Yad4o/skincare-assistant-optimized-lora

Skincare Assistant LoRA (Optimized)

This is an optimized LoRA adapter fine-tuned to provide informational guidance and support for skincare and dermatological queries.

⚠️ MEDICAL DISCLAIMER

This model is for informational purposes only. It is NOT a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of your physician or other qualified health provider with any questions you may have regarding a medical condition.

Model Details

• Base Model: unsloth/Qwen3.5-0.8B
• Architecture: Causal LM
• PEFT Method: LoRA (Low-Rank Adaptation)
• Rank ($r$): 32
• Alpha ($\alpha$): 32
• Target Modules: attention and MLP projections (q, k, v, o, gate, up, down) in the language layers.
• Dropout: 0

Training

• Dataset: lora_health_dataset (including multiturn and expert skincare categories).
• Epochs: 3
• Learning Rate: 2e-4
• Batch Size: 8 (per device) * 2 (accumulation) = 16
• Max Sequence Length: 1024
• Framework: Unsloth / TRL
• Hardware: Google Colab GPU

Comparison

This is the optimized successor to Yad4o/skincare-assistant-lora. Key changes include:

• Base Model: Upgraded from Qwen 1.7B to Qwen 3.5 0.8B.
• Rank: Increased from 16 to 32.
• Dataset: Expanded to include multiturn and expert categories.

Limitations

Informational only. Not medical advice. May produce inaccuracies. Not tested on all skin types.

Installation and Inference

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

base = "unsloth/Qwen3.5-0.8B"
tok = AutoTokenizer.from_pretrained(base)
model = AutoModelForCausalLM.from_pretrained(base, device_map="auto", torch_dtype=torch.float16)
model = PeftModel.from_pretrained(model, "Yad4o/skincare-assistant-optimized-lora")

msgs = [{"role": "user", "content": "What are the benefits of Niacinamide for oily skin?"}]
text = tok.apply_chat_template(msgs, tokenize=False, add_generation_prompt=True)
inputs = tok(text, return_tensors="pt").to(model.device)
out = model.generate(**inputs, max_new_tokens=300)
print(tok.decode(out[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))