---
title: Multi-strategy workflow with reflection
 | Developer Documentation
---

In this notebook we’ll demonstrate a workflow that attempts 3 different query strategies in parallel and picks the best one.

As shown in the diagram below:

- First the quality of the query is judged. If it’s a bad query, a `BadQueryEvent` is emitted and the `improve_query` step will try to improve the quality of the query before trying again. This is reflection.
- Once an acceptable query has been found, three simultaneous events are emitted: a `NaiveRAGEvent`, a `HighTopKEvent`, and a `RerankEvent`.
- Each of these events is picked up by a dedicated step that tries a different RAG strategy on the same index. All 3 emit a `ResponseEvent`
- The `judge` step waits until it has collected all three `ResponseEvents`, then it compares them. It finally emits the best response as a `StopEvent`

![Screenshot 2024-08-16 at 12.41.23 PM.png](data:image/png;base64,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)

## Install dependencies

We need LlamaIndex, the file reader (for reading PDFs), the workflow visualizer (to draw the diagram above), and OpenAI to embed the data and query an LLM.

```
!pip install llama-index-core llama-index-llms-openai llama-index-utils-workflow llama-index-readers-file llama-index-embeddings-openai
```

## Get the data

We are using 3 long PDFs of San Francisco’s annual budget from 2016 through 2018.

```
!mkdir data
!wget "https://www.dropbox.com/scl/fi/xt3squt47djba0j7emmjb/2016-CSF_Budget_Book_2016_FINAL_WEB_with-cover-page.pdf?rlkey=xs064cjs8cb4wma6t5pw2u2bl&dl=0" -O "data/2016-CSF_Budget_Book_2016_FINAL_WEB_with-cover-page.pdf"
!wget "https://www.dropbox.com/scl/fi/jvw59g5nscu1m7f96tjre/2017-Proposed-Budget-FY2017-18-FY2018-19_1.pdf?rlkey=v988oigs2whtcy87ti9wti6od&dl=0" -O "data/2017-Proposed-Budget-FY2017-18-FY2018-19_1.pdf"
!wget "https://www.dropbox.com/scl/fi/izknlwmbs7ia0lbn7zzyx/2018-o0181-18.pdf?rlkey=p5nv2ehtp7272ege3m9diqhei&dl=0" -O "data/2018-o0181-18.pdf"
```

## Bring in dependencies

Now we import all our dependencies

```
import os
from llama_index.core import (
    SimpleDirectoryReader,
    VectorStoreIndex,
    StorageContext,
    load_index_from_storage,
)
from llama_index.core.workflow import (
    step,
    Context,
    Workflow,
    Event,
    StartEvent,
    StopEvent,
)
from llama_index.llms.openai import OpenAI
from llama_index.core.postprocessor.rankGPT_rerank import RankGPTRerank
from llama_index.core.query_engine import RetrieverQueryEngine
from llama_index.core.chat_engine import SimpleChatEngine
from llama_index.utils.workflow import draw_all_possible_flows
```

We also need to set up our OpenAI key.

```
from google.colab import userdata


os.environ["OPENAI_API_KEY"] = userdata.get("openai-key")
```

## Define event classes

Our flow generates quite a few different event types.

```
class JudgeEvent(Event):
    query: str




class BadQueryEvent(Event):
    query: str




class NaiveRAGEvent(Event):
    query: str




class HighTopKEvent(Event):
    query: str




class RerankEvent(Event):
    query: str




class ResponseEvent(Event):
    query: str
    response: str




class SummarizeEvent(Event):
    query: str
    response: str
```

## Define workflow

This is the substance of our workflow, so let’s break it down:

- `load_or_create_index` is a normal RAG function that reads our PDFs from disk and indexes them if they aren’t already indexed. If indexing already happened it will simply restore the existing index from disk.

- `judge_query` does a few things

  - It initializes the LLM and calls `load_or_create_index` to get set up. It stores these things in the context so they are available later.
  - It judges the quality of the query
  - If the query is bad it emits a `BadQueryEvent`
  - If the query is good it emits a `NaiveRAGEvent`, a `HighTopKEvent` and a `RerankerEvent`

- `improve_query` takes the `BadQueryEvent` and uses an LLM to try and expand and deambiguate the query if possible, then it loops back to `judge_query`

- `naive_rag`, `high_top_k` and `rerank` accept their respective events and attempt 3 different RAG strategies. Each emits a `ResponseEvent` with their result and a `source` parameter that says which strategy was used

- `judge` fires every time a `ResponseEvent` is emitted, but it uses `collect_events` to buffer them until it has received all 3. Then it sends the responses to an LLM and asks it to select the “best” one. It emits the best response as a StopEvent

```
class ComplicatedWorkflow(Workflow):
    def load_or_create_index(self, directory_path, persist_dir):
        # Check if the index already exists
        if os.path.exists(persist_dir):
            print("Loading existing index...")
            # Load the index from disk
            storage_context = StorageContext.from_defaults(
                persist_dir=persist_dir
            )
            index = load_index_from_storage(storage_context)
        else:
            print("Creating new index...")
            # Load documents from the specified directory
            documents = SimpleDirectoryReader(directory_path).load_data()


            # Create a new index from the documents
            index = VectorStoreIndex.from_documents(documents)


            # Persist the index to disk
            index.storage_context.persist(persist_dir=persist_dir)


        return index


    @step
    async def judge_query(
        self, ctx: Context, ev: StartEvent | JudgeEvent
    ) -> BadQueryEvent | NaiveRAGEvent | HighTopKEvent | RerankEvent:
        # initialize
        llm = await ctx.store.get("llm", default=None)
        if llm is None:
            await ctx.store.set("llm", OpenAI(model="gpt-4o", temperature=0.1))
            await ctx.store.set(
                "index", self.load_or_create_index("data", "storage")
            )


            # we use a chat engine so it remembers previous interactions
            await ctx.store.set("judge", SimpleChatEngine.from_defaults())


        response = await ctx.store.get("judge").chat(
            f"""
            Given a user query, determine if this is likely to yield good results from a RAG system as-is. If it's good, return 'good', if it's bad, return 'bad'.
            Good queries use a lot of relevant keywords and are detailed. Bad queries are vague or ambiguous.


            Here is the query: {ev.query}
            """
        )
        if response == "bad":
            # try again
            return BadQueryEvent(query=ev.query)
        else:
            # send query to all 3 strategies
            self.send_event(NaiveRAGEvent(query=ev.query))
            self.send_event(HighTopKEvent(query=ev.query))
            self.send_event(RerankEvent(query=ev.query))


    @step
    async def improve_query(
        self, ctx: Context, ev: BadQueryEvent
    ) -> JudgeEvent:
        response = await ctx.store.get("llm").complete(
            f"""
            This is a query to a RAG system: {ev.query}


            The query is bad because it is too vague. Please provide a more detailed query that includes specific keywords and removes any ambiguity.
        """
        )
        return JudgeEvent(query=str(response))


    @step
    async def naive_rag(
        self, ctx: Context, ev: NaiveRAGEvent
    ) -> ResponseEvent:
        index = await ctx.store.get("index")
        engine = index.as_query_engine(similarity_top_k=5)
        response = engine.query(ev.query)
        print("Naive response:", response)
        return ResponseEvent(
            query=ev.query, source="Naive", response=str(response)
        )


    @step
    async def high_top_k(
        self, ctx: Context, ev: HighTopKEvent
    ) -> ResponseEvent:
        index = await ctx.store.get("index")
        engine = index.as_query_engine(similarity_top_k=20)
        response = engine.query(ev.query)
        print("High top k response:", response)
        return ResponseEvent(
            query=ev.query, source="High top k", response=str(response)
        )


    @step
    async def rerank(self, ctx: Context, ev: RerankEvent) -> ResponseEvent:
        index = await ctx.store.get("index")
        reranker = RankGPTRerank(top_n=5, llm=await ctx.store.get("llm"))
        retriever = index.as_retriever(similarity_top_k=20)
        engine = RetrieverQueryEngine.from_args(
            retriever=retriever,
            node_postprocessors=[reranker],
        )
        response = engine.query(ev.query)
        print("Reranker response:", response)
        return ResponseEvent(
            query=ev.query, source="Reranker", response=str(response)
        )


    @step
    async def judge(self, ctx: Context, ev: ResponseEvent) -> StopEvent:
        ready = ctx.collect_events(ev, [ResponseEvent] * 3)
        if ready is None:
            return None


        response = await ctx.store.get("judge").chat(
            f"""
            A user has provided a query and 3 different strategies have been used
            to try to answer the query. Your job is to decide which strategy best
            answered the query. The query was: {ev.query}


            Response 1 ({ready[0].source}): {ready[0].response}
            Response 2 ({ready[1].source}): {ready[1].response}
            Response 3 ({ready[2].source}): {ready[2].response}


            Please provide the number of the best response (1, 2, or 3).
            Just provide the number, with no other text or preamble.
        """
        )


        best_response = int(str(response))
        print(
            f"Best response was number {best_response}, which was from {ready[best_response-1].source}"
        )
        return StopEvent(result=str(ready[best_response - 1].response))
```

## Draw flow diagram

This is how we get the diagram we showed at the start.

```
draw_all_possible_flows(
    ComplicatedWorkflow, filename="complicated_workflow.html"
)
```

## Run the workflow

Let’s take the workflow for a spin:

- The judge\_query event returned nothing. This is because it used `send_event` instead. So the query was judged “good”.
- All 3 RAG steps run and generate different answers to the query
- The `judge` step runs 3 times. The first 2 times it produces no event, because it has not collected the requisite 3 `ResponseEvent`s.
- On the third time it selects the best response and returns a `StopEvent`

```
c = ComplicatedWorkflow(timeout=120, verbose=True)
result = await c.run(
    # query="How has spending on police changed in San Francisco's budgets from 2016 to 2018?"
    # query="How has spending on healthcare changed in San Francisco?"
    query="How has spending changed?"
)
print(result)
```

```
Running step judge_query
Creating new index...
Step judge_query produced no event
Running step naive_rag
Naive response: Spending has increased over the years due to various factors such as new voter-approved minimum spending requirements, the creation of new voter-approved baselines, and growth in baseline funded requirements. Additionally, there have been notable changes in spending across different service areas and departments, with increases in funding for areas like public protection, transportation, and public works.
Step naive_rag produced event ResponseEvent
Running step rerank
Reranker response: Spending has increased over the years, with notable changes in the allocation of funds to various service areas and departments. The budget reflects adjustments in spending to address evolving needs and priorities, resulting in a rise in overall expenditures across different categories.
Step rerank produced event ResponseEvent
Running step high_top_k
High top k response: Spending has increased over the years, with the total budget showing growth in various areas such as aid assistance/grants, materials & supplies, equipment, debt service, services of other departments, and professional & contractual services. Additionally, there have been new investments in programs like workforce development, economic development, film services, and finance and administration. The budget allocations have been adjusted to accommodate changing needs and priorities, reflecting an overall increase in spending across different departments and programs.
Step high_top_k produced event ResponseEvent
Running step judge
Step judge produced no event
Running step judge
Step judge produced no event
Running step judge
Best response was number 3, which was from High top k
Step judge produced event StopEvent
Spending has increased over the years, with the total budget showing growth in various areas such as aid assistance/grants, materials & supplies, equipment, debt service, services of other departments, and professional & contractual services. Additionally, there have been new investments in programs like workforce development, economic development, film services, and finance and administration. The budget allocations have been adjusted to accommodate changing needs and priorities, reflecting an overall increase in spending across different departments and programs.
```
