✨ Generate inference types for existing tasks

packages/tasks/src/tasks/audio-classification/inference.ts

@@ -1,7 +1,7 @@
 /**
  * Inference code generated from the JSON schema spec in ./spec
  *
- * Generated on 2024-01-19T14:59:10.562Z
+ * Generated on 2024-01-19T16:16:01.752Z
  */
 
 /**

packages/tasks/src/tasks/automatic-speech-recognition/inference.ts

@@ -0,0 +1,31 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Generated on 2024-01-19T16:16:01.752Z
+ */
+
+/**
+ * Inputs for Automatic Speech Recognition inference
+ */
+export interface AutomaticSpeechRecognitionInput {
+	/**
+	 * The input audio data
+	 */
+	inputs: any;
+	/**
+	 * Additional inference parameters
+	 */
+	parameters?: { [key: string]: any };
+	[property: string]: any;
+}
+
+/**
+ * Outputs of inference for the Automatic Speech Recognition task
+ */
+export interface AutomaticSpeechRecognitionOutput {
+	/**
+	 * The recognized text.
+	 */
+	text: string;
+	[property: string]: any;
+}

packages/tasks/src/tasks/depth-estimation/inference.ts

@@ -0,0 +1,33 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Generated on 2024-01-19T16:16:01.752Z
+ */
+
+/**
+ * Inputs for Depth Estimation inference
+ */
+export interface DepthEstimationInput {
+	/**
+	 * The input image data
+	 */
+	inputs: any;
+	/**
+	 * Additional inference parameters
+	 */
+	parameters?: DepthEstimationParameters;
+	[property: string]: any;
+}
+
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Depth Estimation
+ */
+export interface DepthEstimationParameters {
+	/**
+	 * When specified, limits the output to the top K most probable classes.
+	 */
+	topK?: number;
+	[property: string]: any;
+}

packages/tasks/src/tasks/document-question-answering/inference.ts

@@ -0,0 +1,101 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Generated on 2024-01-19T16:16:01.752Z
+ */
+
+/**
+ * Inputs for Document Question Answering inference
+ */
+export interface DocumentQuestionAnsweringInput {
+	/**
+	 * The
+	 */
+	inputs: DocumentAndQuestion[] | DocumentAndQuestion;
+	/**
+	 * Additional inference parameters
+	 */
+	parameters?: DocumentQuestionAnsweringParameters;
+	[property: string]: any;
+}
+
+export interface DocumentAndQuestion {
+	/**
+	 * The image on which the question is asked
+	 */
+	image?: any;
+	/**
+	 * A question to ask of the document
+	 */
+	question?: string;
+	[property: string]: any;
+}
+
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Document Question Answering
+ */
+export interface DocumentQuestionAnsweringParameters {
+	/**
+	 * If the words in the document are too long to fit with the question for the model, it will
+	 * be split in several chunks with some overlap. This argument controls the size of that
+	 * overlap.
+	 */
+	docStride?: number;
+	/**
+	 * Whether to accept impossible as an answer
+	 */
+	handleImpossibleAnswer?: boolean;
+	/**
+	 * Language to use while running OCR. Defaults to english.
+	 */
+	lang?: string;
+	/**
+	 * The maximum length of predicted answers (e.g., only answers with a shorter length are
+	 * considered).
+	 */
+	maxAnswerLen?: number;
+	/**
+	 * The maximum length of the question after tokenization. It will be truncated if needed.
+	 */
+	maxQuestionLen?: number;
+	/**
+	 * The maximum length of the total sentence (context + question) in tokens of each chunk
+	 * passed to the model. The context will be split in several chunks (using doc_stride as
+	 * overlap) if needed.
+	 */
+	maxSeqLen?: number;
+	/**
+	 * The number of answers to return (will be chosen by order of likelihood). Can return less
+	 * than top_k answers if there are not enough options available within the context.
+	 */
+	topK?: number;
+	/**
+	 * A list of words and bounding boxes (normalized 0->1000). If provided, the inference will
+	 * skip the OCR step and use the provided bounding boxes instead.
+	 */
+	wordBoxes?: Array<number[] | string>;
+	[property: string]: any;
+}
+
+/**
+ * Outputs of inference for the Document Question Answering task
+ */
+export interface DocumentQuestionAnsweringOutput {
+	/**
+	 * The answer to the question.
+	 */
+	answer: string;
+	end: number;
+	/**
+	 * The probability associated to the answer.
+	 */
+	score: number;
+	start: number;
+	/**
+	 * The index of each word/box pair that is in the answer
+	 */
+	words: number[];
+	[property: string]: any;
+}

packages/tasks/src/tasks/feature-extraction/inference.ts

@@ -0,0 +1,20 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Generated on 2024-01-19T16:16:01.752Z
+ */
+
+/**
+ * Inputs for Feature Extraction inference
+ */
+export interface FeatureExtractionInput {
+	/**
+	 * One or several texts to get the features of
+	 */
+	inputs: string[] | string;
+	/**
+	 * Additional inference parameters
+	 */
+	parameters?: { [key: string]: any };
+	[property: string]: any;
+}

packages/tasks/src/tasks/fill-mask/inference.ts

@@ -0,0 +1,63 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Generated on 2024-01-19T16:16:01.752Z
+ */
+
+/**
+ * Inputs for Fill Mask inference
+ */
+export interface FillMaskInput {
+	/**
+	 * One or several texts with masked tokens
+	 */
+	inputs: string[] | string;
+	/**
+	 * Additional inference parameters
+	 */
+	parameters?: FillMaskParameters;
+	[property: string]: any;
+}
+
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Fill Mask
+ */
+export interface FillMaskParameters {
+	/**
+	 * When passed, the model will limit the scores to the passed targets instead of looking up
+	 * in the whole vocabulary. If the provided targets are not in the model vocab, they will be
+	 * tokenized and the first resulting token will be used (with a warning, and that might be
+	 * slower).
+	 */
+	targets?: string[] | string;
+	/**
+	 * When passed, overrides the number of predictions to return.
+	 */
+	topK?: number;
+	[property: string]: any;
+}
+
+/**
+ * Outputs of inference for the Fill Mask task
+ */
+export interface FillMaskOutput {
+	/**
+	 * The corresponding probability
+	 */
+	score: number;
+	/**
+	 * The corresponding input with the mask token prediction.
+	 */
+	sequence: string;
+	/**
+	 * The predicted token id (to replace the masked one).
+	 */
+	token: number;
+	/**
+	 * The predicted token (to replace the masked one).
+	 */
+	tokenStr: string;
+	[property: string]: any;
+}

packages/tasks/src/tasks/image-classification/inference.ts

@@ -0,0 +1,48 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Generated on 2024-01-19T16:16:01.752Z
+ */
+
+/**
+ * Inputs for Image Classification inference
+ */
+export interface ImageClassificationInput {
+	/**
+	 * On or several image files to classify
+	 */
+	inputs: any;
+	/**
+	 * Additional inference parameters
+	 */
+	parameters?: ImageClassificationParameters;
+	[property: string]: any;
+}
+
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Image Classification
+ */
+export interface ImageClassificationParameters {
+	/**
+	 * When specified, limits the output to the top K most probable classes.
+	 */
+	topK?: number;
+	[property: string]: any;
+}
+
+/**
+ * Outputs of inference for the Image Classification task
+ */
+export interface ImageClassificationOutput {
+	/**
+	 * The predicted class label (model specific).
+	 */
+	label: string;
+	/**
+	 * The corresponding probability.
+	 */
+	score: number;
+	[property: string]: any;
+}

packages/tasks/src/tasks/image-segmentation/inference.ts

@@ -0,0 +1,64 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Generated on 2024-01-19T16:16:01.752Z
+ */
+
+/**
+ * Inputs for Image Segmentation inference
+ */
+export interface ImageSegmentationInput {
+	/**
+	 * One or several image files to perform segmentation on
+	 */
+	inputs: any;
+	/**
+	 * Additional inference parameters
+	 */
+	parameters?: ImageSegmentationParameters;
+	[property: string]: any;
+}
+
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Image Segmentation
+ */
+export interface ImageSegmentationParameters {
+	/**
+	 * Threshold to use when turning the predicted masks into binary values.
+	 */
+	maskThreshold?: number;
+	/**
+	 * Mask overlap threshold to eliminate small, disconnected segments.
+	 */
+	overlapMaskAreaThreshold?: number;
+	/**
+	 * Segmentation task to be performed, depending on model capabilities.
+	 */
+	subtask?: Subtask;
+	/**
+	 * Probability threshold to filter out predicted masks.
+	 */
+	threshold?: number;
+	[property: string]: any;
+}
+
+export type Subtask = "instance" | "panoptic" | "semantic";
+
+/**
+ * Outputs of inference for the Image Segmentation task
+ *
+ * A predicted mask / segment
+ */
+export interface ImageSegmentationOutput {
+	/**
+	 * The label of the predicted segment
+	 */
+	label: string;
+	/**
+	 * The corresponding mask as a black-and-white image
+	 */
+	mask: any;
+	[property: string]: any;
+}

packages/tasks/src/tasks/image-to-image/inference.ts

@@ -0,0 +1,20 @@
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Generated on 2024-01-19T16:16:01.752Z
+ */
+
+/**
+ * Inputs for Image To Image inference
+ */
+export interface ImageToImageInput {
+	/**
+	 * One or more images to generate images from
+	 */
+	inputs: any;
+	/**
+	 * Additional inference parameters
+	 */
+	parameters?: any;
+	[property: string]: any;
+}