TinyLlama-ECommerce-Chatbot

This is a fine-tuned version of TinyLlama/TinyLlama-1.1B-Chat-v1.0 for e-commerce customer service chatbot applications.

Model Details

• Base Model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
• Fine-tuning Method: LoRA (Low-Rank Adaptation)
• Task: E-commerce customer service conversation
• Training Data: ChatML formatted e-commerce conversations

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

# Load base model and tokenizer
base_model = AutoModelForCausalLM.from_pretrained("TinyLlama/TinyLlama-1.1B-Chat-v1.0")
tokenizer = AutoTokenizer.from_pretrained("ShenghaoYummy/TinyLlama-ECommerce-Chatbot")

# Load the fine-tuned model
model = PeftModel.from_pretrained(base_model, "ShenghaoYummy/TinyLlama-ECommerce-Chatbot")

# Generate response
def generate_response(prompt):
    inputs = tokenizer(prompt, return_tensors="pt")
    outputs = model.generate(**inputs, max_new_tokens=150, temperature=0.8)
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

# Example usage
prompt = "<|system|>\nYou are a helpful e-commerce assistant.<|im_end|>\n<|user|>\nWhat's your return policy?<|im_end|>\n<|assistant|>\n"
response = generate_response(prompt)
print(response)

Training Details

This model was fine-tuned using:

• LoRA with rank 16
• Learning rate optimization via Optuna
• MLflow experiment tracking
• DVC pipeline management

Limitations

• Optimized for e-commerce customer service scenarios
• May not perform well on general conversation topics
• Responses are based on training data patterns

Citation

If you use this model, please cite: @misc{ecommerce-chatbot-tinyllama, title={Fine-tuned TinyLlama for E-commerce Customer Service}, author={Your Name}, year={2024}, url={https://huggingface.co/ShenghaoYummy/TinyLlama-ECommerce-Chatbot} }