rag.py

### We first load a knowledge base on which we want to perform RAG

import datasets
from langchain.docstore.document import Document
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_community.retrievers import BM25Retriever
from huggingface_hub import login
import os

knowledge_base = datasets.load_dataset("SuccessfulCrab/web_content", split="train")

source_docs = [
    Document(page_content=doc["text"])
    for doc in knowledge_base
]

text_splitter = RecursiveCharacterTextSplitter(
    chunk_size=500,
    chunk_overlap=50,
    add_start_index=True,
    strip_whitespace=True,
    separators=["\n\n", "\n", ".", " ", ""],
)
docs_processed = text_splitter.split_documents(source_docs)

### Since we need to add a vectordb as an attribute of the tool, we cannot simply use the simple tool constructor with a @tool decorator.
### Therefore we will follow the advanced setup highlighted in the tools tutorial.

from smolagents import Tool

class RetrieverTool(Tool):
    name = "retriever"
    description = "Uses semantic search to retrieve the parts of transformers documentation that could be most relevant to answer your query."
    inputs = {
        "query": {
            "type": "string",
            "description": "The query to perform. This should be semantically close to your target documents. Use the affirmative form rather than a question.",
        }
    }
    output_type = "string"

    def __init__(self, docs, **kwargs):
        super().__init__(**kwargs)
        self.retriever = BM25Retriever.from_documents(
            docs, k=10
        )

    def forward(self, query: str) -> str:
        assert isinstance(query, str), "Your search query must be a string"

        docs = self.retriever.invoke(
            query,
        )
        return "\nRetrieved documents:\n" + "".join(
            [
                f"\n\n===== Document {str(i)} =====\n" + doc.page_content
                for i, doc in enumerate(docs)
            ]
        )

retriever_tool = RetrieverTool(docs_processed)