import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

# Load base model
base_model_name = "NousResearch/Llama-2-7b-chat-hf"
model = AutoModelForCausalLM.from_pretrained(
    base_model_name,
    torch_dtype=torch.float16,
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(base_model_name)

# Load LoRA adapter
model = PeftModel.from_pretrained(model, "Damon07/llama-2-7b-chat-medical")

# Generate response
prompt = "### Human: I have chest pain. What should I do?\n### Assistant:"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=150, temperature=0.7)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)

Training Details

Base Model: NousResearch/Llama-2-7b-chat-hf
Fine-tuning Method: LoRA (Low-Rank Adaptation)
Dataset: lavita/ChatDoctor-HealthCareMagic-100k (5000 samples)
Task: Medical Question Answering / Healthcare Chat

LoRA Rank: 64
LoRA Alpha: 16
Training Epochs: 1
Learning Rate: 2e-4
Batch Size: 1 (with gradient accumulation steps: 4)

Disclaimer

This model is for educational and research purposes only. Always consult with qualified healthcare professionals for medical advice.