@huggingface/inference 2.1.3 → 2.2.1

package/README.md

@@ -188,6 +188,22 @@ await hf.documentQuestionAnswering({
   }
 })
 
+// Tabular
+
+await hf.tabularRegression({
+  model: "scikit-learn/Fish-Weight",
+  inputs: {
+    data: {
+      "Height":["11.52", "12.48", "12.3778"],
+      "Length1":["23.2", "24", "23.9"],
+      "Length2":["25.4", "26.3", "26.5"],
+      "Length3":["30", "31.2", "31.1"],
+      "Species":["Bream", "Bream", "Bream"],
+      "Width":["4.02", "4.3056", "4.6961"]
+    },
+  },
+})
+
 // Custom call, for models with custom parameters / outputs
 await hf.request({
   model: 'my-custom-model',
@@ -222,7 +238,7 @@ const { generated_text } = await gpt2.textGeneration({inputs: 'The answer to the
 - [x] Question answering
 - [x] Table question answering
 - [x] Text classification
-- [x] Text generation
+- [x] Text generation - [demo](https://huggingface.co/spaces/huggingfacejs/streaming-text-generation)
 - [x] Text2Text generation
 - [x] Token classification
 - [x] Named entity recognition
@@ -243,11 +259,14 @@ const { generated_text } = await gpt2.textGeneration({inputs: 'The answer to the
 - [x] Object detection
 - [x] Image segmentation
 - [x] Text to image
-- [x] Image to text
+- [x] Image to text - [demo](https://huggingface.co/spaces/huggingfacejs/image-to-text)
 
 ### Multimodal
-- [x] Document question answering
-- [x] Visual question answering
+- [x] Document question answering - [demo](https://huggingface.co/spaces/huggingfacejs/doc-vis-qa)
+- [x] Visual question answering - [demo](https://huggingface.co/spaces/huggingfacejs/doc-vis-qa)
+
+### Tabular
+- [x] Tabular regression
 
 ## Tree-shaking
 

package/dist/index.js

@@ -38,10 +38,12 @@ __export(src_exports, {
   streamingRequest: () => streamingRequest,
   summarization: () => summarization,
   tableQuestionAnswering: () => tableQuestionAnswering,
+  tabularRegression: () => tabularRegression,
   textClassification: () => textClassification,
   textGeneration: () => textGeneration,
   textGenerationStream: () => textGenerationStream,
   textToImage: () => textToImage,
+  textToSpeech: () => textToSpeech,
   tokenClassification: () => tokenClassification,
   translation: () => translation,
   visualQuestionAnswering: () => visualQuestionAnswering,
@@ -68,10 +70,12 @@ __export(tasks_exports, {
   streamingRequest: () => streamingRequest,
   summarization: () => summarization,
   tableQuestionAnswering: () => tableQuestionAnswering,
+  tabularRegression: () => tabularRegression,
   textClassification: () => textClassification,
   textGeneration: () => textGeneration,
   textGenerationStream: () => textGenerationStream,
   textToImage: () => textToImage,
+  textToSpeech: () => textToSpeech,
   tokenClassification: () => tokenClassification,
   translation: () => translation,
   visualQuestionAnswering: () => visualQuestionAnswering,
@@ -286,7 +290,11 @@ async function* streamingRequest(args, options) {
       onChunk(value);
       for (const event of events) {
         if (event.data.length > 0) {
-          yield JSON.parse(event.data);
+          const data = JSON.parse(event.data);
+          if (typeof data === "object" && data !== null && "error" in data) {
+            throw new Error(data.error);
+          }
+          yield data;
         }
       }
       events = [];
@@ -326,6 +334,16 @@ async function automaticSpeechRecognition(args, options) {
   return res;
 }
 
+// src/tasks/audio/textToSpeech.ts
+async function textToSpeech(args, options) {
+  const res = await request(args, options);
+  const isValidOutput = res && res instanceof Blob;
+  if (!isValidOutput) {
+    throw new InferenceOutputError("Expected Blob");
+  }
+  return res;
+}
+
 // src/tasks/cv/imageClassification.ts
 async function imageClassification(args, options) {
   const res = await request(args, options);
@@ -543,7 +561,7 @@ async function zeroShotClassification(args, options) {
   return res;
 }
 
-// src/utils/base64FromBytes.ts
+// ../shared/src/base64FromBytes.ts
 function base64FromBytes(arr) {
   if (globalThis.Buffer) {
     return globalThis.Buffer.from(arr).toString("base64");
@@ -556,6 +574,10 @@ function base64FromBytes(arr) {
   }
 }
 
+// ../shared/src/isBackend.ts
+var isBrowser = typeof window !== "undefined" && typeof window.document !== "undefined";
+var isWebWorker = typeof self === "object" && self.constructor && self.constructor.name === "DedicatedWorkerGlobalScope";
+
 // src/tasks/multimodal/documentQuestionAnswering.ts
 async function documentQuestionAnswering(args, options) {
   const reqArgs = {
@@ -594,6 +616,16 @@ async function visualQuestionAnswering(args, options) {
   return res;
 }
 
+// src/tasks/tabular/tabularRegression.ts
+async function tabularRegression(args, options) {
+  const res = await request(args, options);
+  const isValidOutput = Array.isArray(res) && res.every((x) => typeof x === "number");
+  if (!isValidOutput) {
+    throw new InferenceOutputError("Expected number[]");
+  }
+  return res;
+}
+
 // src/HfInference.ts
 var HfInference = class {
   accessToken;
@@ -653,10 +685,12 @@ var HfInferenceEndpoint = class {
   streamingRequest,
   summarization,
   tableQuestionAnswering,
+  tabularRegression,
   textClassification,
   textGeneration,
   textGenerationStream,
   textToImage,
+  textToSpeech,
   tokenClassification,
   translation,
   visualQuestionAnswering,

package/dist/index.mjs

@@ -23,10 +23,12 @@ __export(tasks_exports, {
   streamingRequest: () => streamingRequest,
   summarization: () => summarization,
   tableQuestionAnswering: () => tableQuestionAnswering,
+  tabularRegression: () => tabularRegression,
   textClassification: () => textClassification,
   textGeneration: () => textGeneration,
   textGenerationStream: () => textGenerationStream,
   textToImage: () => textToImage,
+  textToSpeech: () => textToSpeech,
   tokenClassification: () => tokenClassification,
   translation: () => translation,
   visualQuestionAnswering: () => visualQuestionAnswering,
@@ -241,7 +243,11 @@ async function* streamingRequest(args, options) {
       onChunk(value);
       for (const event of events) {
         if (event.data.length > 0) {
-          yield JSON.parse(event.data);
+          const data = JSON.parse(event.data);
+          if (typeof data === "object" && data !== null && "error" in data) {
+            throw new Error(data.error);
+          }
+          yield data;
         }
       }
       events = [];
@@ -281,6 +287,16 @@ async function automaticSpeechRecognition(args, options) {
   return res;
 }
 
+// src/tasks/audio/textToSpeech.ts
+async function textToSpeech(args, options) {
+  const res = await request(args, options);
+  const isValidOutput = res && res instanceof Blob;
+  if (!isValidOutput) {
+    throw new InferenceOutputError("Expected Blob");
+  }
+  return res;
+}
+
 // src/tasks/cv/imageClassification.ts
 async function imageClassification(args, options) {
   const res = await request(args, options);
@@ -498,7 +514,7 @@ async function zeroShotClassification(args, options) {
   return res;
 }
 
-// src/utils/base64FromBytes.ts
+// ../shared/src/base64FromBytes.ts
 function base64FromBytes(arr) {
   if (globalThis.Buffer) {
     return globalThis.Buffer.from(arr).toString("base64");
@@ -511,6 +527,10 @@ function base64FromBytes(arr) {
   }
 }
 
+// ../shared/src/isBackend.ts
+var isBrowser = typeof window !== "undefined" && typeof window.document !== "undefined";
+var isWebWorker = typeof self === "object" && self.constructor && self.constructor.name === "DedicatedWorkerGlobalScope";
+
 // src/tasks/multimodal/documentQuestionAnswering.ts
 async function documentQuestionAnswering(args, options) {
   const reqArgs = {
@@ -549,6 +569,16 @@ async function visualQuestionAnswering(args, options) {
   return res;
 }
 
+// src/tasks/tabular/tabularRegression.ts
+async function tabularRegression(args, options) {
+  const res = await request(args, options);
+  const isValidOutput = Array.isArray(res) && res.every((x) => typeof x === "number");
+  if (!isValidOutput) {
+    throw new InferenceOutputError("Expected number[]");
+  }
+  return res;
+}
+
 // src/HfInference.ts
 var HfInference = class {
   accessToken;
@@ -607,10 +637,12 @@ export {
   streamingRequest,
   summarization,
   tableQuestionAnswering,
+  tabularRegression,
   textClassification,
   textGeneration,
   textGenerationStream,
   textToImage,
+  textToSpeech,
   tokenClassification,
   translation,
   visualQuestionAnswering,

package/node_modules/@huggingface/shared/package.json

@@ -0,0 +1,8 @@
+{
+  "name": "@huggingface/shared",
+  "version": "1.0.0",
+  "private": true,
+  "main": "src/index.ts",
+  "source": "src/index.ts",
+  "types": "./src/index.ts"
+}

package/node_modules/@huggingface/shared/src/index.ts

@@ -0,0 +1,3 @@
+export * from './base64FromBytes';
+export * from './isBackend';
+export * from './isFrontend';

package/node_modules/@huggingface/shared/tsconfig.json

@@ -0,0 +1,14 @@
+{
+	"compilerOptions": {
+		"esModuleInterop": true,
+		"noEmit": false,
+		"module": "ESNext",
+		"target": "ESNext",
+		"moduleResolution": "Node",
+		"noImplicitAny": true,
+		"strict": true,
+		"strictNullChecks": true,
+		"skipLibCheck": true,
+        "composite": true
+	}
+}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@huggingface/inference",
-  "version": "2.1.3",
+  "version": "2.2.1",
   "license": "MIT",
   "author": "Tim Mikeladze <tim.mikeladze@gmail.com>",
   "description": "Typescript wrapper for the Hugging Face Inference API",
@@ -40,8 +40,12 @@
     "@types/node": "18.13.0",
     "typescript": "^5.0.4",
     "vite": "^4.1.4",
-    "vitest": "^0.29.8"
+    "vitest": "^0.29.8",
+    "@huggingface/shared": "1.0.0"
   },
+  "bundledDependencies": [
+    "@huggingface/shared"
+  ],
   "resolutions": {},
   "scripts": {
     "build": "tsup src/index.ts --format cjs,esm --clean",
@@ -49,6 +53,7 @@
     "lint:check": "eslint --ext .cjs,.ts .",
     "format": "prettier --write .",
     "format:check": "prettier --check .",
+    "preversion": "pnpm --filter doc-internal run fix-cdn-versions && git add ../../README.md",
     "test": "vitest run --config vitest.config.ts",
     "test:browser": "vitest run --browser.name=chrome --browser.headless --config vitest.config.ts",
     "type-check": "tsc"

package/src/tasks/audio/textToSpeech.ts

@@ -0,0 +1,25 @@
+import { InferenceOutputError } from "../../lib/InferenceOutputError";
+import type { BaseArgs, Options } from "../../types";
+import { request } from "../custom/request";
+
+export type TextToSpeechArgs = BaseArgs & {
+	/**
+	 * The text to generate an audio from
+	 */
+	inputs: string;
+};
+
+export type TextToSpeechOutput = Blob;
+
+/**
+ * This task synthesize an audio of a voice pronouncing a given text.
+ * Recommended model: espnet/kan-bayashi_ljspeech_vits
+ */
+export async function textToSpeech(args: TextToSpeechArgs, options?: Options): Promise<TextToSpeechOutput> {
+	const res = await request<TextToSpeechOutput>(args, options);
+	const isValidOutput = res && res instanceof Blob;
+	if (!isValidOutput) {
+		throw new InferenceOutputError("Expected Blob");
+	}
+	return res;
+}

package/src/tasks/custom/streamingRequest.ts

@@ -65,7 +65,11 @@ export async function* streamingRequest<T>(
 			onChunk(value);
 			for (const event of events) {
 				if (event.data.length > 0) {
-					yield JSON.parse(event.data) as T;
+					const data = JSON.parse(event.data);
+					if (typeof data === "object" && data !== null && "error" in data) {
+						throw new Error(data.error);
+					}
+					yield data as T;
 				}
 			}
 			events = [];

package/src/tasks/index.ts

@@ -5,8 +5,9 @@ export * from "./custom/streamingRequest";
 // Audio tasks
 export * from "./audio/audioClassification";
 export * from "./audio/automaticSpeechRecognition";
+export * from "./audio/textToSpeech";
 
-// Commputer Vision tasks
+// Computer Vision tasks
 export * from "./cv/imageClassification";
 export * from "./cv/imageSegmentation";
 export * from "./cv/imageToText";
@@ -31,3 +32,6 @@ export * from "./nlp/zeroShotClassification";
 // Multimodal tasks
 export * from "./multimodal/documentQuestionAnswering";
 export * from "./multimodal/visualQuestionAnswering";
+
+// Tabular tasks
+export * from "./tabular/tabularRegression";

package/src/tasks/multimodal/documentQuestionAnswering.ts

@@ -2,7 +2,7 @@ import { InferenceOutputError } from "../../lib/InferenceOutputError";
 import type { BaseArgs, Options } from "../../types";
 import { request } from "../custom/request";
 import type { RequestArgs } from "../../types";
-import { base64FromBytes } from "../../utils/base64FromBytes";
+import { base64FromBytes } from "@huggingface/shared";
 import { toArray } from "../../utils/toArray";
 
 export type DocumentQuestionAnsweringArgs = BaseArgs & {

package/src/tasks/multimodal/visualQuestionAnswering.ts

@@ -1,7 +1,7 @@
 import { InferenceOutputError } from "../../lib/InferenceOutputError";
 import type { BaseArgs, Options, RequestArgs } from "../../types";
 import { request } from "../custom/request";
-import { base64FromBytes } from "../../utils/base64FromBytes";
+import { base64FromBytes } from "@huggingface/shared";
 
 export type VisualQuestionAnsweringArgs = BaseArgs & {
 	inputs: {

package/src/tasks/nlp/textGeneration.ts

@@ -44,6 +44,10 @@ export type TextGenerationArgs = BaseArgs & {
 		 * (Default: None). Float to define the tokens that are within the sample operation of text generation. Add tokens in the sample for more probable to least probable until the sum of the probabilities is greater than top_p.
 		 */
 		top_p?: number;
+		/**
+		 * (Default: None). Integer. The maximum number of tokens from the input.
+		 */
+		truncate?: number;
 	};
 };
 

package/src/tasks/tabular/tabularRegression.ts

@@ -0,0 +1,34 @@
+import { InferenceOutputError } from "../../lib/InferenceOutputError";
+import type { BaseArgs, Options } from "../../types";
+import { request } from "../custom/request";
+
+export type TabularRegressionArgs = BaseArgs & {
+	inputs: {
+		/**
+		 * A table of data represented as a dict of list where entries are headers and the lists are all the values, all lists must have the same size.
+		 */
+		data: Record<string, string[]>;
+	};
+};
+
+/**
+ * a list of predicted values for each row
+ */
+export type TabularRegressionOutput = number[];
+
+/**
+ * Predicts target value for a given set of features in tabular form.
+ * Typically, you will want to train a regression model on your training data and use it with your new data of the same format.
+ * Example model: scikit-learn/Fish-Weight
+ */
+export async function tabularRegression(
+	args: TabularRegressionArgs,
+	options?: Options
+): Promise<TabularRegressionOutput> {
+	const res = await request<TabularRegressionOutput>(args, options);
+	const isValidOutput = Array.isArray(res) && res.every((x) => typeof x === "number");
+	if (!isValidOutput) {
+		throw new InferenceOutputError("Expected number[]");
+	}
+	return res;
+}