langchain 0.2.2 → 0.2.3

package/dist/output_parsers/tests/structured.int.test.js

@@ -0,0 +1,150 @@
+import { expect, test } from "@jest/globals";
+import { z } from "zod";
+import { OpenAI, ChatOpenAI } from "@langchain/openai";
+import { ChatPromptTemplate, PromptTemplate, SystemMessagePromptTemplate, } from "@langchain/core/prompts";
+import { LLMChain } from "../../chains/index.js";
+import { StructuredOutputParser } from "../structured.js";
+test("StructuredOutputParser deals special chars in prompt with llm model", async () => {
+    const model = new OpenAI({
+        temperature: 0,
+    });
+    const parser = StructuredOutputParser.fromNamesAndDescriptions({
+        question1: "a very on-topic question",
+        question2: "a super weird question",
+        question3: "an on-topic, but slightly creative",
+    });
+    const prompt = new PromptTemplate({
+        template: "context:\n{context}\n---{format_instructions}",
+        inputVariables: ["context"],
+        partialVariables: {
+            format_instructions: parser.getFormatInstructions(),
+        },
+    });
+    const chain = new LLMChain({
+        llm: model,
+        prompt,
+        outputParser: parser,
+        outputKey: "questions",
+    });
+    const result = await chain.call({
+        context: `The U2 ur-myth begins in 1976, when drummer Larry Mullen wanted to form a band.
+      He picked four school friends from Mount Temple Comprehensive School in Dublin.
+      “Larry formed U2,” says Paul McGuinness, U2’s manager from the beginning. “He
+      auditioned the other three and he chose them. The first name of U2 was the Larry
+      Mullen band,” McGuinness laughs. “And he never lets us forget it.” `,
+    });
+    console.log("response", result);
+    expect(result.questions).toHaveProperty("question1");
+    expect(result.questions).toHaveProperty("question2");
+    expect(result.questions).toHaveProperty("question3");
+});
+test("StructuredOutputParser deals special chars in prompt with chat model", async () => {
+    const model = new ChatOpenAI({
+        temperature: 0,
+    });
+    const parser = StructuredOutputParser.fromNamesAndDescriptions({
+        question1: "a very on-topic question",
+        question2: "a super weird question",
+        question3: "an on-topic, but slightly creative",
+    });
+    const prompt = new ChatPromptTemplate({
+        promptMessages: [
+            SystemMessagePromptTemplate.fromTemplate("context:\n{context}\n---"),
+            SystemMessagePromptTemplate.fromTemplate(`{format_instructions}`),
+        ],
+        inputVariables: ["context"],
+        partialVariables: {
+            format_instructions: parser.getFormatInstructions(),
+        },
+    });
+    const chain = new LLMChain({
+        llm: model,
+        prompt,
+        outputParser: parser,
+        outputKey: "questions",
+    });
+    const result = await chain.call({
+        context: `The U2 ur-myth begins in 1976, when drummer Larry Mullen wanted to form a band.
+        He picked four school friends from Mount Temple Comprehensive School in Dublin.
+        “Larry formed U2,” says Paul McGuinness, U2’s manager from the beginning. “He
+        auditioned the other three and he chose them. The first name of U2 was the Larry
+        Mullen band,” McGuinness laughs. “And he never lets us forget it.” `,
+    });
+    console.log("response", result);
+    expect(result.questions).toHaveProperty("question1");
+    expect(result.questions).toHaveProperty("question2");
+    expect(result.questions).toHaveProperty("question3");
+});
+test("StructuredOutputParser deals special chars in prompt with chat model 2", async () => {
+    const model = new ChatOpenAI({
+        temperature: 0,
+    });
+    const parser = StructuredOutputParser.fromNamesAndDescriptions({
+        question1: "a very on-topic question",
+        question2: "a super weird question",
+        question3: "an on-topic, but slightly creative",
+    });
+    const prompt = new ChatPromptTemplate({
+        promptMessages: [
+            SystemMessagePromptTemplate.fromTemplate("context:\n{context}\n---"),
+            SystemMessagePromptTemplate.fromTemplate(`{format_instructions}`),
+        ],
+        inputVariables: ["context"],
+        partialVariables: {
+            format_instructions: parser.getFormatInstructions(),
+        },
+    });
+    const chain = new LLMChain({
+        llm: model,
+        prompt,
+        outputKey: "questions",
+    });
+    const result = await chain.call({
+        context: `The U2 ur-myth begins in 1976, when drummer Larry Mullen wanted to form a band.
+          He picked four school friends from Mount Temple Comprehensive School in Dublin.
+          “Larry formed U2,” says Paul McGuinness, U2’s manager from the beginning. “He
+          auditioned the other three and he chose them. The first name of U2 was the Larry
+          Mullen band,” McGuinness laughs. “And he never lets us forget it.” `,
+    });
+    console.log("response", result);
+    const parsed = await parser.parse(result.questions);
+    expect(parsed).toHaveProperty("question1");
+    expect(parsed).toHaveProperty("question2");
+    expect(parsed).toHaveProperty("question3");
+});
+test("StructuredOutputParser handles a longer and more complex schema", async () => {
+    const parser = StructuredOutputParser.fromZodSchema(z.object({
+        name: z.string().describe("Human name"),
+        surname: z.string().describe("Human surname"),
+        age: z.number().describe("Human age"),
+        appearance: z.string().describe("Human appearance description"),
+        shortBio: z.string().describe("Short bio secription"),
+        university: z.string().optional().describe("University name if attended"),
+        gender: z.string().describe("Gender of the human"),
+        interests: z
+            .array(z.string())
+            .describe("json array of strings human interests"),
+    }));
+    const formatInstructions = parser.getFormatInstructions();
+    const prompt = new PromptTemplate({
+        template: "Generate details of a hypothetical person.\n{format_instructions}\nPerson description: {inputText}",
+        inputVariables: ["inputText"],
+        partialVariables: { format_instructions: formatInstructions },
+    });
+    const model = new OpenAI({ temperature: 0.5, modelName: "gpt-3.5-turbo" });
+    const input = await prompt.format({
+        inputText: "A man, living in Poland.",
+    });
+    const response = await model.invoke(input);
+    console.log("response", response);
+    const parsed = await parser.parse(response);
+    expect(parsed).toHaveProperty("name");
+    expect(parsed).toHaveProperty("surname");
+    expect(parsed).toHaveProperty("age");
+    expect(parsed).toHaveProperty("appearance");
+    expect(parsed).toHaveProperty("shortBio");
+    expect(parsed).toHaveProperty("age");
+    expect(parsed).toHaveProperty("gender");
+    expect(parsed).toHaveProperty("interests");
+    expect(parsed.interests.length).toBeGreaterThan(0);
+});

package/dist/prompts/tests/selectors.test.d.ts

@@ -0,0 +1 @@
+export {};

package/dist/prompts/tests/selectors.test.js

@@ -0,0 +1,59 @@
+import { expect, test } from "@jest/globals";
+import { LengthBasedExampleSelector, SemanticSimilarityExampleSelector, } from "@langchain/core/example_selectors";
+import { PromptTemplate } from "@langchain/core/prompts";
+import { FakeEmbeddings } from "@langchain/core/utils/testing";
+import { MemoryVectorStore } from "../../vectorstores/memory.js";
+test("Test using LengthBasedExampleSelector", async () => {
+    const prompt = new PromptTemplate({
+        template: "{foo} {bar}",
+        inputVariables: ["foo"],
+        partialVariables: { bar: "baz" },
+    });
+    const selector = await LengthBasedExampleSelector.fromExamples([{ foo: "one one one" }], {
+        examplePrompt: prompt,
+        maxLength: 10,
+    });
+    await selector.addExample({ foo: "one two three" });
+    await selector.addExample({ foo: "four five six" });
+    await selector.addExample({ foo: "seven eight nine" });
+    await selector.addExample({ foo: "ten eleven twelve" });
+    const chosen = await selector.selectExamples({ foo: "hello", bar: "world" });
+    expect(chosen).toStrictEqual([
+        { foo: "one one one" },
+        { foo: "one two three" },
+    ]);
+});
+test("Test using SemanticSimilarityExampleSelector", async () => {
+    const vectorStore = await MemoryVectorStore.fromTexts(["Hello world", "Bye bye", "hello nice world", "bye", "hi"], [{ id: 2 }, { id: 1 }, { id: 3 }, { id: 4 }, { id: 5 }], new FakeEmbeddings() // not using  OpenAIEmbeddings() because would be extra dependency
+    );
+    const selector = new SemanticSimilarityExampleSelector({
+        vectorStore,
+    });
+    const chosen = await selector.selectExamples({ id: 1 });
+    expect(chosen).toEqual([{ id: 2 }, { id: 1 }, { id: 3 }, { id: 4 }]);
+});
+test("Test using SemanticSimilarityExampleSelector with metadata filtering", async () => {
+    const vectorStore = await MemoryVectorStore.fromTexts(["Hello world", "Bye bye", "hello nice world", "bye", "hi"], [{ id: 2 }, { id: 1 }, { id: 3 }, { id: 4 }, { id: 5 }], new FakeEmbeddings() // not using  OpenAIEmbeddings() because would be extra dependency
+    );
+    const selector = new SemanticSimilarityExampleSelector({
+        vectorStore,
+        filter: (doc) => doc.metadata.id === 2,
+    });
+    const chosen = await selector.selectExamples({ id: 1 });
+    expect(chosen).toEqual([{ id: 2 }]);
+});
+test("Test using SemanticSimilarityExampleSelector with a passed in retriever", async () => {
+    const vectorStore = await MemoryVectorStore.fromTexts(["Hello world", "Bye bye", "hello nice world", "bye", "hi"], [{ id: 2 }, { id: 1 }, { id: 3 }, { id: 4 }, { id: 5 }], new FakeEmbeddings() // not using  OpenAIEmbeddings() because would be extra dependency
+    );
+    const selector = new SemanticSimilarityExampleSelector({
+        vectorStoreRetriever: vectorStore.asRetriever({ k: 5 }),
+    });
+    const chosen = await selector.selectExamples({ id: 1 });
+    expect(chosen).toEqual([
+        { id: 2 },
+        { id: 1 },
+        { id: 3 },
+        { id: 4 },
+        { id: 5 },
+    ]);
+});

package/dist/retrievers/self_query/tests/memory_self_query.int.test.d.ts

@@ -0,0 +1 @@
+export {};

package/dist/retrievers/self_query/tests/memory_self_query.int.test.js

@@ -0,0 +1,330 @@
+import { test } from "@jest/globals";
+import { Document } from "@langchain/core/documents";
+import { OpenAIEmbeddings, OpenAI } from "@langchain/openai";
+import { SelfQueryRetriever } from "../index.js";
+import { FunctionalTranslator } from "../functional.js";
+import { MemoryVectorStore } from "../../../vectorstores/memory.js";
+test("Memory Vector Store Self Query Retriever Test", async () => {
+    const docs = [
+        new Document({
+            pageContent: "A bunch of scientists bring back dinosaurs and mayhem breaks loose",
+            metadata: { year: 1993, rating: 7.7, genre: "science fiction" },
+        }),
+        new Document({
+            pageContent: "Leo DiCaprio gets lost in a dream within a dream within a dream within a ...",
+            metadata: { year: 2010, director: "Christopher Nolan", rating: 8.2 },
+        }),
+        new Document({
+            pageContent: "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 },
+        }),
+        new Document({
+            pageContent: "A bunch of normal-sized women are supremely wholesome and some men pine after them",
+            metadata: { year: 2019, director: "Greta Gerwig", rating: 8.3 },
+        }),
+        new Document({
+            pageContent: "Toys come alive and have a blast doing so",
+            metadata: { year: 1995, genre: "animated" },
+        }),
+        new Document({
+            pageContent: "Three men walk into the Zone, three men walk out of the Zone",
+            metadata: {
+                year: 1979,
+                director: "Andrei Tarkovsky",
+                genre: "science fiction",
+                rating: 9.9,
+            },
+        }),
+    ];
+    const attributeInfo = [
+        {
+            name: "genre",
+            description: "The genre of the movie",
+            type: "string or array of strings",
+        },
+        {
+            name: "year",
+            description: "The year the movie was released",
+            type: "number",
+        },
+        {
+            name: "director",
+            description: "The director of the movie",
+            type: "string",
+        },
+        {
+            name: "rating",
+            description: "The rating of the movie (1-10)",
+            type: "number",
+        },
+        {
+            name: "length",
+            description: "The length of the movie in minutes",
+            type: "number",
+        },
+    ];
+    const embeddings = new OpenAIEmbeddings();
+    const llm = new OpenAI({
+        modelName: "gpt-3.5-turbo",
+    });
+    const documentContents = "Brief summary of a movie";
+    const vectorStore = await MemoryVectorStore.fromDocuments(docs, embeddings);
+    const selfQueryRetriever = SelfQueryRetriever.fromLLM({
+        llm,
+        vectorStore,
+        documentContents,
+        attributeInfo,
+        structuredQueryTranslator: new FunctionalTranslator(),
+    });
+    const query1 = await selfQueryRetriever.getRelevantDocuments("Which movies are less than 90 minutes?");
+    console.log(query1);
+    expect(query1.length).toEqual(0);
+    const query2 = await selfQueryRetriever.getRelevantDocuments("Which movies are rated higher than 8.5?");
+    console.log(query2);
+    expect(query2.length).toEqual(2);
+    const query3 = await selfQueryRetriever.getRelevantDocuments("Which movies are directed by Greta Gerwig?");
+    console.log(query3);
+    expect(query3.length).toEqual(1);
+});
+test("Memory Vector Store Self Query Retriever Test With Default Filter Or Merge Operator", async () => {
+    const docs = [
+        new Document({
+            pageContent: "A bunch of scientists bring back dinosaurs and mayhem breaks loose",
+            metadata: {
+                year: 1993,
+                rating: 7.7,
+                genre: "science fiction",
+                type: "movie",
+            },
+        }),
+        new Document({
+            pageContent: "Leo DiCaprio gets lost in a dream within a dream within a dream within a ...",
+            metadata: {
+                year: 2010,
+                director: "Christopher Nolan",
+                rating: 8.2,
+                type: "movie",
+            },
+        }),
+        new Document({
+            pageContent: "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,
+                type: "movie",
+            },
+        }),
+        new Document({
+            pageContent: "A bunch of normal-sized women are supremely wholesome and some men pine after them",
+            metadata: {
+                year: 2019,
+                director: "Greta Gerwig",
+                rating: 8.3,
+                type: "movie",
+            },
+        }),
+        new Document({
+            pageContent: "Toys come alive and have a blast doing so",
+            metadata: { year: 1995, genre: "animated", type: "movie" },
+        }),
+        new Document({
+            pageContent: "Three men walk into the Zone, three men walk out of the Zone",
+            metadata: {
+                year: 1979,
+                director: "Andrei Tarkovsky",
+                genre: "science fiction",
+                rating: 9.9,
+                type: "movie",
+            },
+        }),
+        new Document({
+            pageContent: "10x the previous gecs",
+            metadata: {
+                year: 2023,
+                title: "10000 gecs",
+                artist: "100 gecs",
+                rating: 9.9,
+                type: "album",
+            },
+        }),
+    ];
+    const attributeInfo = [
+        {
+            name: "genre",
+            description: "The genre of the movie",
+            type: "string or array of strings",
+        },
+        {
+            name: "year",
+            description: "The year the movie was released",
+            type: "number",
+        },
+        {
+            name: "director",
+            description: "The director of the movie",
+            type: "string",
+        },
+        {
+            name: "rating",
+            description: "The rating of the movie (1-10)",
+            type: "number",
+        },
+        {
+            name: "length",
+            description: "The length of the movie in minutes",
+            type: "number",
+        },
+    ];
+    const embeddings = new OpenAIEmbeddings();
+    const llm = new OpenAI({
+        modelName: "gpt-3.5-turbo",
+    });
+    const documentContents = "Brief summary of a movie";
+    const vectorStore = await MemoryVectorStore.fromDocuments(docs, embeddings);
+    const selfQueryRetriever = SelfQueryRetriever.fromLLM({
+        llm,
+        vectorStore,
+        documentContents,
+        attributeInfo,
+        structuredQueryTranslator: new FunctionalTranslator(),
+        searchParams: {
+            filter: (doc) => doc.metadata && doc.metadata.type === "movie",
+            mergeFiltersOperator: "or",
+            k: docs.length,
+        },
+    });
+    const query1 = await selfQueryRetriever.getRelevantDocuments("Which movies are less than 90 minutes?");
+    console.log(query1);
+    expect(query1.length).toEqual(6);
+    const query2 = await selfQueryRetriever.getRelevantDocuments("Which movies are rated higher than 8.5?");
+    console.log(query2);
+    expect(query2.length).toEqual(7);
+    const query3 = await selfQueryRetriever.getRelevantDocuments("Which movies are directed by Greta Gerwig?");
+    console.log(query3);
+    expect(query3.length).toEqual(6);
+    const query4 = await selfQueryRetriever.getRelevantDocuments("Awawawa au au au wawawawa hello?");
+    console.log(query4);
+    expect(query4.length).toEqual(6); // this one should return documents since default filter takes over
+});
+test("Memory Vector Store Self Query Retriever Test With Default Filter And Merge Operator", async () => {
+    const docs = [
+        new Document({
+            pageContent: "A bunch of scientists bring back dinosaurs and mayhem breaks loose",
+            metadata: {
+                year: 1993,
+                rating: 7.7,
+                genre: "science fiction",
+                type: "movie",
+            },
+        }),
+        new Document({
+            pageContent: "Leo DiCaprio gets lost in a dream within a dream within a dream within a ...",
+            metadata: {
+                year: 2010,
+                director: "Christopher Nolan",
+                rating: 8.2,
+                type: "movie",
+            },
+        }),
+        new Document({
+            pageContent: "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,
+                type: "movie",
+            },
+        }),
+        new Document({
+            pageContent: "A bunch of normal-sized women are supremely wholesome and some men pine after them",
+            metadata: {
+                year: 2019,
+                director: "Greta Gerwig",
+                rating: 8.3,
+                type: "movie",
+            },
+        }),
+        new Document({
+            pageContent: "Toys come alive and have a blast doing so",
+            metadata: { year: 1995, genre: "animated", type: "movie" },
+        }),
+        new Document({
+            pageContent: "Three men walk into the Zone, three men walk out of the Zone",
+            metadata: {
+                year: 1979,
+                director: "Andrei Tarkovsky",
+                genre: "science fiction",
+                rating: 9.9,
+                type: "movie",
+            },
+        }),
+        new Document({
+            pageContent: "10x the previous gecs",
+            metadata: {
+                year: 2023,
+                title: "10000 gecs",
+                artist: "100 gecs",
+                rating: 9.9,
+                type: "album",
+            },
+        }),
+    ];
+    const attributeInfo = [
+        {
+            name: "genre",
+            description: "The genre of the movie",
+            type: "string or array of strings",
+        },
+        {
+            name: "year",
+            description: "The year the movie was released",
+            type: "number",
+        },
+        {
+            name: "director",
+            description: "The director of the movie",
+            type: "string",
+        },
+        {
+            name: "rating",
+            description: "The rating of the movie (1-10)",
+            type: "number",
+        },
+        {
+            name: "length",
+            description: "The length of the movie in minutes",
+            type: "number",
+        },
+    ];
+    const embeddings = new OpenAIEmbeddings();
+    const llm = new OpenAI({
+        modelName: "gpt-3.5-turbo",
+    });
+    const documentContents = "Brief summary of a movie";
+    const vectorStore = await MemoryVectorStore.fromDocuments(docs, embeddings);
+    const selfQueryRetriever = SelfQueryRetriever.fromLLM({
+        llm,
+        vectorStore,
+        documentContents,
+        attributeInfo,
+        structuredQueryTranslator: new FunctionalTranslator(),
+        searchParams: {
+            filter: (doc) => doc.metadata && doc.metadata.type === "movie",
+            mergeFiltersOperator: "and",
+            k: docs.length,
+        },
+    });
+    const query1 = await selfQueryRetriever.getRelevantDocuments("Which movies are less than 90 minutes?");
+    console.log(query1);
+    expect(query1.length).toEqual(0);
+    const query2 = await selfQueryRetriever.getRelevantDocuments("Which movies are rated higher than 8.5?");
+    console.log(query2);
+    expect(query2.length).toEqual(2);
+    const query3 = await selfQueryRetriever.getRelevantDocuments("Which movies are directed by Greta Gerwig?");
+    console.log(query3);
+    expect(query3.length).toEqual(1);
+    const query4 = await selfQueryRetriever.getRelevantDocuments("Awawawa au au au wawawawa hello?");
+    console.log(query4);
+    expect(query4.length).toEqual(0); // this one should return documents since default filter takes over
+});

package/dist/retrievers/tests/chain_extract.int.test.d.ts

@@ -0,0 +1 @@
+export {};

package/dist/retrievers/tests/chain_extract.int.test.js

@@ -0,0 +1,32 @@
+import { test, expect } from "@jest/globals";
+import { OpenAI, OpenAIEmbeddings } from "@langchain/openai";
+import { PromptTemplate } from "@langchain/core/prompts";
+import { LLMChain } from "../../chains/llm_chain.js";
+import { StuffDocumentsChain } from "../../chains/combine_docs_chain.js";
+import { ConversationalRetrievalQAChain } from "../../chains/conversational_retrieval_chain.js";
+import { MemoryVectorStore } from "../../vectorstores/memory.js";
+import { ContextualCompressionRetriever } from "../contextual_compression.js";
+import { LLMChainExtractor } from "../document_compressors/chain_extract.js";
+test("Test LLMChainExtractor", async () => {
+    const model = new OpenAI({ modelName: "gpt-3.5-turbo-instruct" });
+    const prompt = PromptTemplate.fromTemplate("Print {question}, and ignore {chat_history}");
+    const baseCompressor = LLMChainExtractor.fromLLM(model);
+    expect(baseCompressor).toBeDefined();
+    const retriever = new ContextualCompressionRetriever({
+        baseCompressor,
+        baseRetriever: await MemoryVectorStore.fromTexts(["Hello world", "Bye bye", "hello nice world", "bye", "hi"], [{ id: 2 }, { id: 1 }, { id: 3 }, { id: 4 }, { id: 5 }], new OpenAIEmbeddings()).then((vectorStore) => vectorStore.asRetriever()),
+    });
+    const llmChain = new LLMChain({ prompt, llm: model });
+    const combineDocsChain = new StuffDocumentsChain({
+        llmChain,
+        documentVariableName: "foo",
+    });
+    const chain = new ConversationalRetrievalQAChain({
+        retriever,
+        combineDocumentsChain: combineDocsChain,
+        questionGeneratorChain: llmChain,
+    });
+    const res = await chain.call({ question: "foo", chat_history: "bar" });
+    expect(res.text.length).toBeGreaterThan(0);
+    console.log({ res });
+});

package/dist/retrievers/tests/hyde.int.test.d.ts

@@ -0,0 +1 @@
+export {};

package/dist/retrievers/tests/hyde.int.test.js

@@ -0,0 +1,44 @@
+import { expect, test } from "@jest/globals";
+import { OpenAIEmbeddings, OpenAI } from "@langchain/openai";
+import { Document } from "@langchain/core/documents";
+import { MemoryVectorStore } from "../../vectorstores/memory.js";
+import { HydeRetriever } from "../hyde.js";
+test("Hyde retriever", async () => {
+    const embeddings = new OpenAIEmbeddings();
+    const vectorStore = new MemoryVectorStore(embeddings);
+    const llm = new OpenAI();
+    const retriever = new HydeRetriever({
+        vectorStore,
+        llm,
+        k: 1,
+    });
+    await vectorStore.addDocuments([
+        "My name is John.",
+        "My name is Bob.",
+        "My favourite food is pizza.",
+        "My favourite food is pasta.",
+    ].map((pageContent) => new Document({ pageContent })));
+    const results = await retriever.getRelevantDocuments("What is my favourite food?");
+    expect(results.length).toBe(1);
+    console.log(results);
+});
+test("Hyde retriever with default prompt template", async () => {
+    const embeddings = new OpenAIEmbeddings();
+    const vectorStore = new MemoryVectorStore(embeddings);
+    const llm = new OpenAI();
+    const retriever = new HydeRetriever({
+        vectorStore,
+        llm,
+        k: 1,
+        promptTemplate: "websearch",
+    });
+    await vectorStore.addDocuments([
+        "My name is John.",
+        "My name is Bob.",
+        "My favourite food is pizza.",
+        "My favourite food is pasta.",
+    ].map((pageContent) => new Document({ pageContent })));
+    const results = await retriever.getRelevantDocuments("What is my favourite food?");
+    expect(results.length).toBe(1);
+    console.log(results);
+});

package/dist/retrievers/tests/matryoshka_retriever.int.test.d.ts

@@ -0,0 +1 @@
+export {};