@huggingface/transformers 3.0.1 → 3.1.0

package/types/models/pyannote/processing_pyannote.d.ts

@@ -0,0 +1,30 @@
+export class PyAnnoteProcessor extends Processor {
+    static feature_extractor_class: typeof AutoFeatureExtractor;
+    /**
+     * Calls the feature_extractor function with the given audio input.
+     * @param {any} audio The audio input to extract features from.
+     * @returns {Promise<any>} A Promise that resolves with the extracted features.
+     */
+    _call(audio: any): Promise<any>;
+    /**
+     * NOTE: Can return fractional values. `Math.ceil` will ensure correct value.
+     * @param {number} samples The number of frames in the audio.
+     * @returns {number} The number of frames in the audio.
+     */
+    samples_to_frames(samples: number): number;
+    /**
+     * Post-processes the speaker diarization logits output by the model.
+     * @param {import('../../utils/tensor.js').Tensor} logits The speaker diarization logits output by the model.
+     * @param {number} num_samples Number of samples in the input audio.
+     * @returns {Array<Array<{ id: number, start: number, end: number, confidence: number }>>} The post-processed speaker diarization results.
+     */
+    post_process_speaker_diarization(logits: import('../../utils/tensor.js').Tensor, num_samples: number): Array<Array<{
+        id: number;
+        start: number;
+        end: number;
+        confidence: number;
+    }>>;
+}
+import { Processor } from '../../base/processing_utils.js';
+import { AutoFeatureExtractor } from '../auto/feature_extraction_auto.js';
+//# sourceMappingURL=processing_pyannote.d.ts.map

package/types/models/pyannote/processing_pyannote.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"processing_pyannote.d.ts","sourceRoot":"","sources":["../../../src/models/pyannote/processing_pyannote.js"],"names":[],"mappings":"AAIA;IACI,4DAAqD;IAErD;;;;OAIG;IACH,aAHW,GAAG,GACD,QAAQ,GAAG,CAAC,CAIxB;IAED;;;;OAIG;IACH,2BAHW,MAAM,GACJ,MAAM,CAIlB;IAED;;;;;OAKG;IACH,yCAJW,OAAO,uBAAuB,EAAE,MAAM,eACtC,MAAM,GACJ,MAAM,MAAM;QAAE,EAAE,EAAE,MAAM,CAAC;QAAC,KAAK,EAAE,MAAM,CAAC;QAAC,GAAG,EAAE,MAAM,CAAC;QAAC,UAAU,EAAE,MAAM,CAAA;KAAE,CAAC,CAAC,CAwCxF;CACJ;0BAtEyB,gCAAgC;qCACrB,oCAAoC"}

package/types/models/qwen2_vl/image_processing_qwen2_vl.d.ts

@@ -0,0 +1,11 @@
+export class Qwen2VLImageProcessor extends ImageProcessor {
+    _call(images: any, ...args: any[]): Promise<{
+        pixel_values: Tensor;
+        image_grid_thw: Tensor;
+        original_sizes: import("../../base/image_processors_utils.js").HeightWidth[];
+        reshaped_input_sizes: import("../../base/image_processors_utils.js").HeightWidth[];
+    }>;
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+import { Tensor } from "../../utils/tensor.js";
+//# sourceMappingURL=image_processing_qwen2_vl.d.ts.map

package/types/models/qwen2_vl/image_processing_qwen2_vl.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_qwen2_vl.d.ts","sourceRoot":"","sources":["../../../src/models/qwen2_vl/image_processing_qwen2_vl.js"],"names":[],"mappings":"AAKA;IACI;;;;;OA2CC;CACJ;+BAhDM,sCAAsC;uBACjB,uBAAuB"}

package/types/models/qwen2_vl/processing_qwen2_vl.d.ts

@@ -0,0 +1,17 @@
+export class Qwen2VLProcessor extends Processor {
+    static image_processor_class: typeof AutoImageProcessor;
+    static tokenizer_class: typeof AutoTokenizer;
+    /**
+     *
+     * @param {string|string[]} text
+     * @param {RawImage|RawImage[]} images
+     * @param  {...any} args
+     * @returns {Promise<any>}
+     */
+    _call(text: string | string[], images?: RawImage | RawImage[], ...args: any[]): Promise<any>;
+}
+import { Processor } from "../../base/processing_utils.js";
+import { RawImage } from "../../utils/image.js";
+import { AutoImageProcessor } from "../auto/image_processing_auto.js";
+import { AutoTokenizer } from "../../tokenizers.js";
+//# sourceMappingURL=processing_qwen2_vl.d.ts.map

package/types/models/qwen2_vl/processing_qwen2_vl.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"processing_qwen2_vl.d.ts","sourceRoot":"","sources":["../../../src/models/qwen2_vl/processing_qwen2_vl.js"],"names":[],"mappings":"AAKA;IACI,wDAAiD;IACjD,6CAAsC;IAEtC;;;;;;OAMG;IACH,YALW,MAAM,GAAC,MAAM,EAAE,WACf,QAAQ,GAAC,QAAQ,EAAE,WACf,GAAG,KACL,QAAQ,GAAG,CAAC,CAoCxB;CACJ;0BAnDyB,gCAAgC;yBAGjC,sBAAsB;mCAFZ,kCAAkC;8BACvC,qBAAqB"}

package/types/models/rt_detr/image_processing_rt_detr.d.ts

@@ -0,0 +1,8 @@
+export class RTDetrImageProcessor extends ImageProcessor {
+    post_process_object_detection(outputs: {
+        logits: import("../../transformers.js").Tensor;
+        pred_boxes: import("../../transformers.js").Tensor;
+    }, threshold?: number, target_sizes?: [number, number][], is_zero_shot?: boolean): any[];
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_rt_detr.d.ts.map

package/types/models/rt_detr/image_processing_rt_detr.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_rt_detr.d.ts","sourceRoot":"","sources":["../../../src/models/rt_detr/image_processing_rt_detr.js"],"names":[],"mappings":"AAMA;IAMgyE;;;6FAAu6F;CADtsK;+BARM,sCAAsC"}

package/types/models/sam/image_processing_sam.d.ts

@@ -0,0 +1,103 @@
+/**
+ * @typedef {object} SamImageProcessorResult
+ * @property {Tensor} pixel_values
+ * @property {import("../../base/image_processors_utils.js").HeightWidth[]} original_sizes
+ * @property {import("../../base/image_processors_utils.js").HeightWidth[]} reshaped_input_sizes
+ * @property {Tensor} [input_points]
+ * @property {Tensor} [input_labels]
+ * @property {Tensor} [input_boxes]
+ */
+export class SamImageProcessor extends ImageProcessor {
+    /**
+     *
+     * @param {any} input_points
+     * @param {import("../../base/image_processors_utils.js").HeightWidth[]} original_sizes
+     * @param {import("../../base/image_processors_utils.js").HeightWidth[]} reshaped_input_sizes
+     * @returns {Tensor}
+     */
+    reshape_input_points(input_points: any, original_sizes: import("../../base/image_processors_utils.js").HeightWidth[], reshaped_input_sizes: import("../../base/image_processors_utils.js").HeightWidth[], is_bounding_box?: boolean): Tensor;
+    /**
+     *
+     * @param {any} input_labels
+     * @param {Tensor} input_points
+     * @returns {Tensor}
+     */
+    add_input_labels(input_labels: any, input_points: Tensor): Tensor;
+    /**
+     * @param {any[]} images The URL(s) of the image(s) to extract features from.
+     * @param {Object} [options] Additional options for the processor.
+     * @param {any} [options.input_points=null] A 3D or 4D array, representing the input points provided by the user.
+     * - 3D: `[point_batch_size, nb_points_per_image, 2]`. In this case, `batch_size` is assumed to be 1.
+     * - 4D: `[batch_size, point_batch_size, nb_points_per_image, 2]`.
+     * @param {any} [options.input_labels=null] A 2D or 3D array, representing the input labels for the points, used by the prompt encoder to encode the prompt.
+     * - 2D: `[point_batch_size, nb_points_per_image]`. In this case, `batch_size` is assumed to be 1.
+     * - 3D: `[batch_size, point_batch_size, nb_points_per_image]`.
+     * @param {number[][][]} [options.input_boxes=null] A 3D array of shape `(batch_size, num_boxes, 4)`, representing the input boxes provided by the user.
+     * This is used by the prompt encoder to encode the prompt. Generally yields to much better generated masks.
+     * The processor will generate a tensor, with each dimension corresponding respectively to the image batch size,
+     * the number of boxes per image and the coordinates of the top left and botton right point of the box.
+     * In the order (`x1`, `y1`, `x2`, `y2`):
+     * - `x1`: the x coordinate of the top left point of the input box
+     * - `y1`: the y coordinate of the top left point of the input box
+     * - `x2`: the x coordinate of the bottom right point of the input box
+     * - `y2`: the y coordinate of the bottom right point of the input box
+     * @returns {Promise<SamImageProcessorResult>}
+     */
+    _call(images: any[], { input_points, input_labels, input_boxes }?: {
+        input_points?: any;
+        input_labels?: any;
+        input_boxes?: number[][][];
+    }): Promise<SamImageProcessorResult>;
+    /**
+     * Remove padding and upscale masks to the original image size.
+     * @param {Tensor} masks Batched masks from the mask_decoder in (batch_size, num_channels, height, width) format.
+     * @param {[number, number][]} original_sizes The original sizes of each image before it was resized to the model's expected input shape, in (height, width) format.
+     * @param {[number, number][]} reshaped_input_sizes The size of each image as it is fed to the model, in (height, width) format. Used to remove padding.
+     * @param {Object} options Optional parameters for post-processing.
+     * @param {number} [options.mask_threshold] The threshold to use for binarizing the masks.
+     * @param {boolean} [options.binarize] Whether to binarize the masks.
+     * @param {Object} [options.pad_size] The target size the images were padded to before being passed to the model. If `null`, the target size is assumed to be the processor's `pad_size`.
+     * @param {number} [options.pad_size.height] The height the images were padded to.
+     * @param {number} [options.pad_size.width] The width the images were padded to.
+     * @returns {Promise<Tensor[]>} Batched masks in batch_size, num_channels, height, width) format, where (height, width) is given by original_size.
+     */
+    post_process_masks(masks: Tensor, original_sizes: [number, number][], reshaped_input_sizes: [number, number][], { mask_threshold, binarize, pad_size, }?: {
+        mask_threshold?: number;
+        binarize?: boolean;
+        pad_size?: {
+            height?: number;
+            width?: number;
+        };
+    }): Promise<Tensor[]>;
+    /**
+     * Generates a list of crop boxes of different sizes. Each layer has (2**i)**2 boxes for the ith layer.
+     * @param {import("../../utils/image.js").RawImage} image Input original image
+     * @param {number} target_size Target size of the resized image
+     * @param {Object} options Options for generating crop boxes
+     * @param {number} [options.crop_n_layers] If >0, mask prediction will be run again on crops of the image.
+     * Sets the number of layers to run, where each layer has 2**i_layer number of image crops.
+     * @param {number} [options.overlap_ratio] Sets the degree to which crops overlap. In the first crop layer,
+     * crops will overlap by this fraction of the image length. Later layers with more crops scale down this overlap.
+     * @param {number} [options.points_per_crop] Number of points to sample from each crop.
+     * @param {number} [options.crop_n_points_downscale_factor] The number of points-per-side sampled in layer n is
+     * scaled down by crop_n_points_downscale_factor**n.
+     * @returns {Object} An object containing the crop boxes, number of points per crop, cropped images, and input labels.
+     */
+    generate_crop_boxes(image: import("../../utils/image.js").RawImage, target_size: number, { crop_n_layers, overlap_ratio, points_per_crop, crop_n_points_downscale_factor, }?: {
+        crop_n_layers?: number;
+        overlap_ratio?: number;
+        points_per_crop?: number;
+        crop_n_points_downscale_factor?: number;
+    }): any;
+}
+export type SamImageProcessorResult = {
+    pixel_values: Tensor;
+    original_sizes: import("../../base/image_processors_utils.js").HeightWidth[];
+    reshaped_input_sizes: import("../../base/image_processors_utils.js").HeightWidth[];
+    input_points?: Tensor;
+    input_labels?: Tensor;
+    input_boxes?: Tensor;
+};
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+import { Tensor } from "../../utils/tensor.js";
+//# sourceMappingURL=image_processing_sam.d.ts.map

package/types/models/sam/image_processing_sam.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_sam.d.ts","sourceRoot":"","sources":["../../../src/models/sam/image_processing_sam.js"],"names":[],"mappings":"AAWA;;;;;;;;GAQG;AAEH;IAEI;;;;;;OAMG;IACH,mCALW,GAAG,kBACH,OAAO,sCAAsC,EAAE,WAAW,EAAE,wBAC5D,OAAO,sCAAsC,EAAE,WAAW,EAAE,8BAC1D,MAAM,CA4ClB;IAED;;;;;OAKG;IACH,+BAJW,GAAG,gBACH,MAAM,GACJ,MAAM,CAoBlB;IACD;;;;;;;;;;;;;;;;;;;OAmBG;IACH,cAnBW,GAAG,EAAE;QAES,YAAY,GAA1B,GAAG;QAGW,YAAY,GAA1B,GAAG;QAGoB,WAAW,GAAlC,MAAM,EAAE,EAAE,EAAE;QASV,QAAQ,uBAAuB,CAAC,CA+B5C;IAED;;;;;;;;;;;;OAYG;IACH,0BAXW,MAAM,kBACN,CAAC,MAAM,EAAE,MAAM,CAAC,EAAE,wBAClB,CAAC,MAAM,EAAE,MAAM,CAAC,EAAE;QAED,cAAc,GAA/B,MAAM;QACY,QAAQ,GAA1B,OAAO;QACU,QAAQ;YACC,MAAM,GAAhC,MAAM;YACoB,KAAK,GAA/B,MAAM;;QACJ,QAAQ,MAAM,EAAE,CAAC,CAsD7B;IAED;;;;;;;;;;;;;OAaG;IACH,2BAZW,OAAO,sBAAsB,EAAE,QAAQ,eACvC,MAAM;QAEW,aAAa,GAA9B,MAAM;QAEW,aAAa,GAA9B,MAAM;QAEW,eAAe,GAAhC,MAAM;QACW,8BAA8B,GAA/C,MAAM;YAYhB;CACJ;;kBAnOa,MAAM;oBACN,OAAO,sCAAsC,EAAE,WAAW,EAAE;0BAC5D,OAAO,sCAAsC,EAAE,WAAW,EAAE;mBAC5D,MAAM;mBACN,MAAM;kBACN,MAAM;;+BAhBb,sCAAsC;uBAMtC,uBAAuB"}

package/types/models/sam/processing_sam.d.ts

@@ -0,0 +1,9 @@
+export class SamProcessor extends Processor {
+    static image_processor_class: typeof AutoImageProcessor;
+    _call(...args: any[]): Promise<any>;
+    post_process_masks(...args: any[]): any;
+    reshape_input_points(...args: any[]): any;
+}
+import { Processor } from "../../base/processing_utils.js";
+import { AutoImageProcessor } from "../auto/image_processing_auto.js";
+//# sourceMappingURL=processing_sam.d.ts.map

package/types/models/sam/processing_sam.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"processing_sam.d.ts","sourceRoot":"","sources":["../../../src/models/sam/processing_sam.js"],"names":[],"mappings":"AAGA;IACI,wDAAiD;IAEjD,oCAEC;IAED,wCAGC;IAED,0CAGC;CACJ;0BAnByB,gCAAgC;mCACvB,kCAAkC"}

package/types/models/seamless_m4t/feature_extraction_seamless_m4t.d.ts

@@ -0,0 +1,34 @@
+export class SeamlessM4TFeatureExtractor extends FeatureExtractor {
+    constructor(config: any);
+    mel_filters: number[][];
+    window: Float64Array;
+    /**
+     * Computes the log-Mel spectrogram of the provided audio waveform.
+     * @param {Float32Array|Float64Array} waveform The audio waveform to process.
+     * @param {number} max_length The maximum number of frames to return.
+     * @returns {Promise<Tensor>} An object containing the log-Mel spectrogram data as a Float32Array and its dimensions as an array of numbers.
+     */
+    _extract_fbank_features(waveform: Float32Array | Float64Array, max_length: number): Promise<Tensor>;
+    /**
+     * Asynchronously extracts features from a given audio using the provided configuration.
+     * @param {Float32Array|Float64Array} audio The audio data as a Float32Array/Float64Array.
+     * @param {Object} options Optional parameters for feature extraction.
+     * @param {boolean} [options.padding=true] Whether to pad the sequence to a multiple of `pad_to_multiple_of`.
+     * @param {number} [options.pad_to_multiple_of=2] The number to pad the sequence to a multiple of.
+     * @param {boolean} [options.do_normalize_per_mel_bins=true] Whether or not to zero-mean unit-variance normalize the input per mel-channel.
+     * @param {boolean} [options.return_attention_mask=true] Whether to return the attention mask.
+     * @returns {Promise<{ input_features: Tensor, attention_mask?: Tensor }>} A Promise resolving to an object containing the extracted input features and attention masks as Tensors.
+     */
+    _call(audio: Float32Array | Float64Array, { padding, pad_to_multiple_of, do_normalize_per_mel_bins, return_attention_mask, }?: {
+        padding?: boolean;
+        pad_to_multiple_of?: number;
+        do_normalize_per_mel_bins?: boolean;
+        return_attention_mask?: boolean;
+    }): Promise<{
+        input_features: Tensor;
+        attention_mask?: Tensor;
+    }>;
+}
+import { FeatureExtractor } from '../../base/feature_extraction_utils.js';
+import { Tensor } from '../../utils/tensor.js';
+//# sourceMappingURL=feature_extraction_seamless_m4t.d.ts.map

package/types/models/seamless_m4t/feature_extraction_seamless_m4t.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"feature_extraction_seamless_m4t.d.ts","sourceRoot":"","sources":["../../../src/models/seamless_m4t/feature_extraction_seamless_m4t.js"],"names":[],"mappings":"AAIA;IAEI,yBAwBC;IALG,wBAA8B;IAE9B,qBAEE;IAGN;;;;;OAKG;IACH,kCAJW,YAAY,GAAC,YAAY,cACzB,MAAM,GACJ,QAAQ,MAAM,CAAC,CA6B3B;IAED;;;;;;;;;OASG;IACH,aARW,YAAY,GAAC,YAAY;QAEP,OAAO,GAAzB,OAAO;QACU,kBAAkB,GAAnC,MAAM;QACY,yBAAyB,GAA3C,OAAO;QACW,qBAAqB,GAAvC,OAAO;;wBACqB,MAAM;yBAAmB,MAAM;OAuGrE;CACJ;iCAnLuD,wCAAwC;uBACzE,uBAAuB"}

package/types/models/segformer/image_processing_segformer.d.ts

@@ -0,0 +1,10 @@
+export class SegformerImageProcessor extends ImageProcessor {
+    post_process_semantic_segmentation(outputs: any, target_sizes?: [number, number][]): {
+        segmentation: import("../../transformers.js").Tensor;
+        labels: number[];
+    }[];
+}
+export class SegformerFeatureExtractor extends SegformerImageProcessor {
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_segformer.d.ts.map

package/types/models/segformer/image_processing_segformer.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_segformer.d.ts","sourceRoot":"","sources":["../../../src/models/segformer/image_processing_segformer.js"],"names":[],"mappings":"AAMA;IAOqmK;;;QAA21E;CAF/7O;AACD;CAA0E;+BATnE,sCAAsC"}

package/types/models/siglip/image_processing_siglip.d.ts

@@ -0,0 +1,4 @@
+export class SiglipImageProcessor extends ImageProcessor {
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_siglip.d.ts.map

package/types/models/siglip/image_processing_siglip.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_siglip.d.ts","sourceRoot":"","sources":["../../../src/models/siglip/image_processing_siglip.js"],"names":[],"mappings":"AAIA;CAA4D;+BAFrD,sCAAsC"}

package/types/models/speecht5/feature_extraction_speecht5.d.ts

@@ -0,0 +1,4 @@
+export class SpeechT5FeatureExtractor extends FeatureExtractor {
+}
+import { FeatureExtractor } from "../../base/feature_extraction_utils.js";
+//# sourceMappingURL=feature_extraction_speecht5.d.ts.map

package/types/models/speecht5/feature_extraction_speecht5.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"feature_extraction_speecht5.d.ts","sourceRoot":"","sources":["../../../src/models/speecht5/feature_extraction_speecht5.js"],"names":[],"mappings":"AAGA;CAAkE;iCAFjC,wCAAwC"}

package/types/models/speecht5/processing_speecht5.d.ts

@@ -0,0 +1,14 @@
+export class SpeechT5Processor extends Processor {
+    static tokenizer_class: typeof AutoTokenizer;
+    static feature_extractor_class: typeof AutoFeatureExtractor;
+    /**
+     * Calls the feature_extractor function with the given input.
+     * @param {any} input The input to extract features from.
+     * @returns {Promise<any>} A Promise that resolves with the extracted features.
+     */
+    _call(input: any): Promise<any>;
+}
+import { Processor } from "../../base/processing_utils.js";
+import { AutoTokenizer } from "../../tokenizers.js";
+import { AutoFeatureExtractor } from "../auto/feature_extraction_auto.js";
+//# sourceMappingURL=processing_speecht5.d.ts.map

package/types/models/speecht5/processing_speecht5.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"processing_speecht5.d.ts","sourceRoot":"","sources":["../../../src/models/speecht5/processing_speecht5.js"],"names":[],"mappings":"AAIA;IACI,6CAAsC;IACtC,4DAAqD;IAErD;;;;OAIG;IACH,aAHW,GAAG,GACD,QAAQ,GAAG,CAAC,CAIxB;CACJ;0BAhByB,gCAAgC;8BAC5B,qBAAqB;qCACd,oCAAoC"}

package/types/models/swin2sr/image_processing_swin2sr.d.ts

@@ -0,0 +1,5 @@
+export class Swin2SRImageProcessor extends ImageProcessor {
+    pad_image(pixelData: any, imgDims: any, padSize: any, options?: {}): [Float32Array, number[]];
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_swin2sr.d.ts.map

package/types/models/swin2sr/image_processing_swin2sr.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_swin2sr.d.ts","sourceRoot":"","sources":["../../../src/models/swin2sr/image_processing_swin2sr.js"],"names":[],"mappings":"AAIA;IACI,8FAiBC;CACJ;+BArBM,sCAAsC"}

package/types/models/vit/image_processing_vit.d.ts

@@ -0,0 +1,6 @@
+export class ViTImageProcessor extends ImageProcessor {
+}
+export class ViTFeatureExtractor extends ViTImageProcessor {
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_vit.d.ts.map

package/types/models/vit/image_processing_vit.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_vit.d.ts","sourceRoot":"","sources":["../../../src/models/vit/image_processing_vit.js"],"names":[],"mappings":"AAIA;CAAyD;AACzD;CAA8D;+BAHvD,sCAAsC"}

package/types/models/vitmatte/image_processing_vitmatte.d.ts

@@ -0,0 +1,12 @@
+export class VitMatteImageProcessor extends ImageProcessor {
+    /**
+     * Calls the feature extraction process on an array of images, preprocesses
+     * each image, and concatenates the resulting features into a single Tensor.
+     * @param {import("../../utils/image.js").RawImage[]} images The image(s) to extract features from.
+     * @param {import("../../utils/image.js").RawImage[]} trimaps The trimaps(s) to extract features from.
+     * @returns {Promise<import("../../base/image_processors_utils.js").ImageProcessorResult>} An object containing the concatenated pixel values of the preprocessed images.
+     */
+    _call(images: import("../../utils/image.js").RawImage[], trimaps: import("../../utils/image.js").RawImage[]): Promise<import("../../base/image_processors_utils.js").ImageProcessorResult>;
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_vitmatte.d.ts.map

package/types/models/vitmatte/image_processing_vitmatte.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_vitmatte.d.ts","sourceRoot":"","sources":["../../../src/models/vitmatte/image_processing_vitmatte.js"],"names":[],"mappings":"AASA;IACI;;;;;;OAMG;IACH,cAJW,OAAO,sBAAsB,EAAE,QAAQ,EAAE,WACzC,OAAO,sBAAsB,EAAE,QAAQ,EAAE,GACvC,QAAQ,OAAO,sCAAsC,EAAE,oBAAoB,CAAC,CAiCxF;CACJ;+BA/CM,sCAAsC"}

package/types/models/vitpose/image_processing_vitpose.d.ts

@@ -0,0 +1,26 @@
+export class VitPoseImageProcessor extends ImageProcessor {
+    /**
+     * Transform the heatmaps into keypoint predictions and transform them back to the image.
+     * NOTE: This is a naive implementation and does not include advanced post-processing techniques,
+     * so the results may not be as accurate as the original implementation.
+     * @param {import('../../utils/tensor.js').Tensor} outputs The model outputs.
+     * @param {[number, number, number, number][][]} boxes List or array of bounding boxes for each image.
+     * Each box should be a list of 4 floats representing the bounding box coordinates in COCO format (top_left_x, top_left_y, width, height).
+     * @returns {{
+     *   bbox: [number, number, number, number],
+     *   scores: number[],
+     *   labels: number[],
+     *   keypoints: [number, number][]
+     * }[][]} List of keypoints predictions for each image.
+     */
+    post_process_pose_estimation(outputs: import('../../utils/tensor.js').Tensor, boxes: [number, number, number, number][][], { threshold, }?: {
+        threshold?: any;
+    }): {
+        bbox: [number, number, number, number];
+        scores: number[];
+        labels: number[];
+        keypoints: [number, number][];
+    }[][];
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_vitpose.d.ts.map

package/types/models/vitpose/image_processing_vitpose.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_vitpose.d.ts","sourceRoot":"","sources":["../../../src/models/vitpose/image_processing_vitpose.js"],"names":[],"mappings":"AAIA;IAEI;;;;;;;;;;;;;OAaG;IACH,sCAVW,OAAO,uBAAuB,EAAE,MAAM,SACtC,CAAC,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,CAAC,EAAE,EAAE;;QAElC;QACZ,IAAQ,EAAE,CAAC,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,CAAC,CAAC;QAC3C,MAAU,EAAE,MAAM,EAAE,CAAC;QACrB,MAAU,EAAE,MAAM,EAAE,CAAC;QACrB,SAAa,EAAE,CAAC,MAAM,EAAE,MAAM,CAAC,EAAE,CAAA;KAC9B,EAAE,EAAE,CAqEP;CACJ;+BAtFM,sCAAsC"}

package/types/models/wav2vec2/feature_extraction_wav2vec2.d.ts

@@ -0,0 +1,19 @@
+export class Wav2Vec2FeatureExtractor extends FeatureExtractor {
+    /**
+     * @param {Float32Array} input_values
+     * @returns {Float32Array}
+     */
+    _zero_mean_unit_var_norm(input_values: Float32Array): Float32Array;
+    /**
+     * Asynchronously extracts features from a given audio using the provided configuration.
+     * @param {Float32Array|Float64Array} audio The audio data as a Float32Array/Float64Array.
+     * @returns {Promise<{ input_values: Tensor; attention_mask: Tensor }>} A Promise resolving to an object containing the extracted input features and attention mask as Tensors.
+     */
+    _call(audio: Float32Array | Float64Array): Promise<{
+        input_values: Tensor;
+        attention_mask: Tensor;
+    }>;
+}
+import { FeatureExtractor } from "../../base/feature_extraction_utils.js";
+import { Tensor } from "../../utils/tensor.js";
+//# sourceMappingURL=feature_extraction_wav2vec2.d.ts.map

package/types/models/wav2vec2/feature_extraction_wav2vec2.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"feature_extraction_wav2vec2.d.ts","sourceRoot":"","sources":["../../../src/models/wav2vec2/feature_extraction_wav2vec2.js"],"names":[],"mappings":"AAGA;IAEI;;;OAGG;IACH,uCAHW,YAAY,GACV,YAAY,CAQxB;IAED;;;;OAIG;IACH,aAHW,YAAY,GAAC,YAAY;sBACC,MAAM;wBAAkB,MAAM;OAsBlE;CACJ;iCA3CuD,wCAAwC;uBACzE,uBAAuB"}

package/types/models/wav2vec2/processing_wav2vec2.d.ts

@@ -0,0 +1,12 @@
+export class Wav2Vec2ProcessorWithLM extends Processor {
+    static feature_extractor_class: typeof AutoFeatureExtractor;
+    /**
+     * Calls the feature_extractor function with the given audio input.
+     * @param {any} audio The audio input to extract features from.
+     * @returns {Promise<any>} A Promise that resolves with the extracted features.
+     */
+    _call(audio: any): Promise<any>;
+}
+import { Processor } from "../../base/processing_utils.js";
+import { AutoFeatureExtractor } from "../auto/feature_extraction_auto.js";
+//# sourceMappingURL=processing_wav2vec2.d.ts.map

package/types/models/wav2vec2/processing_wav2vec2.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"processing_wav2vec2.d.ts","sourceRoot":"","sources":["../../../src/models/wav2vec2/processing_wav2vec2.js"],"names":[],"mappings":"AAGA;IACI,4DAAqD;IAErD;;;;OAIG;IACH,aAHW,GAAG,GACD,QAAQ,GAAG,CAAC,CAIxB;CACJ;0BAdyB,gCAAgC;qCACrB,oCAAoC"}

package/types/models/wespeaker/feature_extraction_wespeaker.d.ts

@@ -0,0 +1,23 @@
+export class WeSpeakerFeatureExtractor extends FeatureExtractor {
+    constructor(config: any);
+    mel_filters: number[][];
+    window: Float64Array;
+    min_num_frames: any;
+    /**
+     * Computes the log-Mel spectrogram of the provided audio waveform.
+     * @param {Float32Array|Float64Array} waveform The audio waveform to process.
+     * @returns {Promise<Tensor>} An object containing the log-Mel spectrogram data as a Float32Array and its dimensions as an array of numbers.
+     */
+    _extract_fbank_features(waveform: Float32Array | Float64Array): Promise<Tensor>;
+    /**
+     * Asynchronously extracts features from a given audio using the provided configuration.
+     * @param {Float32Array|Float64Array} audio The audio data as a Float32Array/Float64Array.
+     * @returns {Promise<{ input_features: Tensor }>} A Promise resolving to an object containing the extracted input features as a Tensor.
+     */
+    _call(audio: Float32Array | Float64Array): Promise<{
+        input_features: Tensor;
+    }>;
+}
+import { FeatureExtractor } from '../../base/feature_extraction_utils.js';
+import { Tensor } from '../../utils/tensor.js';
+//# sourceMappingURL=feature_extraction_wespeaker.d.ts.map

package/types/models/wespeaker/feature_extraction_wespeaker.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"feature_extraction_wespeaker.d.ts","sourceRoot":"","sources":["../../../src/models/wespeaker/feature_extraction_wespeaker.js"],"names":[],"mappings":"AAKA;IAEI,yBAyBC;IANG,wBAA8B;IAE9B,qBAEE;IACF,oBAAgD;IAGpD;;;;OAIG;IACH,kCAHW,YAAY,GAAC,YAAY,GACvB,QAAQ,MAAM,CAAC,CA2B3B;IAGD;;;;OAIG;IACH,aAHW,YAAY,GAAC,YAAY,GACvB,QAAQ;QAAE,cAAc,EAAE,MAAM,CAAA;KAAE,CAAC,CA4B/C;CACJ;iCAnGuD,wCAAwC;uBACzE,uBAAuB"}

package/types/models/whisper/feature_extraction_whisper.d.ts

@@ -0,0 +1,21 @@
+export class WhisperFeatureExtractor extends FeatureExtractor {
+    constructor(config: any);
+    window: Float64Array;
+    /**
+     * Computes the log-Mel spectrogram of the provided audio waveform.
+     * @param {Float32Array|Float64Array} waveform The audio waveform to process.
+     * @returns {Promise<Tensor>} An object containing the log-Mel spectrogram data as a Float32Array and its dimensions as an array of numbers.
+     */
+    _extract_fbank_features(waveform: Float32Array | Float64Array): Promise<Tensor>;
+    /**
+     * Asynchronously extracts features from a given audio using the provided configuration.
+     * @param {Float32Array|Float64Array} audio The audio data as a Float32Array/Float64Array.
+     * @returns {Promise<{ input_features: Tensor }>} A Promise resolving to an object containing the extracted input features as a Tensor.
+     */
+    _call(audio: Float32Array | Float64Array): Promise<{
+        input_features: Tensor;
+    }>;
+}
+import { FeatureExtractor } from '../../base/feature_extraction_utils.js';
+import { Tensor } from '../../utils/tensor.js';
+//# sourceMappingURL=feature_extraction_whisper.d.ts.map

package/types/models/whisper/feature_extraction_whisper.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"feature_extraction_whisper.d.ts","sourceRoot":"","sources":["../../../src/models/whisper/feature_extraction_whisper.js"],"names":[],"mappings":"AAKA;IAEI,yBAeC;IADG,qBAAwD;IAG5D;;;;OAIG;IACH,kCAHW,YAAY,GAAC,YAAY,GACvB,QAAQ,MAAM,CAAC,CA0B3B;IAED;;;;OAIG;IACH,aAHW,YAAY,GAAC,YAAY,GACvB,QAAQ;QAAE,cAAc,EAAE,MAAM,CAAA;KAAE,CAAC,CAwB/C;CACJ;iCAnFuD,wCAAwC;uBACzE,uBAAuB"}

package/types/models/whisper/processing_whisper.d.ts

@@ -0,0 +1,17 @@
+/**
+ * Represents a WhisperProcessor that extracts features from an audio input.
+ */
+export class WhisperProcessor extends Processor {
+    static tokenizer_class: typeof AutoTokenizer;
+    static feature_extractor_class: typeof AutoFeatureExtractor;
+    /**
+     * Calls the feature_extractor function with the given audio input.
+     * @param {any} audio The audio input to extract features from.
+     * @returns {Promise<any>} A Promise that resolves with the extracted features.
+     */
+    _call(audio: any): Promise<any>;
+}
+import { Processor } from "../../base/processing_utils.js";
+import { AutoTokenizer } from "../../tokenizers.js";
+import { AutoFeatureExtractor } from "../auto/feature_extraction_auto.js";
+//# sourceMappingURL=processing_whisper.d.ts.map

package/types/models/whisper/processing_whisper.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"processing_whisper.d.ts","sourceRoot":"","sources":["../../../src/models/whisper/processing_whisper.js"],"names":[],"mappings":"AAIA;;GAEG;AACH;IACI,6CAAsC;IACtC,4DAAqD;IAErD;;;;OAIG;IACH,aAHW,GAAG,GACD,QAAQ,GAAG,CAAC,CAIxB;CACJ;0BAjByB,gCAAgC;8BAD5B,qBAAqB;qCADd,oCAAoC"}

package/types/models/yolos/image_processing_yolos.d.ts

@@ -0,0 +1,10 @@
+export class YolosImageProcessor extends ImageProcessor {
+    post_process_object_detection(outputs: {
+        logits: import("../../transformers.js").Tensor;
+        pred_boxes: import("../../transformers.js").Tensor;
+    }, threshold?: number, target_sizes?: [number, number][], is_zero_shot?: boolean): any[];
+}
+export class YolosFeatureExtractor extends YolosImageProcessor {
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_yolos.d.ts.map

package/types/models/yolos/image_processing_yolos.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_yolos.d.ts","sourceRoot":"","sources":["../../../src/models/yolos/image_processing_yolos.js"],"names":[],"mappings":"AAKA;IAO+tE;;;6FAAu6F;CAFroK;AACD;CAAkE;+BAR3D,sCAAsC"}