README.md

metadata

language:
  - en
license: mit
library_name: transformers
tags:
  - medical
datasets:
  - Mohammed-Altaf/medical-instruction-120k
model-index:
  - name: Medical-ChatBot
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: AI2 Reasoning Challenge (25-Shot)
          type: ai2_arc
          config: ARC-Challenge
          split: test
          args:
            num_few_shot: 25
        metrics:
          - type: acc_norm
            value: 30.55
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Mohammed-Altaf/Medical-ChatBot
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: HellaSwag (10-Shot)
          type: hellaswag
          split: validation
          args:
            num_few_shot: 10
        metrics:
          - type: acc_norm
            value: 38.63
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Mohammed-Altaf/Medical-ChatBot
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU (5-Shot)
          type: cais/mmlu
          config: all
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 25.98
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Mohammed-Altaf/Medical-ChatBot
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: TruthfulQA (0-shot)
          type: truthful_qa
          config: multiple_choice
          split: validation
          args:
            num_few_shot: 0
        metrics:
          - type: mc2
            value: 41.25
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Mohammed-Altaf/Medical-ChatBot
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: Winogrande (5-shot)
          type: winogrande
          config: winogrande_xl
          split: validation
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 55.41
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Mohammed-Altaf/Medical-ChatBot
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GSM8k (5-shot)
          type: gsm8k
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 0.99
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Mohammed-Altaf/Medical-ChatBot
          name: Open LLM Leaderboard

Please note that the chatbot is designed for research purposes only and is not intended for use in real medical settings. While the chatbot has been trained to provide accurate and helpful responses, it is not a substitute for professional medical advice, diagnosis, or treatment. The information provided by the chatbot should not be used to make medical decisions, and any health concerns should be addressed by a licensed healthcare provider.

Quickstart

import torch
from transformers import GPT2LMHeadModel, GPT2Tokenizer


path = "Mohammed-Altaf/Medical-ChatBot"
device = "cuda" if torch.cuda.is_available() else "cpu"
tokenizer = GPT2Tokenizer.from_pretrained(path)
model = GPT2LMHeadModel.from_pretrained(path).to(device)

prompt_input = (
    "The conversation between human and AI assistant.\n"
    "[|Human|] {input}\n"
    "[|AI|]"
)
sentence = prompt_input.format_map({'input': "what is parkinson's disease?"})
inputs = tokenizer(sentence, return_tensors="pt").to(device)

with torch.no_grad():
    beam_output = model.generate(**inputs,
                                min_new_tokens=1, 
                                max_length=512,
                                num_beams=3,
                                repetition_penalty=1.2,
                                early_stopping=True,
                                eos_token_id=198 
                                )
    print(tokenizer.decode(beam_output[0], skip_special_tokens=True))

Example Outputs

The conversation between human and AI assistant.
[|Human|] what is parkinson's disease?
[|AI|] Parkinson's disease is a neurodegenerative disorder that affects movement. It is caused by the loss of dopamine-producing cells in the brain.

The conversation between human and AI assistant.
[|Human|] what type of honey is best for a bad covid cough?
[|AI|] Manuka honey has been shown to have anti-inflammatory and antibacterial properties that can help alleviate symptoms of a bad covid cough.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	32.13
AI2 Reasoning Challenge (25-Shot)	30.55
HellaSwag (10-Shot)	38.63
MMLU (5-Shot)	25.98
TruthfulQA (0-shot)	41.25
Winogrande (5-shot)	55.41
GSM8k (5-shot)	0.99