def ask_doctor(question): messages = [{"role": "user", "content": question}] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device)

with torch.no_grad():
    outputs = model.generate(
        **inputs,
        max_new_tokens=200,
        pad_token_id=tokenizer.eos_token_id,
        temperature=0.7,
        do_sample=True,
    )
return tokenizer.decode(
    outputs[0][inputs["input_ids"].shape[-1]:],
    skip_special_tokens=True
)

Example

print(ask_doctor("I have a headache and fever from 2 days. What should I do?"))

Example Output

Patient: I have a headache and fever from 2 days. What should I do?

Model: Based on your symptoms, you may be experiencing a viral infection. I recommend rest, staying hydrated, and taking paracetamol for fever. If fever exceeds 103°F or symptoms worsen, please consult a doctor immediately.

Training Details

Parameter	Value
Base Model	SmolLM2-1.7B-Instruct
Method	LoRA
Dataset	avaliev/chat_doctor
Steps	1494
Start Loss	1.5952
Final Loss	1.1899
Framework	HuggingFace Transformers + PEFT

Limitations

Small model (1.7B) — answers may be generic sometimes
Not a replacement for real medical advice
Best used for educational and research purposes

Author

Gaurav — Fine-tuned as part of ML learning project

[More Information Needed]

More Information [optional]

[More Information Needed]

Model Card Authors [optional]

[More Information Needed]

Model Card Contact

[More Information Needed]

Framework versions

PEFT 0.19.1