LLaMA-Ecommerce

This is a merged model combining NousResearch/Llama-2-7b-hf with DSMI/LLaMA-E LoRA adapter.

Model Description

LLaMA-E is specialized for e-commerce content generation tasks including:

• Product descriptions
• Advertisements
• Product titles
• E-commerce Q&A
• Purchase intent analysis

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("askcatalystai/llama-ecommerce")
tokenizer = AutoTokenizer.from_pretrained("askcatalystai/llama-ecommerce")

prompt = "***Instruction: Write a product description\n***Input: Blue cotton t-shirt, comfortable fit\n***Response:"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=200)
print(tokenizer.decode(outputs[0]))

Original Work

Based on research: LLaMA-E: Empowering E-commerce Authoring with Multi-Aspect Instruction Following

License

This model is subject to the Llama 2 Community License.