@huggingface/transformers 3.0.2 → 3.1.1

package/types/generation/streamers.d.ts

@@ -16,18 +16,23 @@ export class TextStreamer extends BaseStreamer {
     /**
      *
      * @param {import('../tokenizers.js').PreTrainedTokenizer} tokenizer
+     * @param {Object} options
+     * @param {boolean} [options.skip_prompt=false] Whether to skip the prompt tokens
+     * @param {function(string): void} [options.callback_function=null] Function to call when a piece of text is ready to display
+     * @param {function(bigint[]): void} [options.token_callback_function=null] Function to call when a new token is generated
+     * @param {Object} [options.decode_kwargs={}] Additional keyword arguments to pass to the tokenizer's decode method
      */
     constructor(tokenizer: import('../tokenizers.js').PreTrainedTokenizer, { skip_prompt, callback_function, token_callback_function, decode_kwargs, ...kwargs }?: {
         skip_prompt?: boolean;
-        callback_function?: any;
-        token_callback_function?: any;
-        decode_kwargs?: {};
+        callback_function?: (arg0: string) => void;
+        token_callback_function?: (arg0: bigint[]) => void;
+        decode_kwargs?: any;
     });
     tokenizer: import("../tokenizers.js").PreTrainedTokenizer;
     skip_prompt: boolean;
-    callback_function: any;
-    token_callback_function: any;
-    decode_kwargs: {};
+    callback_function: (x: any) => void;
+    token_callback_function: (arg0: bigint[]) => void;
+    decode_kwargs: any;
     token_cache: any[];
     print_len: number;
     next_tokens_are_prompt: boolean;
@@ -52,7 +57,7 @@ export class WhisperTextStreamer extends TextStreamer {
      * @param {Object} options
      * @param {boolean} [options.skip_prompt=false] Whether to skip the prompt tokens
      * @param {function(string): void} [options.callback_function=null] Function to call when a piece of text is ready to display
-     * @param {function(string): void} [options.token_callback_function=null] Function to call when a new token is generated
+     * @param {function(bigint[]): void} [options.token_callback_function=null] Function to call when a new token is generated
      * @param {function(number): void} [options.on_chunk_start=null] Function to call when a new chunk starts
      * @param {function(number): void} [options.on_chunk_end=null] Function to call when a chunk ends
      * @param {function(): void} [options.on_finalize=null] Function to call when the stream is finalized
@@ -63,7 +68,7 @@ export class WhisperTextStreamer extends TextStreamer {
     constructor(tokenizer: import('../tokenizers.js').WhisperTokenizer, { skip_prompt, callback_function, token_callback_function, on_chunk_start, on_chunk_end, on_finalize, time_precision, skip_special_tokens, decode_kwargs, }?: {
         skip_prompt?: boolean;
         callback_function?: (arg0: string) => void;
-        token_callback_function?: (arg0: string) => void;
+        token_callback_function?: (arg0: bigint[]) => void;
         on_chunk_start?: (arg0: number) => void;
         on_chunk_end?: (arg0: number) => void;
         on_finalize?: () => void;

package/types/generation/streamers.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"streamers.d.ts","sourceRoot":"","sources":["../../src/generation/streamers.js"],"names":[],"mappings":"AASA;IACI;;;OAGG;IACH,WAFW,MAAM,EAAE,EAAE,QAIpB;IAED;;OAEG;IACH,YAEC;CACJ;AAMD;;GAEG;AACH;IACI;;;OAGG;IACH,uBAFW,OAAO,kBAAkB,EAAE,mBAAmB;;;;;OAoBxD;IAVG,0DAA0B;IAC1B,qBAA8B;IAC9B,uBAA0D;IAC1D,6BAAsD;IACtD,kBAAoD;IAGpD,mBAAqB;IACrB,kBAAkB;IAClB,gCAAkC;IA6DtC;;;;OAIG;IACH,wBAHW,MAAM,cACN,OAAO,QASjB;CACJ;AAED;;;;;;;GAOG;AACH;IACI;;;;;;;;;;;;OAYG;IACH,uBAZW,OAAO,kBAAkB,EAAE,gBAAgB;QAEzB,WAAW,GAA7B,OAAO;QAC0B,iBAAiB,UAAzC,MAAM,KAAG,IAAI;QACW,uBAAuB,UAA/C,MAAM,KAAG,IAAI;QACW,cAAc,UAAtC,MAAM,KAAG,IAAI;QACW,YAAY,UAApC,MAAM,KAAG,IAAI;QACK,WAAW,SAA1B,IAAI;QACC,cAAc,GAA/B,MAAM;QACY,mBAAmB,GAArC,OAAO;QACU,aAAa;OA4BxC;IATG,wBAAgD;IAEhD,uBA1BgB,MAAM,KAAG,IAAI,CA0BO;IACpC,qBA1BgB,MAAM,KAAG,IAAI,CA0BG;IAChC,mBA1BmB,IAAI,CA0BO;IAE9B,uBAAoC;IAEpC,+BAAkC;CAiCzC"}
+{"version":3,"file":"streamers.d.ts","sourceRoot":"","sources":["../../src/generation/streamers.js"],"names":[],"mappings":"AASA;IACI;;;OAGG;IACH,WAFW,MAAM,EAAE,EAAE,QAIpB;IAED;;OAEG;IACH,YAEC;CACJ;AAMD;;GAEG;AACH;IACI;;;;;;;;OAQG;IACH,uBAPW,OAAO,kBAAkB,EAAE,mBAAmB;QAE5B,WAAW,GAA7B,OAAO;QAC0B,iBAAiB,UAAzC,MAAM,KAAG,IAAI;QACa,uBAAuB,UAAjD,MAAM,EAAE,KAAG,IAAI;QACP,aAAa;OAoBxC;IAVG,0DAA0B;IAC1B,qBAA8B;IAC9B,oCAA0D;IAC1D,gCAdgB,MAAM,EAAE,KAAG,IAAI,CAcuB;IACtD,mBAAoD;IAGpD,mBAAqB;IACrB,kBAAkB;IAClB,gCAAkC;IA6DtC;;;;OAIG;IACH,wBAHW,MAAM,cACN,OAAO,QASjB;CACJ;AAED;;;;;;;GAOG;AACH;IACI;;;;;;;;;;;;OAYG;IACH,uBAZW,OAAO,kBAAkB,EAAE,gBAAgB;QAEzB,WAAW,GAA7B,OAAO;QAC0B,iBAAiB,UAAzC,MAAM,KAAG,IAAI;QACa,uBAAuB,UAAjD,MAAM,EAAE,KAAG,IAAI;QACS,cAAc,UAAtC,MAAM,KAAG,IAAI;QACW,YAAY,UAApC,MAAM,KAAG,IAAI;QACK,WAAW,SAA1B,IAAI;QACC,cAAc,GAA/B,MAAM;QACY,mBAAmB,GAArC,OAAO;QACU,aAAa;OA4BxC;IATG,wBAAgD;IAEhD,uBA1BgB,MAAM,KAAG,IAAI,CA0BO;IACpC,qBA1BgB,MAAM,KAAG,IAAI,CA0BG;IAChC,mBA1BmB,IAAI,CA0BO;IAE9B,uBAAoC;IAEpC,+BAAkC;CAiCzC"}

package/types/models/audio_spectrogram_transformer/feature_extraction_audio_spectrogram_transformer.d.ts

@@ -0,0 +1,25 @@
+export class ASTFeatureExtractor extends FeatureExtractor {
+    constructor(config: any);
+    mel_filters: number[][];
+    window: Float64Array;
+    mean: any;
+    std: any;
+    /**
+     * 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.
+     * @returns {Promise<{ input_values: Tensor }>} A Promise resolving to an object containing the extracted input features as a Tensor.
+     */
+    _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_audio_spectrogram_transformer.d.ts.map

package/types/models/audio_spectrogram_transformer/feature_extraction_audio_spectrogram_transformer.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"feature_extraction_audio_spectrogram_transformer.d.ts","sourceRoot":"","sources":["../../../src/models/audio_spectrogram_transformer/feature_extraction_audio_spectrogram_transformer.js"],"names":[],"mappings":"AAKA;IAEI,yBA2BC;IARG,wBAA8B;IAE9B,qBAEE;IAEF,UAA4B;IAC5B,SAA0B;IAG9B;;;;;OAKG;IACH,kCAJW,YAAY,GAAC,YAAY,cACzB,MAAM,GACJ,QAAQ,MAAM,CAAC,CAwB3B;IAGD;;;;OAIG;IACH,aAHW,YAAY,GAAC,YAAY,GACvB,QAAQ;QAAE,YAAY,EAAE,MAAM,CAAA;KAAE,CAAC,CAkB7C;CACJ;iCAzFuD,wCAAwC;uBACzE,uBAAuB"}

package/types/models/auto/feature_extraction_auto.d.ts

@@ -0,0 +1,5 @@
+export class AutoFeatureExtractor {
+    static from_pretrained(pretrained_model_name_or_path: string, options: import("../../utils/hub.js").PretrainedOptions): Promise<FeatureExtractor>;
+}
+import { FeatureExtractor } from '../../base/feature_extraction_utils.js';
+//# sourceMappingURL=feature_extraction_auto.d.ts.map

package/types/models/auto/feature_extraction_auto.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"feature_extraction_auto.d.ts","sourceRoot":"","sources":["../../../src/models/auto/feature_extraction_auto.js"],"names":[],"mappings":"AAMA;IAKsD,kJAoBH;CASlD;iCArCgC,wCAAwC"}

package/types/models/auto/image_processing_auto.d.ts

@@ -0,0 +1,5 @@
+export class AutoImageProcessor {
+    static from_pretrained(pretrained_model_name_or_path: string, options: import("../../utils/hub.js").PretrainedOptions): Promise<ImageProcessor>;
+}
+import { ImageProcessor } from '../../base/image_processors_utils.js';
+//# sourceMappingURL=image_processing_auto.d.ts.map

package/types/models/auto/image_processing_auto.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_auto.d.ts","sourceRoot":"","sources":["../../../src/models/auto/image_processing_auto.js"],"names":[],"mappings":"AAMA;IAuBkkwC,gJAAqyC;CADt2yC;+BAzB8B,sCAAsC"}

package/types/models/auto/processing_auto.d.ts

@@ -0,0 +1,35 @@
+/**
+ * Helper class which is used to instantiate pretrained processors with the `from_pretrained` function.
+ * The chosen processor class is determined by the type specified in the processor config.
+ *
+ * **Example:** Load a processor using `from_pretrained`.
+ * ```javascript
+ * let processor = await AutoProcessor.from_pretrained('openai/whisper-tiny.en');
+ * ```
+ *
+ * **Example:** Run an image through a processor.
+ * ```javascript
+ * let processor = await AutoProcessor.from_pretrained('Xenova/clip-vit-base-patch16');
+ * let image = await RawImage.read('https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/football-match.jpg');
+ * let image_inputs = await processor(image);
+ * // {
+ * //   "pixel_values": {
+ * //     "dims": [ 1, 3, 224, 224 ],
+ * //     "type": "float32",
+ * //     "data": Float32Array [ -1.558687686920166, -1.558687686920166, -1.5440893173217773, ... ],
+ * //     "size": 150528
+ * //   },
+ * //   "original_sizes": [
+ * //     [ 533, 800 ]
+ * //   ],
+ * //   "reshaped_input_sizes": [
+ * //     [ 224, 224 ]
+ * //   ]
+ * // }
+ * ```
+ */
+export class AutoProcessor {
+    static from_pretrained(pretrained_model_name_or_path: string, options: any): Promise<Processor>;
+}
+import { Processor } from '../../base/processing_utils.js';
+//# sourceMappingURL=processing_auto.d.ts.map

package/types/models/auto/processing_auto.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"processing_auto.d.ts","sourceRoot":"","sources":["../../../src/models/auto/processing_auto.js"],"names":[],"mappings":"AAUA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GA6BG;AACH;IAoCuB,gGAwB8qB;CADpsB;0BA/FyB,gCAAgC"}

package/types/models/beit/image_processing_beit.d.ts

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

package/types/models/beit/image_processing_beit.d.ts.map

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

package/types/models/bit/image_processing_bit.d.ts

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

package/types/models/bit/image_processing_bit.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_bit.d.ts","sourceRoot":"","sources":["../../../src/models/bit/image_processing_bit.js"],"names":[],"mappings":"AAIA;CAAyD;+BAFlD,sCAAsC"}

package/types/models/chinese_clip/image_processing_chinese_clip.d.ts

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

package/types/models/chinese_clip/image_processing_chinese_clip.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_chinese_clip.d.ts","sourceRoot":"","sources":["../../../src/models/chinese_clip/image_processing_chinese_clip.js"],"names":[],"mappings":"AAIA;CAAmE;+BAF5D,sCAAsC"}

package/types/models/clap/feature_extraction_clap.d.ts

@@ -0,0 +1,57 @@
+export class ClapFeatureExtractor extends FeatureExtractor {
+    constructor(config: any);
+    mel_filters: number[][];
+    mel_filters_slaney: number[][];
+    window: Float64Array;
+    /**
+     * Extracts the mel spectrogram and prepares it for the mode based on the `truncation` and `padding` arguments.
+     *
+     * Four different path are possible:
+     *   - `truncation="fusion"` and the length of the waveform is greater than the max length: the mel spectrogram
+     *     will be computed on the entire audio. 3 random crops and a dowsampled version of the full mel spectrogram
+     *     are then stacked together. They will later be used for `feature_fusion`.
+     *   - `truncation="rand_trunc"` and the length of the waveform is smaller than the max length: the audio is
+     *     padded based on `padding`.
+     *   - `truncation="fusion"` and the length of the waveform is smaller than the max length: the audio is padded
+     *     based on `padding`, and is repeated `4` times.
+     *   - `truncation="rand_trunc"` and the length of the waveform is greater than the max length: the mel
+     *     spectrogram will be computed on a random crop of the waveform.
+     *
+     * @param {Float32Array|Float64Array} waveform The input waveform.
+     * @param {number} max_length The maximum length of the waveform.
+     * @param {string} truncation The truncation strategy to use.
+     * @param {string} padding The padding strategy to use.
+     * @returns {Promise<Tensor>} An object containing the mel spectrogram data as a Float32Array, its dimensions as an array of numbers, and a boolean indicating whether the waveform was longer than the max length.
+     * @private
+     */
+    private _get_input_mel;
+    /**
+     * Compute the log-mel spectrogram of the provided `waveform` using the Hann window.
+     * In CLAP, two different filter banks are used depending on the truncation pattern:
+     *  - `self.mel_filters`: they correspond to the default parameters of `torchaudio` which can be obtained from
+     *    calling `torchaudio.transforms.MelSpectrogram().mel_scale.fb`. These filters are used when `truncation`
+     *    is set to `"fusion"`.
+     *  - `self.mel_filteres_slaney` : they correspond to the default parameters of `librosa` which used
+     *    `librosa.filters.mel` when computing the mel spectrogram. These filters were only used in the original
+     *    implementation when the truncation mode is not `"fusion"`.
+     *
+     * @param {Float32Array|Float64Array} waveform The audio waveform to process.
+     * @param {number[][]} mel_filters The mel filters to use.
+     * @param {number} [max_length=null] 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, mel_filters: number[][], 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.
+     * @returns {Promise<{ input_features: Tensor }>} A Promise resolving to an object containing the extracted input features as a Tensor.
+     */
+    _call(audio: Float32Array | Float64Array, { max_length, }?: {
+        max_length?: any;
+    }): Promise<{
+        input_features: Tensor;
+    }>;
+}
+import { FeatureExtractor } from '../../base/feature_extraction_utils.js';
+import { Tensor } from '../../utils/tensor.js';
+//# sourceMappingURL=feature_extraction_clap.d.ts.map

package/types/models/clap/feature_extraction_clap.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"feature_extraction_clap.d.ts","sourceRoot":"","sources":["../../../src/models/clap/feature_extraction_clap.js"],"names":[],"mappings":"AAKA;IAEI,yBAyBC;IAtBG,wBAQC;IAED,+BAQC;IAED,qBAAkE;IAKtE;;;;;;;;;;;;;;;;;;;;OAoBG;IACH,uBA0CC;IAED;;;;;;;;;;;;;;OAcG;IACH,kCALW,YAAY,GAAC,YAAY,eACzB,MAAM,EAAE,EAAE,eACV,MAAM,GACJ,QAAQ,MAAM,CAAC,CAoB3B;IAGD;;;;OAIG;IACH,aAHW,YAAY,GAAC,YAAY;;QACvB,QAAQ;QAAE,cAAc,EAAE,MAAM,CAAA;KAAE,CAAC,CAkB/C;CACJ;iCA9JuD,wCAAwC;uBACzE,uBAAuB"}

package/types/models/clip/image_processing_clip.d.ts

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

package/types/models/clip/image_processing_clip.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_clip.d.ts","sourceRoot":"","sources":["../../../src/models/clip/image_processing_clip.js"],"names":[],"mappings":"AAIA;CAA0D;AAC1D;CAAgE;+BAHzD,sCAAsC"}

package/types/models/convnext/image_processing_convnext.d.ts

@@ -0,0 +1,12 @@
+export class ConvNextImageProcessor extends ImageProcessor {
+    constructor(config: any);
+    /**
+     * Percentage of the image to crop. Only has an effect if this.size < 384.
+     */
+    crop_pct: any;
+    resize(image: any): Promise<any>;
+}
+export class ConvNextFeatureExtractor extends ConvNextImageProcessor {
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_convnext.d.ts.map

package/types/models/convnext/image_processing_convnext.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_convnext.d.ts","sourceRoot":"","sources":["../../../src/models/convnext/image_processing_convnext.js"],"names":[],"mappings":"AAIA;IACI,yBAOC;IAJG;;OAEG;IACH,cAAmD;IAGvD,iCA4BC;CACJ;AACD;CAAwE;+BA1CjE,sCAAsC"}

package/types/models/deit/image_processing_deit.d.ts

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

package/types/models/deit/image_processing_deit.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_deit.d.ts","sourceRoot":"","sources":["../../../src/models/deit/image_processing_deit.js"],"names":[],"mappings":"AAIA;CAA0D;AAC1D;CAAgE;+BAHzD,sCAAsC"}

package/types/models/detr/image_processing_detr.d.ts

@@ -0,0 +1,42 @@
+/**
+ * @typedef {object} DetrFeatureExtractorResultProps
+ * @property {import('../../utils/tensor.js').Tensor} pixel_mask
+ * @typedef {import('../../base/image_processors_utils.js').ImageProcessorResult & DetrFeatureExtractorResultProps} DetrFeatureExtractorResult
+ */
+export class DetrImageProcessor 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.
+     * @returns {Promise<DetrFeatureExtractorResult>} An object containing the concatenated pixel values of the preprocessed images.
+     */
+    _call(images: import('../../utils/image.js').RawImage[]): Promise<DetrFeatureExtractorResult>;
+    post_process_object_detection(outputs: {
+        logits: import("../../utils/tensor.js").Tensor;
+        pred_boxes: import("../../utils/tensor.js").Tensor;
+    }, threshold?: number, target_sizes?: [number, number][], is_zero_shot?: boolean): any[];
+    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][]): {
+        segmentation: import("../../utils/tensor.js").Tensor;
+        segments_info: {
+            id: number;
+            label_id: number;
+            score: number;
+        }[];
+    }[];
+    post_process_instance_segmentation(outputs: any, threshold?: number, target_sizes?: [number, number][]): {
+        segmentation: import("../../utils/tensor.js").Tensor;
+        segments_info: {
+            id: number;
+            label_id: number;
+            score: number;
+        }[];
+    }[];
+}
+export class DetrFeatureExtractor extends DetrImageProcessor {
+}
+export type DetrFeatureExtractorResultProps = {
+    pixel_mask: import('../../utils/tensor.js').Tensor;
+};
+export type DetrFeatureExtractorResult = import('../../base/image_processors_utils.js').ImageProcessorResult & DetrFeatureExtractorResultProps;
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_detr.d.ts.map

package/types/models/detr/image_processing_detr.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_detr.d.ts","sourceRoot":"","sources":["../../../src/models/detr/image_processing_detr.js"],"names":[],"mappings":"AAUA;;;;GAIG;AAEH;IACI;;;;;OAKG;IACH,cAHW,OAAO,sBAAsB,EAAE,QAAQ,EAAE,GACvC,QAAQ,0BAA0B,CAAC,CAY/C;IAmBwmB;;;6FAAu6F;IAA8yU;;;;;;;QAA4qG;IAAA;;;;;;;QAAktB;CAH/rjB;AAED;CAAgE;;gBAvClD,OAAO,uBAAuB,EAAE,MAAM;;yCACvC,OAAO,sCAAsC,EAAE,oBAAoB,GAAG,+BAA+B;+BAR3G,sCAAsC"}

package/types/models/donut/image_processing_donut.d.ts

@@ -0,0 +1,7 @@
+export class DonutImageProcessor extends ImageProcessor {
+    pad_image(pixelData: any, imgDims: any, padSize: any, options?: {}): [Float32Array, number[]];
+}
+export class DonutFeatureExtractor extends DonutImageProcessor {
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_donut.d.ts.map

package/types/models/donut/image_processing_donut.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_donut.d.ts","sourceRoot":"","sources":["../../../src/models/donut/image_processing_donut.js"],"names":[],"mappings":"AAIA;IACI,8FAuBC;CACJ;AACD;CAAkE;+BA5B3D,sCAAsC"}

package/types/models/dpt/image_processing_dpt.d.ts

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

package/types/models/dpt/image_processing_dpt.d.ts.map

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

package/types/models/efficientnet/image_processing_efficientnet.d.ts

@@ -0,0 +1,6 @@
+export class EfficientNetImageProcessor extends ImageProcessor {
+    constructor(config: any);
+    include_top: any;
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_efficientnet.d.ts.map

package/types/models/efficientnet/image_processing_efficientnet.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_efficientnet.d.ts","sourceRoot":"","sources":["../../../src/models/efficientnet/image_processing_efficientnet.js"],"names":[],"mappings":"AAIA;IACI,yBAMC;IAJG,iBAAkD;CAKzD;+BAVM,sCAAsC"}

package/types/models/feature_extractors.d.ts

@@ -0,0 +1,10 @@
+export * from "./audio_spectrogram_transformer/feature_extraction_audio_spectrogram_transformer.js";
+export * from "./clap/feature_extraction_clap.js";
+export * from "./pyannote/feature_extraction_pyannote.js";
+export * from "./seamless_m4t/feature_extraction_seamless_m4t.js";
+export * from "./speecht5/feature_extraction_speecht5.js";
+export * from "./wav2vec2/feature_extraction_wav2vec2.js";
+export * from "./wespeaker/feature_extraction_wespeaker.js";
+export * from "./whisper/feature_extraction_whisper.js";
+export { ImageProcessor as ImageFeatureExtractor } from "../base/image_processors_utils.js";
+//# sourceMappingURL=feature_extractors.d.ts.map

package/types/models/feature_extractors.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"feature_extractors.d.ts","sourceRoot":"","sources":["../../src/models/feature_extractors.js"],"names":[],"mappings":""}

package/types/models/florence2/processing_florence2.d.ts

@@ -0,0 +1,39 @@
+export class Florence2Processor extends Processor {
+    static tokenizer_class: typeof AutoTokenizer;
+    static image_processor_class: typeof AutoImageProcessor;
+    constructor(config: any, components: any);
+    /** @type {Map<string, string>} */
+    tasks_answer_post_processing_type: Map<string, string>;
+    /** @type {Map<string, string>} */
+    task_prompts_without_inputs: Map<string, string>;
+    /** @type {Map<string, string>} */
+    task_prompts_with_input: Map<string, string>;
+    regexes: {
+        quad_boxes: RegExp;
+        bboxes: RegExp;
+    };
+    size_per_bin: number;
+    /**
+     * Helper function to construct prompts from input texts
+     * @param {string|string[]} text
+     * @returns {string[]}
+     */
+    construct_prompts(text: string | string[]): string[];
+    /**
+     * Post-process the output of the model to each of the task outputs.
+     * @param {string} text The text to post-process.
+     * @param {string} task The task to post-process the text for.
+     * @param {[number, number]} image_size The size of the image. height x width.
+     */
+    post_process_generation(text: string, task: string, image_size: [number, number]): {
+        [x: string]: string | {
+            [x: string]: any[];
+            labels: any[];
+        };
+    };
+    _call(images: any, 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_florence2.d.ts.map

package/types/models/florence2/processing_florence2.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"processing_florence2.d.ts","sourceRoot":"","sources":["../../../src/models/florence2/processing_florence2.js"],"names":[],"mappings":"AAIA;IACI,6CAAsC;IACtC,wDAAiD;IAEjD,0CAuBC;IAdG,kCAAkC;IAClC,mCADW,IAAI,MAAM,EAAE,MAAM,CAAC,CAC2E;IAEzG,kCAAkC;IAClC,6BADW,IAAI,MAAM,EAAE,MAAM,CAAC,CAC+D;IAE7F,kCAAkC;IAClC,yBADW,IAAI,MAAM,EAAE,MAAM,CAAC,CACuD;IAErF;;;MAGC;IACD,qBAAwB;IAG5B;;;;OAIG;IACH,wBAHW,MAAM,GAAC,MAAM,EAAE,GACb,MAAM,EAAE,CA6BpB;IAED;;;;;OAKG;IACH,8BAJW,MAAM,QACN,MAAM,cACN,CAAC,MAAM,EAAE,MAAM,CAAC;;;;;MAsC1B;IAID,0DAaC;CACJ;0BA/HyB,gCAAgC;8BAE5B,qBAAqB;mCADhB,kCAAkC"}

package/types/models/glpn/image_processing_glpn.d.ts

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

package/types/models/glpn/image_processing_glpn.d.ts.map

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

package/types/models/idefics3/image_processing_idefics3.d.ts

@@ -0,0 +1,40 @@
+export class Idefics3ImageProcessor extends ImageProcessor {
+    constructor(config: any);
+    do_image_splitting: any;
+    max_image_size: any;
+    /**
+     * @typedef {import('../../utils/image.js').RawImage} RawImage
+     * @typedef {import('../../utils/tensor.js').Tensor} Tensor
+     */
+    /**
+     * Calculate size to resize images to, to be multiples of `vision_encoder_max_size` while preserving the aspect ratio.
+     * @param {Tensor} pixel_values Tensor of the image to resize.
+     * @param {number} vision_encoder_max_size Maximum size of the output image. If the image is larger than this size,
+     * it will be split into patches of this size, and the original image will be concatenated with the patches, resized to max_size.
+     */
+    get_resize_for_vision_encoder(pixel_values: import("../../utils/tensor.js").Tensor, vision_encoder_max_size: number): {
+        height: number;
+        width: number;
+    };
+    /** @param {RawImage|RawImage[]|RawImage[][]} images */
+    _call(images: import("../../utils/image.js").RawImage | import("../../utils/image.js").RawImage[] | import("../../utils/image.js").RawImage[][], { do_image_splitting, return_row_col_info, }?: {
+        do_image_splitting?: any;
+        return_row_col_info?: boolean;
+    }): Promise<{
+        rows?: any[][];
+        cols?: any[][];
+        pixel_values: import("../../utils/tensor.js").Tensor;
+        pixel_attention_mask: import("../../utils/tensor.js").Tensor;
+        original_sizes: import("../../base/image_processors_utils.js").HeightWidth[];
+        reshaped_input_sizes: import("../../base/image_processors_utils.js").HeightWidth[];
+    }>;
+    split_image(pixel_values: any, { longest_edge }: {
+        longest_edge: any;
+    }): Promise<{
+        frames: any[];
+        num_splits_h: number;
+        num_splits_w: number;
+    }>;
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_idefics3.d.ts.map

package/types/models/idefics3/image_processing_idefics3.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_idefics3.d.ts","sourceRoot":"","sources":["../../../src/models/idefics3/image_processing_idefics3.js"],"names":[],"mappings":"AAOA;IACI,yBAKC;IAFG,wBAA2D;IAC3D,oBAA2C;IAG/C;;;OAGG;IAEH;;;;;OAKG;IACH,6GAHW,MAAM;;;MAiBhB;IAED,uDAAuD;IACvD;;;;;;;;;;OA2HC;IAED;;;;;;OAiDC;CACJ;+BAtNM,sCAAsC"}

package/types/models/idefics3/processing_idefics3.d.ts

@@ -0,0 +1,19 @@
+export class Idefics3Processor extends Processor {
+    static image_processor_class: typeof AutoImageProcessor;
+    static tokenizer_class: typeof AutoTokenizer;
+    fake_image_token: string;
+    image_token: string;
+    global_img_token: string;
+    /**
+     *
+     * @param {string|string[]} text
+     * @param {RawImage|RawImage[]|RawImage[][]} images
+     * @returns {Promise<any>}
+     */
+    _call(text: string | string[], images?: RawImage | RawImage[] | RawImage[][], options?: {}): 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_idefics3.d.ts.map

package/types/models/idefics3/processing_idefics3.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"processing_idefics3.d.ts","sourceRoot":"","sources":["../../../src/models/idefics3/processing_idefics3.js"],"names":[],"mappings":"AA6DA;IACI,wDAAiD;IACjD,6CAAsC;IAGtC,yBAA+C;IAC/C,oBAAwB;IACxB,yBAAkC;IAElC;;;;;OAKG;IACH,YAJW,MAAM,GAAC,MAAM,EAAE,WACf,QAAQ,GAAC,QAAQ,EAAE,GAAC,QAAQ,EAAE,EAAE,iBAC9B,QAAQ,GAAG,CAAC,CA4DxB;CACJ;0BAtIyB,gCAAgC;yBAGjC,sBAAsB;mCAFZ,kCAAkC;8BACvC,qBAAqB"}

package/types/models/image_processors.d.ts

@@ -0,0 +1,37 @@
+export * from "./beit/image_processing_beit.js";
+export * from "./bit/image_processing_bit.js";
+export * from "./chinese_clip/image_processing_chinese_clip.js";
+export * from "./clip/image_processing_clip.js";
+export * from "./convnext/image_processing_convnext.js";
+export * from "./deit/image_processing_deit.js";
+export * from "./detr/image_processing_detr.js";
+export * from "./donut/image_processing_donut.js";
+export * from "./dpt/image_processing_dpt.js";
+export * from "./efficientnet/image_processing_efficientnet.js";
+export * from "./glpn/image_processing_glpn.js";
+export * from "./idefics3/image_processing_idefics3.js";
+export * from "./janus/image_processing_janus.js";
+export * from "./jina_clip/image_processing_jina_clip.js";
+export * from "./llava_onevision/image_processing_llava_onevision.js";
+export * from "./mask2former/image_processing_mask2former.js";
+export * from "./maskformer/image_processing_maskformer.js";
+export * from "./mobilenet_v1/image_processing_mobilenet_v1.js";
+export * from "./mobilenet_v2/image_processing_mobilenet_v2.js";
+export * from "./mobilenet_v3/image_processing_mobilenet_v3.js";
+export * from "./mobilenet_v4/image_processing_mobilenet_v4.js";
+export * from "./mobilevit/image_processing_mobilevit.js";
+export * from "./nougat/image_processing_nougat.js";
+export * from "./owlv2/image_processing_owlv2.js";
+export * from "./owlvit/image_processing_owlvit.js";
+export * from "./pvt/image_processing_pvt.js";
+export * from "./qwen2_vl/image_processing_qwen2_vl.js";
+export * from "./rt_detr/image_processing_rt_detr.js";
+export * from "./sam/image_processing_sam.js";
+export * from "./segformer/image_processing_segformer.js";
+export * from "./siglip/image_processing_siglip.js";
+export * from "./swin2sr/image_processing_swin2sr.js";
+export * from "./vit/image_processing_vit.js";
+export * from "./vitmatte/image_processing_vitmatte.js";
+export * from "./vitpose/image_processing_vitpose.js";
+export * from "./yolos/image_processing_yolos.js";
+//# sourceMappingURL=image_processors.d.ts.map

package/types/models/image_processors.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processors.d.ts","sourceRoot":"","sources":["../../src/models/image_processors.js"],"names":[],"mappings":""}

package/types/models/janus/image_processing_janus.d.ts

@@ -0,0 +1,7 @@
+export class VLMImageProcessor extends ImageProcessor {
+    constructor(config: any);
+    constant_values: any;
+    pad_image(pixelData: any, imgDims: any, padSize: any, options: any): [Float32Array, number[]];
+}
+import { ImageProcessor } from "../../base/image_processors_utils.js";
+//# sourceMappingURL=image_processing_janus.d.ts.map

package/types/models/janus/image_processing_janus.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"image_processing_janus.d.ts","sourceRoot":"","sources":["../../../src/models/janus/image_processing_janus.js"],"names":[],"mappings":"AAKA;IACI,yBAUC;IADG,qBAAqF;IAGzF,8FAMC;CACJ;+BAtBM,sCAAsC"}