llama-index-packs-amazon-product-extraction 0.2.0

llama-index packs amazon_product_extraction integration

Amazon Product Extraction Pack

This LlamaPack provides an example of our Amazon product extraction pack.

It loads in a website URL, screenshots the page. Then we use OpenAI GPT-4V + prompt engineering to extract the screenshot into a structured JSON output.

Check out the notebook here.

CLI Usage

You can download llamapacks directly using llamaindex-cli, which comes installed with the llama-index python package:

llamaindex-cli download-llamapack AmazonProductExtractionPack --download-dir ./amazon_product_extraction_pack

You can then inspect the files at ./amazon_product_extraction_pack and use them as a template for your own project.

Code Usage

You can download the pack to a the ./amazon_product_extraction_pack directory:

from llama_index.core.llama_pack import download_llama_pack

# download and install dependencies
AmazonProductExtractionPack = download_llama_pack(
    "AmazonProductExtractionPack", "./amazon_product_extraction_pack"
)

From here, you can use the pack, or inspect and modify the pack in ./amazon_product_extraction_pack.

Then, you can set up the pack like so:

# create the pack
# get documents from any data loader
amazon_product_extraction_pack = SentenceWindowRetrieverPack(
    amazon_product_page,
)

The run() function is a light wrapper around program().

response = amazon_product_extraction_pack.run()
display(response.dict())

You can also use modules individually.

# get pydantic program
program = amazon_product_extraction_pack.openai_program

# get multi-modal LLM
mm_llm = amazon_product_extraction_pack.openai_mm_llm