diff --git a/docs/extras/integrations/vectorstores/vectara.ipynb b/docs/extras/integrations/vectorstores/vectara.ipynb index bf99823394f..0741c1b1998 100644 --- a/docs/extras/integrations/vectorstores/vectara.ipynb +++ b/docs/extras/integrations/vectorstores/vectara.ipynb @@ -1,7 +1,6 @@ { "cells": [ { - "attachments": {}, "cell_type": "markdown", "id": "683953b3", "metadata": {}, @@ -60,7 +59,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "eeead681", "metadata": {}, @@ -73,7 +71,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 1, "id": "04a1f1a0", "metadata": {}, "outputs": [], @@ -86,12 +84,12 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 2, "id": "be0a4973", "metadata": {}, "outputs": [], "source": [ - "loader = TextLoader(\"../../../state_of_the_union.txt\")\n", + "loader = TextLoader(\"../../modules/state_of_the_union.txt\")\n", "documents = loader.load()\n", "text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)\n", "docs = text_splitter.split_documents(documents)" @@ -99,7 +97,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 3, "id": "8429667e", "metadata": { "ExecuteTime": { @@ -118,7 +116,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "90dbf3e7", "metadata": {}, @@ -133,7 +130,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 4, "id": "85ef3468", "metadata": {}, "outputs": [], @@ -165,7 +162,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "1f9215c8", "metadata": { @@ -182,7 +178,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 5, "id": "a8c513ab", "metadata": { "ExecuteTime": { @@ -201,7 +197,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 6, "id": "fc516993", "metadata": { "ExecuteTime": { @@ -215,13 +211,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "Tonight. I call on the Senate to: Pass the Freedom to Vote Act. Pass the John Lewis Voting Rights Act. And while you’re at it, pass the Disclose Act so Americans can know who is funding our elections. \n", - "\n", - "Tonight, I’d like to honor someone who has dedicated his life to serve this country: Justice Stephen Breyer—an Army veteran, Constitutional scholar, and retiring Justice of the United States Supreme Court. Justice Breyer, thank you for your service. \n", - "\n", - "One of the most serious constitutional responsibilities a President has is nominating someone to serve on the United States Supreme Court. \n", - "\n", - "And I did that 4 days ago, when I nominated Circuit Court of Appeals Judge Ketanji Brown Jackson. One of our nation’s top legal minds, who will continue Justice Breyer’s legacy of excellence.\n" + "And I did that 4 days ago, when I nominated Circuit Court of Appeals Judge Ketanji Brown Jackson.\n" ] } ], @@ -230,7 +220,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "1bda9bf5", "metadata": {}, @@ -242,7 +231,7 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 7, "id": "8804a21d", "metadata": { "ExecuteTime": { @@ -254,13 +243,13 @@ "source": [ "query = \"What did the president say about Ketanji Brown Jackson\"\n", "found_docs = vectara.similarity_search_with_score(\n", - " query, filter=\"doc.speech = 'state-of-the-union'\"\n", + " query, filter=\"doc.speech = 'state-of-the-union'\", score_threshold=0.2,\n", ")" ] }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 8, "id": "756a6887", "metadata": { "ExecuteTime": { @@ -273,15 +262,9 @@ "name": "stdout", "output_type": "stream", "text": [ - "Tonight. I call on the Senate to: Pass the Freedom to Vote Act. Pass the John Lewis Voting Rights Act. And while you’re at it, pass the Disclose Act so Americans can know who is funding our elections. \n", + "Justice Breyer, thank you for your service. One of the most serious constitutional responsibilities a President has is nominating someone to serve on the United States Supreme Court. And I did that 4 days ago, when I nominated Circuit Court of Appeals Judge Ketanji Brown Jackson. One of our nation’s top legal minds, who will continue Justice Breyer’s legacy of excellence. A former top litigator in private practice.\n", "\n", - "Tonight, I’d like to honor someone who has dedicated his life to serve this country: Justice Stephen Breyer—an Army veteran, Constitutional scholar, and retiring Justice of the United States Supreme Court. Justice Breyer, thank you for your service. \n", - "\n", - "One of the most serious constitutional responsibilities a President has is nominating someone to serve on the United States Supreme Court. \n", - "\n", - "And I did that 4 days ago, when I nominated Circuit Court of Appeals Judge Ketanji Brown Jackson. One of our nation’s top legal minds, who will continue Justice Breyer’s legacy of excellence.\n", - "\n", - "Score: 0.4917977\n" + "Score: 0.786569\n" ] } ], @@ -292,7 +275,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "1f9876a8", "metadata": {}, @@ -302,7 +284,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 9, "id": "47784de5", "metadata": {}, "outputs": [ @@ -310,22 +292,43 @@ "name": "stdout", "output_type": "stream", "text": [ - "(Document(page_content='We must forever conduct our struggle on the high plane of dignity and discipline.', metadata={'section': '1'}), 0.7962591)\n", - "(Document(page_content='We must not allow our\\ncreative protests to degenerate into physical violence. . . .', metadata={'section': '1'}), 0.25983918)\n" + "With this threshold of 1.2 we have 0 documents\n" ] } ], "source": [ "query = \"We must forever conduct our struggle\"\n", + "min_score = 1.2\n", "found_docs = vectara.similarity_search_with_score(\n", - " query, filter=\"doc.speech = 'I-have-a-dream'\"\n", + " query, filter=\"doc.speech = 'I-have-a-dream'\", score_threshold=min_score,\n", ")\n", - "print(found_docs[0])\n", - "print(found_docs[1])" + "print(f\"With this threshold of {min_score} we have {len(found_docs)} documents\")" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "3e22949f", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "With this threshold of 0.2 we have 3 documents\n" + ] + } + ], + "source": [ + "query = \"We must forever conduct our struggle\"\n", + "min_score = 0.2\n", + "found_docs = vectara.similarity_search_with_score(\n", + " query, filter=\"doc.speech = 'I-have-a-dream'\", score_threshold=min_score,\n", + ")\n", + "print(f\"With this threshold of {min_score} we have {len(found_docs)} documents\")\n" ] }, { - "attachments": {}, "cell_type": "markdown", "id": "691a82d6", "metadata": {}, @@ -349,7 +352,7 @@ { "data": { "text/plain": [ - "VectaraRetriever(vectorstore=, search_type='similarity', search_kwargs={'lambda_val': 0.025, 'k': 5, 'filter': '', 'n_sentence_context': '0'})" + "VectaraRetriever(tags=['Vectara'], metadata=None, vectorstore=, search_type='similarity', search_kwargs={'lambda_val': 0.025, 'k': 5, 'filter': '', 'n_sentence_context': '2'})" ] }, "execution_count": 11, @@ -376,7 +379,7 @@ { "data": { "text/plain": [ - "Document(page_content='Tonight. I call on the Senate to: Pass the Freedom to Vote Act. Pass the John Lewis Voting Rights Act. And while you’re at it, pass the Disclose Act so Americans can know who is funding our elections. \\n\\nTonight, I’d like to honor someone who has dedicated his life to serve this country: Justice Stephen Breyer—an Army veteran, Constitutional scholar, and retiring Justice of the United States Supreme Court. Justice Breyer, thank you for your service. \\n\\nOne of the most serious constitutional responsibilities a President has is nominating someone to serve on the United States Supreme Court. \\n\\nAnd I did that 4 days ago, when I nominated Circuit Court of Appeals Judge Ketanji Brown Jackson. One of our nation’s top legal minds, who will continue Justice Breyer’s legacy of excellence.', metadata={'source': '../../../state_of_the_union.txt'})" + "Document(page_content='Justice Breyer, thank you for your service. One of the most serious constitutional responsibilities a President has is nominating someone to serve on the United States Supreme Court. And I did that 4 days ago, when I nominated Circuit Court of Appeals Judge Ketanji Brown Jackson. One of our nation’s top legal minds, who will continue Justice Breyer’s legacy of excellence. A former top litigator in private practice.', metadata={'source': 'langchain', 'lang': 'eng', 'offset': '596', 'len': '97', 'speech': 'state-of-the-union'})" ] }, "execution_count": 12, @@ -414,7 +417,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.9.1" + "version": "3.10.9" } }, "nbformat": 4, diff --git a/libs/langchain/langchain/vectorstores/vectara.py b/libs/langchain/langchain/vectorstores/vectara.py index eee2f1abe1f..9af1124648b 100644 --- a/libs/langchain/langchain/vectorstores/vectara.py +++ b/libs/langchain/langchain/vectorstores/vectara.py @@ -245,6 +245,7 @@ class Vectara(VectorStore): k: int = 5, lambda_val: float = 0.025, filter: Optional[str] = None, + score_threshold: Optional[float] = None, n_sentence_context: int = 2, **kwargs: Any, ) -> List[Tuple[Document, float]]: @@ -258,6 +259,9 @@ class Vectara(VectorStore): filter can be "doc.rating > 3.0 and part.lang = 'deu'"} see https://docs.vectara.com/docs/search-apis/sql/filter-overview for more details. + score_threshold: minimal score threshold for the result. + If defined, results with score less than this value will be + filtered out. n_sentence_context: number of sentences before/after the matching segment to add, defaults to 2 @@ -305,7 +309,14 @@ class Vectara(VectorStore): result = response.json() - responses = result["responseSet"][0]["response"] + if score_threshold: + responses = [ + r + for r in result["responseSet"][0]["response"] + if r["score"] > score_threshold + ] + else: + responses = result["responseSet"][0]["response"] documents = result["responseSet"][0]["document"] metadatas = [] @@ -316,7 +327,7 @@ class Vectara(VectorStore): md.update(doc_md) metadatas.append(md) - docs = [ + docs_with_score = [ ( Document( page_content=x["text"], @@ -327,7 +338,7 @@ class Vectara(VectorStore): for x, md in zip(responses, metadatas) ] - return docs + return docs_with_score def similarity_search( self, @@ -358,6 +369,7 @@ class Vectara(VectorStore): k=k, lambda_val=lambda_val, filter=filter, + score_threshold=None, n_sentence_context=n_sentence_context, **kwargs, ) diff --git a/libs/langchain/tests/integration_tests/vectorstores/test_vectara.py b/libs/langchain/tests/integration_tests/vectorstores/test_vectara.py index 8fa3cd7f40d..2e79001a1c5 100644 --- a/libs/langchain/tests/integration_tests/vectorstores/test_vectara.py +++ b/libs/langchain/tests/integration_tests/vectorstores/test_vectara.py @@ -68,6 +68,19 @@ def test_vectara_add_documents() -> None: assert output2[0].page_content == "retrieval augmented generation" assert output2[0].metadata["abbr"] == "rag" + # test without filter but with similarity score + # this is similar to the first test, but given the score threshold + # we only get one result + output3 = docsearch.similarity_search_with_score( + "large language model", + k=2, + score_threshold=0.1, + n_sentence_context=0, + ) + assert len(output3) == 1 + assert output3[0][0].page_content == "large language model" + assert output3[0][0].metadata["abbr"] == "llm" + for doc_id in doc_ids: docsearch._delete_doc(doc_id)