@min-pack/tfjs-node 2.17.2 → 3.8.1-patch.0

package/tfjs-types/models/llava/processing_llava.d.ts

@@ -0,0 +1,12 @@
+export class LlavaProcessor extends Processor {
+    static tokenizer_class: typeof AutoTokenizer;
+    static image_processor_class: typeof AutoImageProcessor;
+    /**
+     * @typedef {import('../../utils/image.js').RawImage} RawImage
+     */
+    _call(images: import("../../utils/image.js").RawImage | import("../../utils/image.js").RawImage[], text?: any, kwargs?: {}): Promise<any>;
+}
+import { Processor } from "../../base/processing_utils.js";
+import { AutoTokenizer } from "../../tokenizers.js";
+import { AutoImageProcessor } from "../auto/image_processing_auto.js";
+//# sourceMappingURL=processing_llava.d.ts.map

package/tfjs-types/models/llava_onevision/image_processing_llava_onevision.d.ts

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

package/tfjs-types/models/mask2former/image_processing_mask2former.d.ts

@@ -0,0 +1,4 @@
+export class Mask2FormerImageProcessor extends MaskFormerImageProcessor {
+}
+import { MaskFormerImageProcessor } from "../maskformer/image_processing_maskformer.js";
+//# sourceMappingURL=image_processing_mask2former.d.ts.map

package/tfjs-types/models/maskformer/image_processing_maskformer.d.ts

@@ -0,0 +1,22 @@
+export class MaskFormerImageProcessor extends ImageProcessor {
+    post_process_panoptic_segmentation(outputs: any, threshold?: number, mask_threshold?: number, overlap_mask_area_threshold?: number, label_ids_to_fuse?: Set<number>, target_sizes?: [number, number][]): Array<{
+        segmentation: import("../../transformers.js").Tensor;
+        segments_info: Array<{
+            id: number;
+            label_id: number;
+            score: number;
+        }>;
+    }>;
+    post_process_instance_segmentation(outputs: any, threshold?: number, target_sizes?: [number, number][]): Array<{
+        segmentation: import("../../transformers.js").Tensor;
+        segments_info: Array<{
+            id: number;
+            label_id: number;
+            score: number;
+        }>;
+    }>;
+}
+export class MaskFormerFeatureExtractor extends MaskFormerImageProcessor {
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_maskformer.d.ts.map

package/tfjs-types/models/mgp_str/processing_mgp_str.d.ts

@@ -0,0 +1,64 @@
+export class MgpstrProcessor extends Processor {
+    static tokenizer_class: typeof AutoTokenizer;
+    static image_processor_class: typeof AutoImageProcessor;
+    /**
+     * @returns {import('../../tokenizers.js').MgpstrTokenizer} The character tokenizer.
+     */
+    get char_tokenizer(): import("../../tokenizers.js").MgpstrTokenizer;
+    /**
+     * @returns {import('../../tokenizers.js').GPT2Tokenizer} The BPE tokenizer.
+     */
+    get bpe_tokenizer(): import("../../tokenizers.js").GPT2Tokenizer;
+    /**
+     * @returns {import('../../tokenizers.js').BertTokenizer} The WordPiece tokenizer.
+     */
+    get wp_tokenizer(): import("../../tokenizers.js").BertTokenizer;
+    /**
+     * Helper function to decode the model prediction logits.
+     * @param {import('../../utils/tensor.js').Tensor} pred_logits Model prediction logits.
+     * @param {string} format Type of model prediction. Must be one of ['char', 'bpe', 'wp'].
+     * @returns {[string[], number[]]} The decoded sentences and their confidence scores.
+     */
+    _decode_helper(pred_logits: import("../../utils/tensor.js").Tensor, format: string): [string[], number[]];
+    /**
+     * Convert a list of lists of char token ids into a list of strings by calling char tokenizer.
+     * @param {number[][]} sequences List of tokenized input ids.
+     * @returns {string[]} The list of char decoded sentences.
+     */
+    char_decode(sequences: number[][]): string[];
+    /**
+     * Convert a list of lists of BPE token ids into a list of strings by calling BPE tokenizer.
+     * @param {number[][]} sequences List of tokenized input ids.
+     * @returns {string[]} The list of BPE decoded sentences.
+     */
+    bpe_decode(sequences: number[][]): string[];
+    /**
+     * Convert a list of lists of word piece token ids into a list of strings by calling word piece tokenizer.
+     * @param {number[][]} sequences List of tokenized input ids.
+     * @returns {string[]} The list of wp decoded sentences.
+     */
+    wp_decode(sequences: number[][]): string[];
+    /**
+     * Convert a list of lists of token ids into a list of strings by calling decode.
+     * @param {[import('../../utils/tensor.js').Tensor, import('../../utils/tensor.js').Tensor, import('../../utils/tensor.js').Tensor]} sequences List of tokenized input ids.
+     * @returns {{generated_text: string[], scores: number[], char_preds: string[], bpe_preds: string[], wp_preds: string[]}}
+     * Dictionary of all the outputs of the decoded results.
+     * - generated_text: The final results after fusion of char, bpe, and wp.
+     * - scores: The final scores after fusion of char, bpe, and wp.
+     * - char_preds: The list of character decoded sentences.
+     * - bpe_preds: The list of BPE decoded sentences.
+     * - wp_preds: The list of wp decoded sentences.
+     */
+    batch_decode([char_logits, bpe_logits, wp_logits]: [import("../../utils/tensor.js").Tensor, import("../../utils/tensor.js").Tensor, import("../../utils/tensor.js").Tensor]): {
+        generated_text: string[];
+        scores: number[];
+        char_preds: string[];
+        bpe_preds: string[];
+        wp_preds: string[];
+    };
+    _call(images: any, text?: any): Promise<any>;
+}
+import { Processor } from "../../base/processing_utils.js";
+import { AutoTokenizer } from "../../tokenizers.js";
+import { AutoImageProcessor } from "../auto/image_processing_auto.js";
+//# sourceMappingURL=processing_mgp_str.d.ts.map

package/tfjs-types/models/mobilenet_v1/image_processing_mobilenet_v1.d.ts

@@ -0,0 +1,6 @@
+export class MobileNetV1ImageProcessor extends ImageProcessor {
+}
+export class MobileNetV1FeatureExtractor extends MobileNetV1ImageProcessor {
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_mobilenet_v1.d.ts.map

package/tfjs-types/models/mobilenet_v2/image_processing_mobilenet_v2.d.ts

@@ -0,0 +1,6 @@
+export class MobileNetV2ImageProcessor extends ImageProcessor {
+}
+export class MobileNetV2FeatureExtractor extends MobileNetV2ImageProcessor {
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_mobilenet_v2.d.ts.map

package/tfjs-types/models/mobilenet_v3/image_processing_mobilenet_v3.d.ts

@@ -0,0 +1,6 @@
+export class MobileNetV3ImageProcessor extends ImageProcessor {
+}
+export class MobileNetV3FeatureExtractor extends MobileNetV3ImageProcessor {
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_mobilenet_v3.d.ts.map

package/tfjs-types/models/mobilenet_v4/image_processing_mobilenet_v4.d.ts

@@ -0,0 +1,6 @@
+export class MobileNetV4ImageProcessor extends ImageProcessor {
+}
+export class MobileNetV4FeatureExtractor extends MobileNetV4ImageProcessor {
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_mobilenet_v4.d.ts.map

package/tfjs-types/models/mobilevit/image_processing_mobilevit.d.ts

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

package/tfjs-types/models/moonshine/feature_extraction_moonshine.d.ts

@@ -0,0 +1,13 @@
+export class MoonshineFeatureExtractor extends FeatureExtractor {
+    /**
+     * Asynchronously extracts input values from a given audio using the provided configuration.
+     * @param {Float32Array|Float64Array} audio The audio data as a Float32Array/Float64Array.
+     * @returns {Promise<{ input_values: Tensor; }>} The extracted input values.
+     */
+    _call(audio: Float32Array | Float64Array): Promise<{
+        input_values: Tensor;
+    }>;
+}
+import { FeatureExtractor } from '../../base/feature_extraction_utils.js';
+import { Tensor } from '../../utils/tensor.js';
+//# sourceMappingURL=feature_extraction_moonshine.d.ts.map

package/tfjs-types/models/moonshine/processing_moonshine.d.ts

@@ -0,0 +1,17 @@
+/**
+ * Represents a MoonshineProcessor that extracts features from an audio input.
+ */
+export class MoonshineProcessor 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_moonshine.d.ts.map

package/tfjs-types/models/nougat/image_processing_nougat.d.ts

@@ -0,0 +1,4 @@
+export class NougatImageProcessor extends DonutImageProcessor {
+}
+import { DonutImageProcessor } from "../donut/image_processing_donut.js";
+//# sourceMappingURL=image_processing_nougat.d.ts.map

package/tfjs-types/models/owlv2/image_processing_owlv2.d.ts

@@ -0,0 +1,4 @@
+export class Owlv2ImageProcessor extends OwlViTImageProcessor {
+}
+import { OwlViTImageProcessor } from "../owlvit/image_processing_owlvit.js";
+//# sourceMappingURL=image_processing_owlv2.d.ts.map

package/tfjs-types/models/owlvit/image_processing_owlvit.d.ts

@@ -0,0 +1,10 @@
+export class OwlViTImageProcessor 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 OwlViTFeatureExtractor extends OwlViTImageProcessor {
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_owlvit.d.ts.map

package/tfjs-types/models/owlvit/processing_owlvit.d.ts

@@ -0,0 +1,8 @@
+export class OwlViTProcessor extends Processor {
+    static tokenizer_class: typeof AutoTokenizer;
+    static image_processor_class: typeof AutoImageProcessor;
+}
+import { Processor } from "../../base/processing_utils.js";
+import { AutoTokenizer } from "../../tokenizers.js";
+import { AutoImageProcessor } from "../auto/image_processing_auto.js";
+//# sourceMappingURL=processing_owlvit.d.ts.map

package/tfjs-types/models/paligemma/processing_paligemma.d.ts

@@ -0,0 +1,12 @@
+export class PaliGemmaProcessor extends Processor {
+    static tokenizer_class: typeof AutoTokenizer;
+    static image_processor_class: typeof AutoImageProcessor;
+    /**
+     * @typedef {import('../../utils/image.js').RawImage} RawImage
+     */
+    _call(images: import("../../utils/image.js").RawImage | import("../../utils/image.js").RawImage[], text?: any, kwargs?: {}): Promise<any>;
+}
+import { Processor } from "../../base/processing_utils.js";
+import { AutoTokenizer } from "../../tokenizers.js";
+import { AutoImageProcessor } from "../auto/image_processing_auto.js";
+//# sourceMappingURL=processing_paligemma.d.ts.map

package/tfjs-types/models/parakeet/feature_extraction_parakeet.d.ts

@@ -0,0 +1,22 @@
+export class ParakeetFeatureExtractor extends FeatureExtractor {
+    constructor(config: any);
+    window: Float64Array<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; attention_mask: Tensor; }>} A Promise resolving to an object containing the extracted input features as a Tensor.
+     */
+    _call(audio: Float32Array | Float64Array): Promise<{
+        input_features: Tensor;
+        attention_mask: Tensor;
+    }>;
+}
+import { FeatureExtractor } from '../../base/feature_extraction_utils.js';
+import { Tensor } from '../../utils/tensor.js';
+//# sourceMappingURL=feature_extraction_parakeet.d.ts.map

package/tfjs-types/models/phi3_v/image_processing_phi3_v.d.ts

@@ -0,0 +1,17 @@
+export class Phi3VImageProcessor extends ImageProcessor {
+    constructor(config: any);
+    _num_crops: any;
+    calc_num_image_tokens_from_image_size(width: any, height: any): number;
+    _call(images: any, { num_crops, }?: {
+        num_crops?: any;
+    }): Promise<{
+        pixel_values: Tensor;
+        original_sizes: any[];
+        reshaped_input_sizes: any[];
+        image_sizes: Tensor;
+        num_img_tokens: number[];
+    }>;
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+import { Tensor } from "../../utils/tensor.js";
+//# sourceMappingURL=image_processing_phi3_v.d.ts.map

package/tfjs-types/models/phi3_v/processing_phi3_v.d.ts

@@ -0,0 +1,21 @@
+export class Phi3VProcessor extends Processor {
+    static image_processor_class: typeof AutoImageProcessor;
+    static tokenizer_class: typeof AutoTokenizer;
+    /**
+     *
+     * @param {string|string[]} text
+     * @param {RawImage|RawImage[]} images
+     * @param  { { padding?: boolean, truncation?: boolean, num_crops?: number } | undefined } options
+     * @returns {Promise<any>}
+     */
+    _call(text: string | string[], images?: RawImage | RawImage[], { padding, truncation, num_crops, }?: {
+        padding?: boolean;
+        truncation?: boolean;
+        num_crops?: number;
+    } | undefined): 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_phi3_v.d.ts.map

package/tfjs-types/models/pixtral/image_processing_pixtral.d.ts

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

package/tfjs-types/models/pixtral/processing_pixtral.d.ts

@@ -0,0 +1,12 @@
+export class PixtralProcessor extends Processor {
+    static tokenizer_class: typeof AutoTokenizer;
+    static image_processor_class: typeof AutoImageProcessor;
+    /**
+     * @typedef {import('../../utils/image.js').RawImage} RawImage
+     */
+    _call(images: import("../../utils/image.js").RawImage | import("../../utils/image.js").RawImage[], text?: any, kwargs?: {}): Promise<any>;
+}
+import { Processor } from "../../base/processing_utils.js";
+import { AutoTokenizer } from "../../tokenizers.js";
+import { AutoImageProcessor } from "../auto/image_processing_auto.js";
+//# sourceMappingURL=processing_pixtral.d.ts.map

package/tfjs-types/models/processors.d.ts

@@ -0,0 +1,25 @@
+export * from "./florence2/processing_florence2.js";
+export * from "./gemma3n/processing_gemma3n.js";
+export * from "./grounding_dino/processing_grounding_dino.js";
+export * from "./idefics3/processing_idefics3.js";
+export * from "./janus/processing_janus.js";
+export * from "./jina_clip/processing_jina_clip.js";
+export * from "./llava/processing_llava.js";
+export * from "./mgp_str/processing_mgp_str.js";
+export * from "./moonshine/processing_moonshine.js";
+export * from "./owlvit/processing_owlvit.js";
+export * from "./paligemma/processing_paligemma.js";
+export * from "./phi3_v/processing_phi3_v.js";
+export * from "./pixtral/processing_pixtral.js";
+export * from "./pyannote/processing_pyannote.js";
+export * from "./qwen2_vl/processing_qwen2_vl.js";
+export * from "./sam/processing_sam.js";
+export * from "./sam2/processing_sam2.js";
+export * from "./smolvlm/processing_smolvlm.js";
+export * from "./speecht5/processing_speecht5.js";
+export * from "./ultravox/processing_ultravox.js";
+export * from "./voxtral/processing_voxtral.js";
+export * from "./wav2vec2/processing_wav2vec2.js";
+export * from "./wav2vec2_with_lm/processing_wav2vec2_with_lm.js";
+export * from "./whisper/processing_whisper.js";
+//# sourceMappingURL=processors.d.ts.map

package/tfjs-types/models/pvt/image_processing_pvt.d.ts

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

package/tfjs-types/models/pyannote/feature_extraction_pyannote.d.ts

@@ -0,0 +1,31 @@
+export class PyAnnoteFeatureExtractor extends FeatureExtractor {
+    /**
+     * 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; }>} The extracted input features.
+     */
+    _call(audio: Float32Array | Float64Array): Promise<{
+        input_values: Tensor;
+    }>;
+    /**
+     * 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 { FeatureExtractor } from '../../base/feature_extraction_utils.js';
+import { Tensor } from '../../utils/tensor.js';
+//# sourceMappingURL=feature_extraction_pyannote.d.ts.map

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

@@ -0,0 +1,19 @@
+export class PyAnnoteProcessor extends Processor {
+    static feature_extractor_class: typeof PyAnnoteFeatureExtractor;
+    /**
+     * 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>;
+    post_process_speaker_diarization(logits: import("../../transformers.js").Tensor, num_samples: number): Array<Array<{
+        id: number;
+        start: number;
+        end: number;
+        confidence: number;
+    }>>;
+    get sampling_rate(): any;
+}
+import { Processor } from '../../base/processing_utils.js';
+import { PyAnnoteFeatureExtractor } from './feature_extraction_pyannote.js';
+//# sourceMappingURL=processing_pyannote.d.ts.map

package/tfjs-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/tfjs-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/tfjs-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/tfjs-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/tfjs-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/tfjs-types/models/sam2/image_processing_sam2.d.ts

@@ -0,0 +1,2 @@
+export { SamImageProcessor as Sam2ImageProcessor } from "../sam/image_processing_sam.js";
+//# sourceMappingURL=image_processing_sam2.d.ts.map

package/tfjs-types/models/sam2/processing_sam2.d.ts

@@ -0,0 +1,6 @@
+export class Sam2Processor extends SamProcessor {
+}
+export class Sam2VideoProcessor extends Sam2Processor {
+}
+import { SamProcessor } from "../sam/processing_sam.js";
+//# sourceMappingURL=processing_sam2.d.ts.map

package/tfjs-types/models/sam3/image_processing_sam3.d.ts

@@ -0,0 +1,2 @@
+export { Sam2ImageProcessor as Sam3ImageProcessor } from "../sam2/image_processing_sam2.js";
+//# sourceMappingURL=image_processing_sam3.d.ts.map

package/tfjs-types/models/sapiens/image_processing_sapiens.d.ts

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

package/tfjs-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<ArrayBufferLike>;
+    /**
+     * 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