Model Card for Medical Chatbot LLaMA3 8b 4-bit Fine-tuned Model

This is a fine-tuned version of the LLaMA3 model designed to assist with medical queries and preliminary health advice through a chatbot. It uses a 4-bit quantization to reduce memory usage while maintaining efficiency for chatbot interactions.

Model Details

Model Description

This model is a LLaMA3-8b based chatbot fine-tuned specifically for medical symptom analysis and preliminary assessments. It uses 4-bit quantization, which allows for reduced memory usage, making it suitable for deployment on resource-constrained environments. The chatbot can respond to medical inquiries and provide initial health recommendations, though it should not be used as a substitute for professional medical advice. This model is trained on 1M datapoints dataset consisting of Question answering related to Medical field. It would be helpful for both general person who wants info related to medical issues and also for healthcare providers.

Developed by: Himanshu Kumar
Finetuned from model: LLaMA3 8b
Language(s) (NLP): English
License: apache-2.0
Model type: Causal Language Model (LLM), Fine-tuned with LoRA and 4-bit quantization

Uses

Direct Use

The model is primarily intended for use in medical chatbots to handle preliminary health-related queries and symptom analysis. It is designed for interactive applications that aim to provide users with health-related information and advice.

Downstream Use

This model can be further fine-tuned or adapted to other medical domains or integrated into larger healthcare-related systems. It can also be used for other conversational AI tasks in the medical domain, such as appointment scheduling, follow-up care reminders, or patient triage systems.

Out-of-Scope Use

This model should not be used for critical medical decision-making or as a substitute for professional medical advice.

Bias, Risks, and Limitations

The model may inherit biases present in the training data, and the responses should be carefully monitored, particularly in sensitive contexts like healthcare. It is recommended that the model's responses be reviewed by healthcare professionals.

Recommendations

Users should exercise caution and verify any medical information provided by the model with qualified professionals. The model's limitations in handling complex medical cases must be understood before deployment in real-world scenarios.

How to Get Started with the Model

Use the following code to get started with the model:

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("abhiyanta/chatDoctor")
tokenizer = AutoTokenizer.from_pretrained("abhiyanta/chatDoctor")

inputs = tokenizer(
[
    alpaca_prompt.format(
        "I have fever, what should i do?", # instruction
        "", # input
        "", # output - leave this blank for generation!
    )
], return_tensors = "pt").to("cuda")

from transformers import TextStreamer
text_streamer = TextStreamer(tokenizer)
_ = model.generate(**inputs, streamer = text_streamer, max_new_tokens = 128)

abhiyanta
/

chatDoctor