nbase_url: \"https://api.openai.com\"\ngpt_version: \"gpt-4\"\n```\n\n### Step 4: Run the Evaluation\n\nExecute the evaluator with the script `script/run.sh`, modifying parameters as necessary. Example command:\n\n```bash\npython eval_code/reviewer.py \\\n    --config configs/eval.yaml \\\n    --model_name Baichuan-13B-Chat \\\n    --eval_set DotaBench \\\n    --turn_type multi \\\n    --n_processes 2 \\\n    --n_repeat 2 \\\n    --turn_num 2\n```\n\n**Parameter Explanation**\n\n- `--config`: Path to the configuration file.\n- `--model_name`: Name of the model being evaluated.\n- `--eval_set`: Evaluation dataset being used. Choose either `DoctorFLAN` or `DotaBench`.\n- `--turn_type`: Type of interaction (single or multi-turn).\n- `--n_processes`: Number of processes for parallel processing.\n- `--n_repeat`: Number of repetitions for each sample.\n- `--turn_num`: Number of turns for multi-turn evaluations.\n\n## Contributing\n\nContributions are welcome! Feel free to submit issues or pull requests on GitHub to help improve this project.\n\n## Citation\nThe code in this repository is mostly developed for or derived from the paper below. \n```text\n@article{xie2024llms,\n  title={LLMs for Doctors: Leveraging Medical LLMs to Assist Doctors, Not Replace Them},\n  author={Xie, Wenya and Xiao, Qingying and Zheng, Yu and Wang, Xidong and Chen, Junying and Ji, Ke and Gao, Anningzhe and Wan, Xiang and Jiang, Feng and Wang, Benyou},\n  journal={arXiv preprint arXiv:2406.18034},\n  year={2024}\n}\n```\n## License\n\nThis project is licensed under the MIT License.\n\n## Contact Us\n\nFor inquiries, please create an issue in this repository or email the authors: wenyaxie023@gmail.com\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffreedomintelligence%2Fdotagpt","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ffreedomintelligence%2Fdotagpt","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffreedomintelligence%2Fdotagpt/lists"}