Instructions to use LangAGI-Lab/DOCTOR with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use LangAGI-Lab/DOCTOR with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="LangAGI-Lab/DOCTOR")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("LangAGI-Lab/DOCTOR") model = AutoModelForCausalLM.from_pretrained("LangAGI-Lab/DOCTOR") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use LangAGI-Lab/DOCTOR with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "LangAGI-Lab/DOCTOR" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "LangAGI-Lab/DOCTOR", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/LangAGI-Lab/DOCTOR
- SGLang
How to use LangAGI-Lab/DOCTOR with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "LangAGI-Lab/DOCTOR" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "LangAGI-Lab/DOCTOR", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "LangAGI-Lab/DOCTOR" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "LangAGI-Lab/DOCTOR", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use LangAGI-Lab/DOCTOR with Docker Model Runner:
docker model run hf.co/LangAGI-Lab/DOCTOR
metadata
license: apache-2.0
datasets:
- DLI-Lab/DONUT
widget:
- text: >-
A: Hi, Viggo. How are you doing today?\nB: Hey, Yovani. I’m doing all
right. Thanks for asking.\nA: No problem. I saw that you left your coffee
mug on the counter this morning. Did you forget to take it with you?\nB:
Yeah, I did. Thanks for grabbing it for me.\nA: No problem at all. I know
how busy you are and I didn’t want you to have to come back for it
later.\nB: You’re a lifesaver, Yovani. Seriously, thank you so much.
- example_title: example 1
A dialogue commonsense reasoner that generates Chain-of-Thought knowledge in a multi-hop manner given a dialogue history. Our DOCTOR is trained with DONUT which is also available on huggingface.
Links for Reference
- Demo:https://dialoguecot.web.app/
- Repository:https://github.com/kyle8581/DialogueCoT
- Paper:https://arxiv.org/abs/2310.09343
- Point of Contact:mapoout@yonsei.ac.kr
For more details, you can look at our paper Dialogue Chain-of-Thought Distillation for Commonsense-aware Conversational Agents.
If you find the following model helpful, please consider citing our paper!
BibTeX:
@misc{chae2023dialogue,
title={Dialogue Chain-of-Thought Distillation for Commonsense-aware Conversational Agents},
author={Hyungjoo Chae and Yongho Song and Kai Tzu-iunn Ong and Taeyoon Kwon and Minjin Kim and Youngjae Yu and Dongha Lee and Dongyeop Kang and Jinyoung Yeo},
year={2023},
eprint={2310.09343},
archivePrefix={arXiv},
primaryClass={cs.CL}
}