@huggingface/tasks - npm Package Compare versions

Comparing version 0.3.1 to 0.3.2

dist/index.d.ts

@@ -107,11 +107,2 @@ declare const MODALITIES: readonly ["cv", "nlp", "audio", "tabular", "multimodal", "rl", "other"];
     };
-    conversational: {
-        name: string;
-        subtasks: {
-            type: string;
-            name: string;
-        }[];
-        modality: "nlp";
-        color: "green";
-    };
     "feature-extraction": {
@@ -396,2 +387,7 @@ name: string;
     };
+    "image-feature-extraction": {
+        name: string;
+        modality: "cv";
+        color: "indigo";
+    };
     other: {
@@ -406,5 +402,6 @@ name: string;
 type PipelineType = keyof typeof PIPELINE_DATA;
-declare const PIPELINE_TYPES: ("other" | "text-classification" | "token-classification" | "table-question-answering" | "question-answering" | "zero-shot-classification" | "translation" | "summarization" | "conversational" | "feature-extraction" | "text-generation" | "text2text-generation" | "fill-mask" | "sentence-similarity" | "text-to-speech" | "text-to-audio" | "automatic-speech-recognition" | "audio-to-audio" | "audio-classification" | "voice-activity-detection" | "depth-estimation" | "image-classification" | "object-detection" | "image-segmentation" | "text-to-image" | "image-to-text" | "image-to-image" | "image-to-video" | "unconditional-image-generation" | "video-classification" | "reinforcement-learning" | "robotics" | "tabular-classification" | "tabular-regression" | "tabular-to-text" | "table-to-text" | "multiple-choice" | "text-retrieval" | "time-series-forecasting" | "text-to-video" | "image-text-to-text" | "visual-question-answering" | "document-question-answering" | "zero-shot-image-classification" | "graph-ml" | "mask-generation" | "zero-shot-object-detection" | "text-to-3d" | "image-to-3d")[];
+type WidgetType = PipelineType | "conversational";
+declare const PIPELINE_TYPES: ("other" | "text-classification" | "token-classification" | "table-question-answering" | "question-answering" | "zero-shot-classification" | "translation" | "summarization" | "feature-extraction" | "text-generation" | "text2text-generation" | "fill-mask" | "sentence-similarity" | "text-to-speech" | "text-to-audio" | "automatic-speech-recognition" | "audio-to-audio" | "audio-classification" | "voice-activity-detection" | "depth-estimation" | "image-classification" | "object-detection" | "image-segmentation" | "text-to-image" | "image-to-text" | "image-to-image" | "image-to-video" | "unconditional-image-generation" | "video-classification" | "reinforcement-learning" | "robotics" | "tabular-classification" | "tabular-regression" | "tabular-to-text" | "table-to-text" | "multiple-choice" | "text-retrieval" | "time-series-forecasting" | "text-to-video" | "image-text-to-text" | "visual-question-answering" | "document-question-answering" | "zero-shot-image-classification" | "graph-ml" | "mask-generation" | "zero-shot-object-detection" | "text-to-3d" | "image-to-3d" | "image-feature-extraction")[];
 declare const SUBTASK_TYPES: string[];
-declare const PIPELINE_TYPES_SET: Set<"other" | "text-classification" | "token-classification" | "table-question-answering" | "question-answering" | "zero-shot-classification" | "translation" | "summarization" | "conversational" | "feature-extraction" | "text-generation" | "text2text-generation" | "fill-mask" | "sentence-similarity" | "text-to-speech" | "text-to-audio" | "automatic-speech-recognition" | "audio-to-audio" | "audio-classification" | "voice-activity-detection" | "depth-estimation" | "image-classification" | "object-detection" | "image-segmentation" | "text-to-image" | "image-to-text" | "image-to-image" | "image-to-video" | "unconditional-image-generation" | "video-classification" | "reinforcement-learning" | "robotics" | "tabular-classification" | "tabular-regression" | "tabular-to-text" | "table-to-text" | "multiple-choice" | "text-retrieval" | "time-series-forecasting" | "text-to-video" | "image-text-to-text" | "visual-question-answering" | "document-question-answering" | "zero-shot-image-classification" | "graph-ml" | "mask-generation" | "zero-shot-object-detection" | "text-to-3d" | "image-to-3d">;
+declare const PIPELINE_TYPES_SET: Set<"other" | "text-classification" | "token-classification" | "table-question-answering" | "question-answering" | "zero-shot-classification" | "translation" | "summarization" | "feature-extraction" | "text-generation" | "text2text-generation" | "fill-mask" | "sentence-similarity" | "text-to-speech" | "text-to-audio" | "automatic-speech-recognition" | "audio-to-audio" | "audio-classification" | "voice-activity-detection" | "depth-estimation" | "image-classification" | "object-detection" | "image-segmentation" | "text-to-image" | "image-to-text" | "image-to-image" | "image-to-video" | "unconditional-image-generation" | "video-classification" | "reinforcement-learning" | "robotics" | "tabular-classification" | "tabular-regression" | "tabular-to-text" | "table-to-text" | "multiple-choice" | "text-retrieval" | "time-series-forecasting" | "text-to-video" | "image-text-to-text" | "visual-question-answering" | "document-question-answering" | "zero-shot-image-classification" | "graph-ml" | "mask-generation" | "zero-shot-object-detection" | "text-to-3d" | "image-to-3d" | "image-feature-extraction">;
 
@@ -489,2 +486,17 @@ /**
 
+declare const SPECIAL_TOKENS_ATTRIBUTES: readonly ["bos_token", "eos_token", "unk_token", "sep_token", "pad_token", "cls_token", "mask_token"];
+/**
+ * Public interface for a tokenizer's special tokens mapping
+ */
+type SpecialTokensMap = {
+    [key in (typeof SPECIAL_TOKENS_ATTRIBUTES)[number]]?: string;
+};
+/**
+ * Public interface for tokenizer config
+ */
+interface TokenizerConfig extends SpecialTokensMap {
+    use_default_system_prompt?: boolean;
+    chat_template?: string;
+}
+
 declare enum InferenceDisplayability {
@@ -544,2 +556,3 @@ /**
         };
+        tokenizer?: TokenizerConfig;
     };
@@ -1058,3 +1071,3 @@ /**
 
-type PerLanguageMapping = Map<PipelineType, string[] | WidgetExample[]>;
+type PerLanguageMapping = Map<WidgetType, string[] | WidgetExample[]>;
 declare const MAPPING_DEFAULT_WIDGET: Map<string, PerLanguageMapping>;
@@ -1227,2 +1240,2 @@
 
-export { ALL_DISPLAY_MODEL_LIBRARY_KEYS, ALL_MODEL_LIBRARY_KEYS, ExampleRepo, InferenceDisplayability, LIBRARY_TASK_MAPPING_EXCLUDING_TRANSFORMERS, LibraryUiElement, MAPPING_DEFAULT_WIDGET, MODALITIES, MODALITY_LABELS, MODEL_LIBRARIES_UI_ELEMENTS, Modality, ModelData, ModelLibraryKey, PIPELINE_DATA, PIPELINE_TYPES, PIPELINE_TYPES_SET, PipelineData, PipelineType, SUBTASK_TYPES, TASKS_DATA, TASKS_MODEL_LIBRARIES, TaskData, TaskDataCustom, TaskDemo, TaskDemoEntry, TransformersInfo, WidgetExample, WidgetExampleAssetAndPromptInput, WidgetExampleAssetAndTextInput, WidgetExampleAssetAndZeroShotInput, WidgetExampleAssetInput, WidgetExampleAttribute, WidgetExampleOutput, WidgetExampleOutputAnswerScore, WidgetExampleOutputLabels, WidgetExampleOutputText, WidgetExampleOutputUrl, WidgetExampleSentenceSimilarityInput, WidgetExampleStructuredDataInput, WidgetExampleTableDataInput, WidgetExampleTextAndContextInput, WidgetExampleTextAndTableInput, WidgetExampleTextInput, WidgetExampleZeroShotTextInput, index as snippets };
+export { ALL_DISPLAY_MODEL_LIBRARY_KEYS, ALL_MODEL_LIBRARY_KEYS, ExampleRepo, InferenceDisplayability, LIBRARY_TASK_MAPPING_EXCLUDING_TRANSFORMERS, LibraryUiElement, MAPPING_DEFAULT_WIDGET, MODALITIES, MODALITY_LABELS, MODEL_LIBRARIES_UI_ELEMENTS, Modality, ModelData, ModelLibraryKey, PIPELINE_DATA, PIPELINE_TYPES, PIPELINE_TYPES_SET, PipelineData, PipelineType, SPECIAL_TOKENS_ATTRIBUTES, SUBTASK_TYPES, SpecialTokensMap, TASKS_DATA, TASKS_MODEL_LIBRARIES, TaskData, TaskDataCustom, TaskDemo, TaskDemoEntry, TokenizerConfig, TransformersInfo, WidgetExample, WidgetExampleAssetAndPromptInput, WidgetExampleAssetAndTextInput, WidgetExampleAssetAndZeroShotInput, WidgetExampleAssetInput, WidgetExampleAttribute, WidgetExampleOutput, WidgetExampleOutputAnswerScore, WidgetExampleOutputLabels, WidgetExampleOutputText, WidgetExampleOutputUrl, WidgetExampleSentenceSimilarityInput, WidgetExampleStructuredDataInput, WidgetExampleTableDataInput, WidgetExampleTextAndContextInput, WidgetExampleTextAndTableInput, WidgetExampleTextInput, WidgetExampleZeroShotTextInput, WidgetType, index as snippets };

package.json

 {
   "name": "@huggingface/tasks",
   "packageManager": "pnpm@8.10.5",
-  "version": "0.3.1",
+  "version": "0.3.2",
   "description": "List of ML tasks for huggingface.co/tasks",
@@ -6,0 +6,0 @@ "repository": "https://github.com/huggingface/huggingface.js.git",

src/default-widget-inputs.ts

 import type { WidgetExample } from "./widget-example";
-import type { PipelineType } from "./pipelines";
+import type { WidgetType } from "./pipelines";
 
 type LanguageCode = string;
 
-type PerLanguageMapping = Map<PipelineType, string[] | WidgetExample[]>;
+type PerLanguageMapping = Map<WidgetType, string[] | WidgetExample[]>;
 
@@ -8,0 +8,0 @@ /// NOTE TO CONTRIBUTORS:

src/index.ts

@@ -8,2 +8,3 @@ export { LIBRARY_TASK_MAPPING_EXCLUDING_TRANSFORMERS } from "./library-to-tasks";
 	PIPELINE_TYPES,
+	type WidgetType,
 	type PipelineType,
@@ -20,2 +21,3 @@ type PipelineData,
 export type { ModelData, TransformersInfo } from "./model-data";
+export type { SpecialTokensMap, TokenizerConfig } from "./tokenizer-data";
 export type {
@@ -42,4 +44,5 @@ WidgetExample,
 export { InferenceDisplayability } from "./model-data";
+export { SPECIAL_TOKENS_ATTRIBUTES } from "./tokenizer-data";
 
 import * as snippets from "./snippets";
 export { snippets };

src/library-to-tasks.ts

@@ -30,3 +30,3 @@ import type { ModelLibraryKey } from "./model-libraries";
 	open_clip: ["zero-shot-classification", "zero-shot-image-classification"],
-	paddlenlp: ["conversational", "fill-mask", "summarization", "zero-shot-classification"],
+	paddlenlp: ["fill-mask", "summarization", "zero-shot-classification"],
 	peft: ["text-generation"],
@@ -33,0 +33,0 @@ "pyannote-audio": ["automatic-speech-recognition"],

src/model-data.ts

 import type { PipelineType } from "./pipelines";
 import type { WidgetExample } from "./widget-example";
+import type { TokenizerConfig } from "./tokenizer-data";
 
@@ -56,2 +57,3 @@ export enum InferenceDisplayability {
 		};
+		tokenizer?: TokenizerConfig;
 	};
@@ -58,0 +60,0 @@ /**

src/pipelines.ts

@@ -228,13 +228,2 @@ export const MODALITIES = ["cv", "nlp", "audio", "tabular", "multimodal", "rl", "other"] as const;
 	},
-	conversational: {
-		name: "Conversational",
-		subtasks: [
-			{
-				type: "dialogue-generation",
-				name: "Dialogue Generation",
-			},
-		],
-		modality: "nlp",
-		color: "green",
-	},
 	"feature-extraction": {
@@ -253,2 +242,10 @@ name: "Feature Extraction",
 			{
+				type: "dialogue-generation",
+				name: "Dialogue Generation",
+			},
+			{
+				type: "conversational",
+				name: "Conversational",
+			},
+			{
 				type: "language-modeling",
@@ -598,3 +595,3 @@ name: "Language Modeling",
 	"image-text-to-text": {
-		name: "Image + Text to Image (VLLMs)",
+		name: "Image + Text to Text (VLLMs)",
 		modality: "multimodal",
@@ -657,2 +654,7 @@ color: "red",
 	},
+	"image-feature-extraction": {
+		name: "Image Feature Extraction",
+		modality: "cv",
+		color: "indigo",
+	},
 	other: {
@@ -669,2 +671,4 @@ name: "Other",
 
+export type WidgetType = PipelineType | "conversational";
+
 export const PIPELINE_TYPES = Object.keys(PIPELINE_DATA) as PipelineType[];
@@ -671,0 +675,0 @@

src/snippets/curl.ts

@@ -37,3 +37,2 @@ import type { ModelData } from "../model-data.js";
 	summarization: snippetBasic,
-	conversational: snippetBasic,
 	"feature-extraction": snippetBasic,
@@ -40,0 +39,0 @@ "text-generation": snippetBasic,

src/snippets/inputs.ts

@@ -12,9 +12,2 @@ import type { ModelData } from "../model-data";
 
-const inputsConversational = () =>
-	`{
-		"past_user_inputs": ["Which movie is the best ?"],
-		"generated_responses": ["It is Die Hard for sure."],
-		"text": "Can you explain why ?"
-	}`;
-
 const inputsTableQuestionAnswering = () =>
@@ -100,3 +93,2 @@ `{
 	"automatic-speech-recognition": inputsAutomaticSpeechRecognition,
-	conversational: inputsConversational,
 	"document-question-answering": inputsVisualQuestionAnswering,
@@ -103,0 +95,0 @@ "feature-extraction": inputsFeatureExtraction,

src/snippets/js.ts

@@ -124,3 +124,2 @@ import type { ModelData } from "../model-data.js";
 	summarization: snippetBasic,
-	conversational: snippetBasic,
 	"feature-extraction": snippetBasic,
@@ -127,0 +126,0 @@ "text-generation": snippetBasic,

src/snippets/python.ts

@@ -119,3 +119,2 @@ import type { ModelData } from "../model-data.js";
 	summarization: snippetBasic,
-	conversational: snippetBasic,
 	"feature-extraction": snippetBasic,
@@ -122,0 +121,0 @@ "text-generation": snippetBasic,

src/tasks/index.ts

@@ -1,2 +0,3 @@
-import { type PipelineType, PIPELINE_DATA } from "../pipelines";
+import type { PipelineType } from "../pipelines";
+import { PIPELINE_DATA } from "../pipelines";
 
@@ -6,3 +7,2 @@ import audioClassification from "./audio-classification/data";
 import automaticSpeechRecognition from "./automatic-speech-recognition/data";
-import conversational from "./conversational/data";
 import documentQuestionAnswering from "./document-question-answering/data";
@@ -49,3 +49,2 @@ import featureExtraction from "./feature-extraction/data";
 	"automatic-speech-recognition": ["espnet", "nemo", "speechbrain", "transformers", "transformers.js"],
-	conversational: ["transformers"],
 	"depth-estimation": ["transformers", "transformers.js"],
@@ -57,2 +56,3 @@ "document-question-answering": ["transformers", "transformers.js"],
 	"image-classification": ["keras", "timm", "transformers", "transformers.js"],
+	"image-feature-extraction": ["timm", "transformers"],
 	"image-segmentation": ["transformers", "transformers.js"],
@@ -129,3 +129,2 @@ "image-text-to-text": ["transformers"],
 	"automatic-speech-recognition": getData("automatic-speech-recognition", automaticSpeechRecognition),
-	conversational: getData("conversational", conversational),
 	"depth-estimation": getData("depth-estimation", depthEstimation),
@@ -176,2 +175,3 @@ "document-question-answering": getData("document-question-answering", documentQuestionAnswering),
 	"image-to-3d": getData("image-to-3d", placeholder),
+	"image-feature-extraction": getData("image-feature-extraction", placeholder),
 } as const;
@@ -178,0 +178,0 @@

No alert changes

Worsened metrics

Total package byte prevSize

914270

decreased by-0.82%

Number of package files

178

decreased by-0.56%

Lines of code

21873

decreased by-0.64%

No dependency changes