diff --git a/docs/modules/indexes/retrievers/examples/chroma_self_query_retriever.ipynb b/docs/modules/indexes/retrievers/examples/chroma_self_query_retriever.ipynb new file mode 100644 index 00000000..b8f79fb4 --- /dev/null +++ b/docs/modules/indexes/retrievers/examples/chroma_self_query_retriever.ipynb @@ -0,0 +1,302 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "13afcae7", + "metadata": {}, + "source": [ + "# Self-querying retriever with Chroma\n", + "In the notebook we'll demo the `SelfQueryRetriever` wrapped around a Chroma vector store. " + ] + }, + { + "cell_type": "markdown", + "id": "68e75fb9", + "metadata": {}, + "source": [ + "## Creating a Pinecone index\n", + "First we'll want to create a Chroma VectorStore and seed it with some data. We've created a small demo set of documents that contain summaries of movies.\n", + "\n", + "NOTE: The self-query retriever requires you to have `lark` installed (`pip install lark`)" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "63a8af5b", + "metadata": {}, + "outputs": [], + "source": [ + "# !pip install lark" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "cb4a5787", + "metadata": {}, + "outputs": [], + "source": [ + "from langchain.schema import Document\n", + "from langchain.embeddings.openai import OpenAIEmbeddings\n", + "from langchain.vectorstores import Chroma\n", + "\n", + "embeddings = OpenAIEmbeddings()" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "bcbe04d9", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Using embedded DuckDB without persistence: data will be transient\n" + ] + } + ], + "source": [ + "docs = [\n", + " Document(page_content=\"A bunch of scientists bring back dinosaurs and mayhem breaks loose\", metadata={\"year\": 1993, \"rating\": 7.7, \"genre\": \"science fiction\"}),\n", + " Document(page_content=\"Leo DiCaprio gets lost in a dream within a dream within a dream within a ...\", metadata={\"year\": 2010, \"director\": \"Christopher Nolan\", \"rating\": 8.2}),\n", + " Document(page_content=\"A psychologist / detective gets lost in a series of dreams within dreams within dreams and Inception reused the idea\", metadata={\"year\": 2006, \"director\": \"Satoshi Kon\", \"rating\": 8.6}),\n", + " Document(page_content=\"A bunch of normal-sized women are supremely wholesome and some men pine after them\", metadata={\"year\": 2019, \"director\": \"Greta Gerwig\", \"rating\": 8.3}),\n", + " Document(page_content=\"Toys come alive and have a blast doing so\", metadata={\"year\": 1995, \"genre\": \"animated\"}),\n", + " Document(page_content=\"Three men walk into the Zone, three men walk out of the Zone\", metadata={\"year\": 1979, \"rating\": 9.9, \"director\": \"Andrei Tarkovsky\", \"genre\": \"science fiction\", \"rating\": 9.9})\n", + "]\n", + "vectorstore = Chroma.from_documents(\n", + " docs, embeddings\n", + ")" + ] + }, + { + "cell_type": "markdown", + "id": "5ecaab6d", + "metadata": {}, + "source": [ + "# Creating our self-querying retriever\n", + "Now we can instantiate our retriever. To do this we'll need to provide some information upfront about the metadata fields that our documents support and a short description of the document contents." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "86e34dbf", + "metadata": {}, + "outputs": [], + "source": [ + "from langchain.llms import OpenAI\n", + "from langchain.retrievers.self_query.base import SelfQueryRetriever\n", + "from langchain.chains.query_constructor.base import AttributeInfo\n", + "\n", + "metadata_field_info=[\n", + " AttributeInfo(\n", + " name=\"genre\",\n", + " description=\"The genre of the movie\", \n", + " type=\"string or list[string]\", \n", + " ),\n", + " AttributeInfo(\n", + " name=\"year\",\n", + " description=\"The year the movie was released\", \n", + " type=\"integer\", \n", + " ),\n", + " AttributeInfo(\n", + " name=\"director\",\n", + " description=\"The name of the movie director\", \n", + " type=\"string\", \n", + " ),\n", + " AttributeInfo(\n", + " name=\"rating\",\n", + " description=\"A 1-10 rating for the movie\",\n", + " type=\"float\"\n", + " ),\n", + "]\n", + "document_content_description = \"Brief summary of a movie\"\n", + "llm = OpenAI(temperature=0)\n", + "retriever = SelfQueryRetriever.from_llm(llm, vectorstore, document_content_description, metadata_field_info, verbose=True)" + ] + }, + { + "cell_type": "markdown", + "id": "ea9df8d4", + "metadata": {}, + "source": [ + "# Testing it out\n", + "And now we can try actually using our retriever!" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "38a126e9", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "query='dinosaur' filter=None\n" + ] + }, + { + "data": { + "text/plain": [ + "[Document(page_content='A bunch of scientists bring back dinosaurs and mayhem breaks loose', metadata={'year': 1993, 'rating': 7.7, 'genre': 'science fiction'}),\n", + " Document(page_content='Toys come alive and have a blast doing so', metadata={'year': 1995, 'genre': 'animated'}),\n", + " Document(page_content='A psychologist / detective gets lost in a series of dreams within dreams within dreams and Inception reused the idea', metadata={'year': 2006, 'director': 'Satoshi Kon', 'rating': 8.6}),\n", + " Document(page_content='Leo DiCaprio gets lost in a dream within a dream within a dream within a ...', metadata={'year': 2010, 'director': 'Christopher Nolan', 'rating': 8.2})]" + ] + }, + "execution_count": 6, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# This example only specifies a relevant query\n", + "retriever.get_relevant_documents(\"What are some movies about dinosaurs\")" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "fc3f1e6e", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "query=' ' filter=Comparison(comparator=, attribute='rating', value=8.5)\n" + ] + }, + { + "data": { + "text/plain": [ + "[Document(page_content='A psychologist / detective gets lost in a series of dreams within dreams within dreams and Inception reused the idea', metadata={'year': 2006, 'director': 'Satoshi Kon', 'rating': 8.6}),\n", + " Document(page_content='Three men walk into the Zone, three men walk out of the Zone', metadata={'year': 1979, 'rating': 9.9, 'director': 'Andrei Tarkovsky', 'genre': 'science fiction'})]" + ] + }, + "execution_count": 7, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# This example only specifies a filter\n", + "retriever.get_relevant_documents(\"I want to watch a movie rated higher than 8.5\")" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "b19d4da0", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "query='women' filter=Comparison(comparator=, attribute='director', value='Greta Gerwig')\n" + ] + }, + { + "data": { + "text/plain": [ + "[Document(page_content='A bunch of normal-sized women are supremely wholesome and some men pine after them', metadata={'year': 2019, 'director': 'Greta Gerwig', 'rating': 8.3})]" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# This example specifies a query and a filter\n", + "retriever.get_relevant_documents(\"Has Greta Gerwig directed any movies about women\")" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "f900e40e", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "query=' ' filter=Operation(operator=, arguments=[Comparison(comparator=, attribute='genre', value='science fiction'), Comparison(comparator=, attribute='rating', value=8.5)])\n" + ] + }, + { + "data": { + "text/plain": [ + "[Document(page_content='Three men walk into the Zone, three men walk out of the Zone', metadata={'year': 1979, 'rating': 9.9, 'director': 'Andrei Tarkovsky', 'genre': 'science fiction'})]" + ] + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# This example specifies a composite filter\n", + "retriever.get_relevant_documents(\"What's a highly rated (above 8.5) science fiction film?\")" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "12a51522", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "query='toys' filter=Operation(operator=, arguments=[Comparison(comparator=, attribute='year', value=1990), Comparison(comparator=, attribute='year', value=2005), Comparison(comparator=, attribute='genre', value='animated')])\n" + ] + }, + { + "data": { + "text/plain": [ + "[Document(page_content='Toys come alive and have a blast doing so', metadata={'year': 1995, 'genre': 'animated'})]" + ] + }, + "execution_count": 10, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# This example specifies a query and composite filter\n", + "retriever.get_relevant_documents(\"What's a movie after 1990 but before 2005 that's all about toys, and preferably is animated\")" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.3" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/docs/modules/indexes/retrievers/examples/self_query_retriever.ipynb b/docs/modules/indexes/retrievers/examples/self_query_retriever.ipynb index 665adf9f..2725c97d 100644 --- a/docs/modules/indexes/retrievers/examples/self_query_retriever.ipynb +++ b/docs/modules/indexes/retrievers/examples/self_query_retriever.ipynb @@ -17,8 +17,6 @@ "## Creating a Pinecone index\n", "First we'll want to create a Pinecone VectorStore and seed it with some data. We've created a small demo set of documents that contain summaries of movies.\n", "\n", - "NOTE: The self-query retriever currently only has built-in support for Pinecone VectorStore.\n", - "\n", "NOTE: The self-query retriever requires you to have `lark` installed (`pip install lark`)" ] }, @@ -322,7 +320,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.9.1" + "version": "3.11.3" } }, "nbformat": 4, diff --git a/langchain/retrievers/__init__.py b/langchain/retrievers/__init__.py index a56c9f96..6fc0c806 100644 --- a/langchain/retrievers/__init__.py +++ b/langchain/retrievers/__init__.py @@ -6,6 +6,7 @@ from langchain.retrievers.knn import KNNRetriever from langchain.retrievers.metal import MetalRetriever from langchain.retrievers.pinecone_hybrid_search import PineconeHybridSearchRetriever from langchain.retrievers.remote_retriever import RemoteLangChainRetriever +from langchain.retrievers.self_query.base import SelfQueryRetriever from langchain.retrievers.svm import SVMRetriever from langchain.retrievers.tfidf import TFIDFRetriever from langchain.retrievers.time_weighted_retriever import ( @@ -28,4 +29,5 @@ __all__ = [ "SVMRetriever", "KNNRetriever", "VespaRetriever", + "SelfQueryRetriever", ] diff --git a/langchain/retrievers/self_query/base.py b/langchain/retrievers/self_query/base.py index b74dfaca..bf5ad303 100644 --- a/langchain/retrievers/self_query/base.py +++ b/langchain/retrievers/self_query/base.py @@ -8,15 +8,17 @@ from langchain.base_language import BaseLanguageModel from langchain.chains.query_constructor.base import load_query_constructor_chain from langchain.chains.query_constructor.ir import StructuredQuery, Visitor from langchain.chains.query_constructor.schema import AttributeInfo +from langchain.retrievers.self_query.chroma import ChromaTranslator from langchain.retrievers.self_query.pinecone import PineconeTranslator from langchain.schema import BaseRetriever, Document -from langchain.vectorstores import Pinecone, VectorStore +from langchain.vectorstores import Chroma, Pinecone, VectorStore def _get_builtin_translator(vectorstore_cls: Type[VectorStore]) -> Visitor: """Get the translator class corresponding to the vector store class.""" BUILTIN_TRANSLATORS: Dict[Type[VectorStore], Type[Visitor]] = { - Pinecone: PineconeTranslator + Pinecone: PineconeTranslator, + Chroma: ChromaTranslator, } if vectorstore_cls not in BUILTIN_TRANSLATORS: raise ValueError( diff --git a/langchain/retrievers/self_query/chroma.py b/langchain/retrievers/self_query/chroma.py new file mode 100644 index 00000000..02457de3 --- /dev/null +++ b/langchain/retrievers/self_query/chroma.py @@ -0,0 +1,53 @@ +"""Logic for converting internal query language to a valid Chroma query.""" +from typing import Dict, Tuple, Union + +from langchain.chains.query_constructor.ir import ( + Comparator, + Comparison, + Operation, + Operator, + StructuredQuery, + Visitor, +) + + +class ChromaTranslator(Visitor): + """Logic for converting internal query language elements to valid filters.""" + + allowed_operators = [Operator.AND, Operator.OR] + """Subset of allowed logical operators.""" + + def _format_func(self, func: Union[Operator, Comparator]) -> str: + if isinstance(func, Operator) and self.allowed_operators is not None: + if func not in self.allowed_operators: + raise ValueError( + f"Received disallowed operator {func}. Allowed " + f"comparators are {self.allowed_operators}" + ) + if isinstance(func, Comparator) and self.allowed_comparators is not None: + if func not in self.allowed_comparators: + raise ValueError( + f"Received disallowed comparator {func}. Allowed " + f"comparators are {self.allowed_comparators}" + ) + return f"${func.value}" + + def visit_operation(self, operation: Operation) -> Dict: + args = [arg.accept(self) for arg in operation.arguments] + return {self._format_func(operation.operator): args} + + def visit_comparison(self, comparison: Comparison) -> Dict: + return { + comparison.attribute: { + self._format_func(comparison.comparator): comparison.value + } + } + + def visit_structured_query( + self, structured_query: StructuredQuery + ) -> Tuple[str, dict]: + if structured_query.filter is None: + kwargs = {} + else: + kwargs = {"filter": structured_query.filter.accept(self)} + return structured_query.query, kwargs