Mbox Reader

If you’re opening this Notebook on colab, you will probably need to install LlamaIndex 🦙.

!pip install llama-index

%env OPENAI_API_KEY=sk-************

from llama_index import MboxReader, VectorStoreIndex

documents = MboxReader().load_data(
    "mbox_data_dir", max_count=1000
)  # Returns list of documents

index = VectorStoreIndex.from_documents(
    documents
)  # Initialize index with documents

query_engine = index.as_query_engine()
res = query_engine.query("When did i have that call with the London office?")

> [query] Total LLM token usage: 100 tokens
> [query] Total embedding token usage: 10 tokens

res.response

> There is a call scheduled with the London office at 12am GMT on the 10th of February.