langchain 0.0.202 → 0.0.204-rc.0

package/dist/embeddings/gradient_ai.js

@@ -1,99 +1 @@
-import { Gradient } from "@gradientai/nodejs-sdk";
-import { getEnvironmentVariable } from "../util/env.js";
-import { chunkArray } from "../util/chunk.js";
-import { Embeddings } from "./base.js";
-/**
- * Class for generating embeddings using the Gradient AI's API. Extends the
- * Embeddings class and implements GradientEmbeddingsParams and
- */
-export class GradientEmbeddings extends Embeddings {
-    constructor(fields) {
-        super(fields);
-        Object.defineProperty(this, "gradientAccessKey", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "workspaceId", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "batchSize", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: 128
-        });
-        // eslint-disable-next-line @typescript-eslint/no-explicit-any
-        Object.defineProperty(this, "model", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        this.gradientAccessKey =
-            fields?.gradientAccessKey ??
-                getEnvironmentVariable("GRADIENT_ACCESS_TOKEN");
-        this.workspaceId =
-            fields?.workspaceId ?? getEnvironmentVariable("GRADIENT_WORKSPACE_ID");
-        if (!this.gradientAccessKey) {
-            throw new Error("Missing Gradient AI Access Token");
-        }
-        if (!this.workspaceId) {
-            throw new Error("Missing Gradient AI Workspace ID");
-        }
-    }
-    /**
-     * Method to generate embeddings for an array of documents. Splits the
-     * documents into batches and makes requests to the Gradient API to generate
-     * embeddings.
-     * @param texts Array of documents to generate embeddings for.
-     * @returns Promise that resolves to a 2D array of embeddings for each document.
-     */
-    async embedDocuments(texts) {
-        await this.setModel();
-        const mappedTexts = texts.map((text) => ({ input: text }));
-        const batches = chunkArray(mappedTexts, this.batchSize);
-        const batchRequests = batches.map((batch) => this.caller.call(async () => this.model.generateEmbeddings({
-            inputs: batch,
-        })));
-        const batchResponses = await Promise.all(batchRequests);
-        const embeddings = [];
-        for (let i = 0; i < batchResponses.length; i += 1) {
-            const batch = batches[i];
-            const { embeddings: batchResponse } = batchResponses[i];
-            for (let j = 0; j < batch.length; j += 1) {
-                embeddings.push(batchResponse[j].embedding);
-            }
-        }
-        return embeddings;
-    }
-    /**
-     * Method to generate an embedding for a single document. Calls the
-     * embedDocuments method with the document as the input.
-     * @param text Document to generate an embedding for.
-     * @returns Promise that resolves to an embedding for the document.
-     */
-    async embedQuery(text) {
-        const data = await this.embedDocuments([text]);
-        return data[0];
-    }
-    /**
-     * Method to set the model to use for generating embeddings.
-     * @sets the class' `model` value to that of the retrieved Embeddings Model.
-     */
-    async setModel() {
-        if (this.model)
-            return;
-        const gradient = new Gradient({
-            accessToken: this.gradientAccessKey,
-            workspaceId: this.workspaceId,
-        });
-        this.model = await gradient.getEmbeddingsModel({
-            slug: "bge-large",
-        });
-    }
-}
+export * from "@langchain/community/embeddings/gradient_ai";

package/dist/embeddings/hf.cjs

@@ -1,77 +1,17 @@
 "use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.HuggingFaceInferenceEmbeddings = void 0;
-const inference_1 = require("@huggingface/inference");
-const base_js_1 = require("./base.cjs");
-const env_js_1 = require("../util/env.cjs");
-/**
- * Class that extends the Embeddings class and provides methods for
- * generating embeddings using Hugging Face models through the
- * HuggingFaceInference API.
- */
-class HuggingFaceInferenceEmbeddings extends base_js_1.Embeddings {
-    constructor(fields) {
-        super(fields ?? {});
-        Object.defineProperty(this, "apiKey", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "model", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "endpointUrl", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "client", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        this.model = fields?.model ?? "BAAI/bge-base-en-v1.5";
-        this.apiKey =
-            fields?.apiKey ?? (0, env_js_1.getEnvironmentVariable)("HUGGINGFACEHUB_API_KEY");
-        this.endpointUrl = fields?.endpointUrl;
-        this.client = this.endpointUrl
-            ? new inference_1.HfInference(this.apiKey).endpoint(this.endpointUrl)
-            : new inference_1.HfInference(this.apiKey);
-    }
-    async _embed(texts) {
-        // replace newlines, which can negatively affect performance.
-        const clean = texts.map((text) => text.replace(/\n/g, " "));
-        return this.caller.call(() => this.client.featureExtraction({
-            model: this.model,
-            inputs: clean,
-        }));
-    }
-    /**
-     * Method that takes a document as input and returns a promise that
-     * resolves to an embedding for the document. It calls the _embed method
-     * with the document as the input and returns the first embedding in the
-     * resulting array.
-     * @param document Document to generate an embedding for.
-     * @returns Promise that resolves to an embedding for the document.
-     */
-    embedQuery(document) {
-        return this._embed([document]).then((embeddings) => embeddings[0]);
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    var desc = Object.getOwnPropertyDescriptor(m, k);
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+      desc = { enumerable: true, get: function() { return m[k]; } };
     }
-    /**
-     * Method that takes an array of documents as input and returns a promise
-     * that resolves to a 2D array of embeddings for each document. It calls
-     * the _embed method with the documents as the input.
-     * @param documents Array of documents to generate embeddings for.
-     * @returns Promise that resolves to a 2D array of embeddings for each document.
-     */
-    embedDocuments(documents) {
-        return this._embed(documents);
-    }
-}
-exports.HuggingFaceInferenceEmbeddings = HuggingFaceInferenceEmbeddings;
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __exportStar = (this && this.__exportStar) || function(m, exports) {
+    for (var p in m) if (p !== "default" && !Object.prototype.hasOwnProperty.call(exports, p)) __createBinding(exports, m, p);
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+__exportStar(require("@langchain/community/embeddings/hf"), exports);

package/dist/embeddings/hf.d.ts

@@ -1,41 +1 @@
-import { HfInference, HfInferenceEndpoint } from "@huggingface/inference";
-import { Embeddings, EmbeddingsParams } from "./base.js";
-/**
- * Interface that extends EmbeddingsParams and defines additional
- * parameters specific to the HuggingFaceInferenceEmbeddings class.
- */
-export interface HuggingFaceInferenceEmbeddingsParams extends EmbeddingsParams {
-    apiKey?: string;
-    model?: string;
-    endpointUrl?: string;
-}
-/**
- * Class that extends the Embeddings class and provides methods for
- * generating embeddings using Hugging Face models through the
- * HuggingFaceInference API.
- */
-export declare class HuggingFaceInferenceEmbeddings extends Embeddings implements HuggingFaceInferenceEmbeddingsParams {
-    apiKey?: string;
-    model: string;
-    endpointUrl?: string;
-    client: HfInference | HfInferenceEndpoint;
-    constructor(fields?: HuggingFaceInferenceEmbeddingsParams);
-    _embed(texts: string[]): Promise<number[][]>;
-    /**
-     * Method that takes a document as input and returns a promise that
-     * resolves to an embedding for the document. It calls the _embed method
-     * with the document as the input and returns the first embedding in the
-     * resulting array.
-     * @param document Document to generate an embedding for.
-     * @returns Promise that resolves to an embedding for the document.
-     */
-    embedQuery(document: string): Promise<number[]>;
-    /**
-     * Method that takes an array of documents as input and returns a promise
-     * that resolves to a 2D array of embeddings for each document. It calls
-     * the _embed method with the documents as the input.
-     * @param documents Array of documents to generate embeddings for.
-     * @returns Promise that resolves to a 2D array of embeddings for each document.
-     */
-    embedDocuments(documents: string[]): Promise<number[][]>;
-}
+export * from "@langchain/community/embeddings/hf";

package/dist/embeddings/hf.js

@@ -1,73 +1 @@
-import { HfInference } from "@huggingface/inference";
-import { Embeddings } from "./base.js";
-import { getEnvironmentVariable } from "../util/env.js";
-/**
- * Class that extends the Embeddings class and provides methods for
- * generating embeddings using Hugging Face models through the
- * HuggingFaceInference API.
- */
-export class HuggingFaceInferenceEmbeddings extends Embeddings {
-    constructor(fields) {
-        super(fields ?? {});
-        Object.defineProperty(this, "apiKey", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "model", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "endpointUrl", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "client", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        this.model = fields?.model ?? "BAAI/bge-base-en-v1.5";
-        this.apiKey =
-            fields?.apiKey ?? getEnvironmentVariable("HUGGINGFACEHUB_API_KEY");
-        this.endpointUrl = fields?.endpointUrl;
-        this.client = this.endpointUrl
-            ? new HfInference(this.apiKey).endpoint(this.endpointUrl)
-            : new HfInference(this.apiKey);
-    }
-    async _embed(texts) {
-        // replace newlines, which can negatively affect performance.
-        const clean = texts.map((text) => text.replace(/\n/g, " "));
-        return this.caller.call(() => this.client.featureExtraction({
-            model: this.model,
-            inputs: clean,
-        }));
-    }
-    /**
-     * Method that takes a document as input and returns a promise that
-     * resolves to an embedding for the document. It calls the _embed method
-     * with the document as the input and returns the first embedding in the
-     * resulting array.
-     * @param document Document to generate an embedding for.
-     * @returns Promise that resolves to an embedding for the document.
-     */
-    embedQuery(document) {
-        return this._embed([document]).then((embeddings) => embeddings[0]);
-    }
-    /**
-     * Method that takes an array of documents as input and returns a promise
-     * that resolves to a 2D array of embeddings for each document. It calls
-     * the _embed method with the documents as the input.
-     * @param documents Array of documents to generate embeddings for.
-     * @returns Promise that resolves to a 2D array of embeddings for each document.
-     */
-    embedDocuments(documents) {
-        return this._embed(documents);
-    }
-}
+export * from "@langchain/community/embeddings/hf";

package/dist/embeddings/hf_transformers.cjs

@@ -1,89 +1,17 @@
 "use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.HuggingFaceTransformersEmbeddings = void 0;
-const transformers_1 = require("@xenova/transformers");
-const chunk_js_1 = require("../util/chunk.cjs");
-const base_js_1 = require("./base.cjs");
-/**
- * @example
- * ```typescript
- * const model = new HuggingFaceTransformersEmbeddings({
- *   modelName: "Xenova/all-MiniLM-L6-v2",
- * });
- *
- * // Embed a single query
- * const res = await model.embedQuery(
- *   "What would be a good company name for a company that makes colorful socks?"
- * );
- * console.log({ res });
- *
- * // Embed multiple documents
- * const documentRes = await model.embedDocuments(["Hello world", "Bye bye"]);
- * console.log({ documentRes });
- * ```
- */
-class HuggingFaceTransformersEmbeddings extends base_js_1.Embeddings {
-    constructor(fields) {
-        super(fields ?? {});
-        Object.defineProperty(this, "modelName", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: "Xenova/all-MiniLM-L6-v2"
-        });
-        Object.defineProperty(this, "batchSize", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: 512
-        });
-        Object.defineProperty(this, "stripNewLines", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: true
-        });
-        Object.defineProperty(this, "timeout", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "pipelinePromise", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        this.modelName = fields?.modelName ?? this.modelName;
-        this.stripNewLines = fields?.stripNewLines ?? this.stripNewLines;
-        this.timeout = fields?.timeout;
-    }
-    async embedDocuments(texts) {
-        const batches = (0, chunk_js_1.chunkArray)(this.stripNewLines ? texts.map((t) => t.replace(/\n/g, " ")) : texts, this.batchSize);
-        const batchRequests = batches.map((batch) => this.runEmbedding(batch));
-        const batchResponses = await Promise.all(batchRequests);
-        const embeddings = [];
-        for (let i = 0; i < batchResponses.length; i += 1) {
-            const batchResponse = batchResponses[i];
-            for (let j = 0; j < batchResponse.length; j += 1) {
-                embeddings.push(batchResponse[j]);
-            }
-        }
-        return embeddings;
-    }
-    async embedQuery(text) {
-        const data = await this.runEmbedding([
-            this.stripNewLines ? text.replace(/\n/g, " ") : text,
-        ]);
-        return data[0];
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    var desc = Object.getOwnPropertyDescriptor(m, k);
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+      desc = { enumerable: true, get: function() { return m[k]; } };
     }
-    async runEmbedding(texts) {
-        const pipe = await (this.pipelinePromise ??= (0, transformers_1.pipeline)("feature-extraction", this.modelName));
-        return this.caller.call(async () => {
-            const output = await pipe(texts, { pooling: "mean", normalize: true });
-            return output.tolist();
-        });
-    }
-}
-exports.HuggingFaceTransformersEmbeddings = HuggingFaceTransformersEmbeddings;
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __exportStar = (this && this.__exportStar) || function(m, exports) {
+    for (var p in m) if (p !== "default" && !Object.prototype.hasOwnProperty.call(exports, p)) __createBinding(exports, m, p);
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+__exportStar(require("@langchain/community/embeddings/hf_transformers"), exports);

package/dist/embeddings/hf_transformers.d.ts

@@ -1,47 +1 @@
-import { Embeddings, EmbeddingsParams } from "./base.js";
-export interface HuggingFaceTransformersEmbeddingsParams extends EmbeddingsParams {
-    /** Model name to use */
-    modelName: string;
-    /**
-     * Timeout to use when making requests to OpenAI.
-     */
-    timeout?: number;
-    /**
-     * The maximum number of documents to embed in a single request.
-     */
-    batchSize?: number;
-    /**
-     * Whether to strip new lines from the input text. This is recommended by
-     * OpenAI, but may not be suitable for all use cases.
-     */
-    stripNewLines?: boolean;
-}
-/**
- * @example
- * ```typescript
- * const model = new HuggingFaceTransformersEmbeddings({
- *   modelName: "Xenova/all-MiniLM-L6-v2",
- * });
- *
- * // Embed a single query
- * const res = await model.embedQuery(
- *   "What would be a good company name for a company that makes colorful socks?"
- * );
- * console.log({ res });
- *
- * // Embed multiple documents
- * const documentRes = await model.embedDocuments(["Hello world", "Bye bye"]);
- * console.log({ documentRes });
- * ```
- */
-export declare class HuggingFaceTransformersEmbeddings extends Embeddings implements HuggingFaceTransformersEmbeddingsParams {
-    modelName: string;
-    batchSize: number;
-    stripNewLines: boolean;
-    timeout?: number;
-    private pipelinePromise;
-    constructor(fields?: Partial<HuggingFaceTransformersEmbeddingsParams>);
-    embedDocuments(texts: string[]): Promise<number[][]>;
-    embedQuery(text: string): Promise<number[]>;
-    private runEmbedding;
-}
+export * from "@langchain/community/embeddings/hf_transformers";

package/dist/embeddings/hf_transformers.js

@@ -1,85 +1 @@
-import { pipeline } from "@xenova/transformers";
-import { chunkArray } from "../util/chunk.js";
-import { Embeddings } from "./base.js";
-/**
- * @example
- * ```typescript
- * const model = new HuggingFaceTransformersEmbeddings({
- *   modelName: "Xenova/all-MiniLM-L6-v2",
- * });
- *
- * // Embed a single query
- * const res = await model.embedQuery(
- *   "What would be a good company name for a company that makes colorful socks?"
- * );
- * console.log({ res });
- *
- * // Embed multiple documents
- * const documentRes = await model.embedDocuments(["Hello world", "Bye bye"]);
- * console.log({ documentRes });
- * ```
- */
-export class HuggingFaceTransformersEmbeddings extends Embeddings {
-    constructor(fields) {
-        super(fields ?? {});
-        Object.defineProperty(this, "modelName", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: "Xenova/all-MiniLM-L6-v2"
-        });
-        Object.defineProperty(this, "batchSize", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: 512
-        });
-        Object.defineProperty(this, "stripNewLines", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: true
-        });
-        Object.defineProperty(this, "timeout", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "pipelinePromise", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        this.modelName = fields?.modelName ?? this.modelName;
-        this.stripNewLines = fields?.stripNewLines ?? this.stripNewLines;
-        this.timeout = fields?.timeout;
-    }
-    async embedDocuments(texts) {
-        const batches = chunkArray(this.stripNewLines ? texts.map((t) => t.replace(/\n/g, " ")) : texts, this.batchSize);
-        const batchRequests = batches.map((batch) => this.runEmbedding(batch));
-        const batchResponses = await Promise.all(batchRequests);
-        const embeddings = [];
-        for (let i = 0; i < batchResponses.length; i += 1) {
-            const batchResponse = batchResponses[i];
-            for (let j = 0; j < batchResponse.length; j += 1) {
-                embeddings.push(batchResponse[j]);
-            }
-        }
-        return embeddings;
-    }
-    async embedQuery(text) {
-        const data = await this.runEmbedding([
-            this.stripNewLines ? text.replace(/\n/g, " ") : text,
-        ]);
-        return data[0];
-    }
-    async runEmbedding(texts) {
-        const pipe = await (this.pipelinePromise ??= pipeline("feature-extraction", this.modelName));
-        return this.caller.call(async () => {
-            const output = await pipe(texts, { pooling: "mean", normalize: true });
-            return output.tolist();
-        });
-    }
-}
+export * from "@langchain/community/embeddings/hf_transformers";

package/dist/embeddings/llama_cpp.cjs

@@ -1,83 +1,17 @@
 "use strict";
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.LlamaCppEmbeddings = void 0;
-const llama_cpp_js_1 = require("../util/llama_cpp.cjs");
-const base_js_1 = require("./base.cjs");
-/**
- * @example
- * ```typescript
- * // Initialize LlamaCppEmbeddings with the path to the model file
- * const embeddings = new LlamaCppEmbeddings({
- *   modelPath: "/Replace/with/path/to/your/model/gguf-llama2-q4_0.bin",
- * });
- *
- * // Embed a query string using the Llama embeddings
- * const res = embeddings.embedQuery("Hello Llama!");
- *
- * // Output the resulting embeddings
- * console.log(res);
- *
- * ```
- */
-class LlamaCppEmbeddings extends base_js_1.Embeddings {
-    constructor(inputs) {
-        super(inputs);
-        Object.defineProperty(this, "_model", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "_context", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        const _inputs = inputs;
-        _inputs.embedding = true;
-        this._model = (0, llama_cpp_js_1.createLlamaModel)(_inputs);
-        this._context = (0, llama_cpp_js_1.createLlamaContext)(this._model, _inputs);
-    }
-    /**
-     * Generates embeddings for an array of texts.
-     * @param texts - An array of strings to generate embeddings for.
-     * @returns A Promise that resolves to an array of embeddings.
-     */
-    async embedDocuments(texts) {
-        const tokensArray = [];
-        for (const text of texts) {
-            const encodings = await this.caller.call(() => new Promise((resolve) => {
-                resolve(this._context.encode(text));
-            }));
-            tokensArray.push(encodings);
-        }
-        const embeddings = [];
-        for (const tokens of tokensArray) {
-            const embedArray = [];
-            for (let i = 0; i < tokens.length; i += 1) {
-                const nToken = +tokens[i];
-                embedArray.push(nToken);
-            }
-            embeddings.push(embedArray);
-        }
-        return embeddings;
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    var desc = Object.getOwnPropertyDescriptor(m, k);
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+      desc = { enumerable: true, get: function() { return m[k]; } };
     }
-    /**
-     * Generates an embedding for a single text.
-     * @param text - A string to generate an embedding for.
-     * @returns A Promise that resolves to an array of numbers representing the embedding.
-     */
-    async embedQuery(text) {
-        const tokens = [];
-        const encodings = await this.caller.call(() => new Promise((resolve) => {
-            resolve(this._context.encode(text));
-        }));
-        for (let i = 0; i < encodings.length; i += 1) {
-            const token = +encodings[i];
-            tokens.push(token);
-        }
-        return tokens;
-    }
-}
-exports.LlamaCppEmbeddings = LlamaCppEmbeddings;
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __exportStar = (this && this.__exportStar) || function(m, exports) {
+    for (var p in m) if (p !== "default" && !Object.prototype.hasOwnProperty.call(exports, p)) __createBinding(exports, m, p);
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+__exportStar(require("@langchain/community/embeddings/llama_cpp"), exports);