|
--- |
|
library_name: transformers |
|
tags: |
|
- medical |
|
- text-generation-inference |
|
- llm |
|
- biomistral |
|
license: apache-2.0 |
|
datasets: |
|
- ruslanmv/ai-medical-chatbot |
|
- lavita/ChatDoctor-HealthCareMagic-100k |
|
language: |
|
- en |
|
metrics: |
|
- rouge |
|
- bleu |
|
--- |
|
|
|
# Model Card for BioMistral Multi-Turn Doctor Conversation Model |
|
|
|
## Model Details |
|
|
|
### Model Description |
|
|
|
This model is a fine-tuned version of the BioMistral model, specifically tailored for multi-turn doctor-patient conversations. It leverages the powerful language generation capabilities of BioMistral to provide accurate and context-aware responses in medical dialogue scenarios. |
|
|
|
- **Developed by:** Siyahul Haque T P |
|
- **Model type:** Text-generation (LLM) |
|
- **Language(s) (NLP):** English (en) |
|
- **License:** Apache-2.0 |
|
- **Finetuned from model:** BioMistral |
|
|
|
|
|
|
|
## Uses |
|
|
|
### Direct Use |
|
|
|
This model can be directly used for generating responses in multi-turn medical conversations, making it useful for applications like virtual health assistants and medical chatbots. |
|
|
|
### Downstream Use |
|
|
|
This model can be further fine-tuned or integrated into larger healthcare applications, such as patient management systems or automated symptom checkers. |
|
|
|
### Out-of-Scope Use |
|
|
|
The model is not suitable for use in emergency medical situations, providing final diagnoses, or replacing professional medical advice. |
|
|
|
## Bias, Risks, and Limitations |
|
|
|
The model may reflect biases present in the training data, including underrepresentation of certain medical conditions or demographic groups. The model should not be used as a sole source of medical information and must be supervised by qualified healthcare professionals. |
|
|
|
### Recommendations |
|
|
|
Users should be aware of the potential biases and limitations of the model. It is recommended to use the model as a supplementary tool rather than a primary source of medical advice. |
|
|
|
## How to Get Started with the Model |
|
|
|
```python |
|
from transformers import AutoModelForCausalLM, AutoTokenizer |
|
|
|
# Load the tokenizer and model from the Hugging Face Hub |
|
tokenizer = AutoTokenizer.from_pretrained("BioMistral/BioMistral-7B") |
|
model = AutoModelForCausalLM.from_pretrained("siyah1/BioMistral-7b-Chat-Doctor") |
|
|
|
# Example input: patient describing a symptom |
|
input_text = "Hello, doctor, I have a headache." |
|
|
|
# Tokenize the input text |
|
inputs = tokenizer(input_text, return_tensors="pt") |
|
|
|
# Generate a response from the model |
|
outputs = model.generate(**inputs, max_length=100, num_return_sequences=1) |
|
|
|
# Decode the generated response |
|
response = tokenizer.decode(outputs[0], skip_special_tokens=True) |
|
|
|
# Print the model's response |
|
print("Doctor:", response) |
|
|
