langchain/templates/rag-fusion/rag_fusion/chain.py
Bagatur 480626dc99
docs, community[patch], experimental[patch], langchain[patch], cli[pa… (#15412)
…tch]: import models from community

ran
```bash
git grep -l 'from langchain\.chat_models' | xargs -L 1 sed -i '' "s/from\ langchain\.chat_models/from\ langchain_community.chat_models/g"
git grep -l 'from langchain\.llms' | xargs -L 1 sed -i '' "s/from\ langchain\.llms/from\ langchain_community.llms/g"
git grep -l 'from langchain\.embeddings' | xargs -L 1 sed -i '' "s/from\ langchain\.embeddings/from\ langchain_community.embeddings/g"
git checkout master libs/langchain/tests/unit_tests/llms
git checkout master libs/langchain/tests/unit_tests/chat_models
git checkout master libs/langchain/tests/unit_tests/embeddings/test_imports.py
make format
cd libs/langchain; make format
cd ../experimental; make format
cd ../core; make format
```
2024-01-02 15:32:16 -05:00

51 lines
1.4 KiB
Python

from langchain import hub
from langchain.load import dumps, loads
from langchain.vectorstores import Pinecone
from langchain_community.chat_models import ChatOpenAI
from langchain_community.embeddings import OpenAIEmbeddings
from langchain_core.output_parsers import StrOutputParser
from langchain_core.pydantic_v1 import BaseModel
def reciprocal_rank_fusion(results: list[list], k=60):
fused_scores = {}
for docs in results:
# Assumes the docs are returned in sorted order of relevance
for rank, doc in enumerate(docs):
doc_str = dumps(doc)
if doc_str not in fused_scores:
fused_scores[doc_str] = 0
fused_scores[doc_str] += 1 / (rank + k)
reranked_results = [
(loads(doc), score)
for doc, score in sorted(fused_scores.items(), key=lambda x: x[1], reverse=True)
]
return reranked_results
prompt = hub.pull("langchain-ai/rag-fusion-query-generation")
generate_queries = (
prompt | ChatOpenAI(temperature=0) | StrOutputParser() | (lambda x: x.split("\n"))
)
vectorstore = Pinecone.from_existing_index("rag-fusion", OpenAIEmbeddings())
retriever = vectorstore.as_retriever()
chain = (
{"original_query": lambda x: x}
| generate_queries
| retriever.map()
| reciprocal_rank_fusion
)
# Add typed inputs to chain for playground
class Question(BaseModel):
__root__: str
chain = chain.with_types(input_type=Question)