processing_qwen2_5_vl_audio.py

#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
#           This file was automatically generated from src/transformers/models/qwen2_5_vl/modular_qwen2_5_vl.py.
#               Do NOT edit this file manually as any edits will be overwritten by the generation of
#             the file from the modular. If any change should be done, please apply the change to the
#                          modular_qwen2_5_vl.py file directly. One of our CI enforces this.
#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
# coding=utf-8
# Copyright 2025 The Qwen Team and The HuggingFace Inc. team. All rights reserved.
#
# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
# and OPT implementations in this library. It has been modified from its
# original forms to accommodate minor architectural differences compared
# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from typing import List, Union, Optional

from transformers.feature_extraction_utils import BatchFeature
from transformers.image_utils import ImageInput, VideoInput
from transformers.processing_utils import ProcessingKwargs, ProcessorMixin, Unpack, VideosKwargs
from transformers.tokenization_utils_base import PreTokenizedInput, TextInput
import numpy as np


class Qwen2_5_VLVideosProcessorKwargs(VideosKwargs, total=False):
    fps: Union[List[float], float]


class Qwen2_5_VLProcessorKwargs(ProcessingKwargs, total=False):
    videos_kwargs: Qwen2_5_VLVideosProcessorKwargs
    _defaults = {
        "text_kwargs": {
            "padding": False,
        },
        "videos_kwargs": {"fps": 2.0},
    }


class Qwen2_5_VL_Audio_Processor(ProcessorMixin):
    r"""
    Constructs a Qwen2.5-VL processor which wraps a Qwen2.5-VL image processor and a Qwen2 tokenizer into a single processor.
    [`Qwen2_5_VLProcessor`] offers all the functionalities of [`Qwen2VLImageProcessor`] and [`Qwen2TokenizerFast`]. See the
    [`~Qwen2_5_VLProcessor.__call__`] and [`~Qwen2_5_VLProcessor.decode`] for more information.
    Args:
        image_processor ([`Qwen2VLImageProcessor`], *optional*):
            The image processor is a required input.
        tokenizer ([`Qwen2TokenizerFast`], *optional*):
            The tokenizer is a required input.
        chat_template (`str`, *optional*): A Jinja template which will be used to convert lists of messages
            in a chat into a tokenizable string.
    """

    attributes = ["image_processor", "tokenizer","feature_extractor"]
    valid_kwargs = ["chat_template"]
    feature_extractor_class = "WhisperFeatureExtractor"

    image_processor_class = "AutoImageProcessor"
    tokenizer_class = ("Qwen2Tokenizer", "Qwen2TokenizerFast")

    def __init__(self, image_processor=None, feature_extractor=None, tokenizer=None, chat_template=None, **kwargs):
    # def __init__(self, image_processor=None, tokenizer=None, **kwargs):
        # print(kwargs['chat_template'])
        self.image_token = "<|image_pad|>" if not hasattr(tokenizer, "image_token") else tokenizer.image_token
        self.video_token = "<|video_pad|>" if not hasattr(tokenizer, "video_token") else tokenizer.video_token
        self.audio_token = tokenizer.audio_token if hasattr(tokenizer, "audio_token") else "<|AUDIO|>"
        self.audio_bos_token = tokenizer.audio_bos_token if hasattr(tokenizer, "audio_bos_token") else "<|audio_bos|>"
        self.audio_eos_token = tokenizer.audio_eos_token if hasattr(tokenizer, "audio_eos_token") else "<|audio_eos|>"
        super().__init__(image_processor, feature_extractor, tokenizer, chat_template=chat_template)

    def __call__(
        self,
        images: ImageInput = None,
        text: Union[TextInput, PreTokenizedInput, List[TextInput], List[PreTokenizedInput]] = None,
        videos: VideoInput = None,
        audios: Union[np.ndarray, List[np.ndarray]] = None,
        sampling_rate: Optional[int] = None,
        **kwargs: Unpack[Qwen2_5_VLProcessorKwargs],
    ) -> BatchFeature:
        """
        Main method to prepare for the model one or several sequences(s) and image(s). This method forwards the `text`
        and `kwargs` arguments to Qwen2TokenizerFast's [`~Qwen2TokenizerFast.__call__`] if `text` is not `None` to encode
        the text. To prepare the vision inputs, this method forwards the `vision_infos` and `kwrags` arguments to
        Qwen2VLImageProcessor's [`~Qwen2VLImageProcessor.__call__`] if `vision_infos` is not `None`.

        Args:
            images (`PIL.Image.Image`, `np.ndarray`, `torch.Tensor`, `List[PIL.Image.Image]`, `List[np.ndarray]`, `List[torch.Tensor]`):
                The image or batch of images to be prepared. Each image can be a PIL image, NumPy array or PyTorch
                tensor. Both channels-first and channels-last formats are supported.
            text (`str`, `List[str]`, `List[List[str]]`):
                The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings
                (pretokenized string). If the sequences are provided as list of strings (pretokenized), you must set
                `is_split_into_words=True` (to lift the ambiguity with a batch of sequences).
            videos (`np.ndarray`, `torch.Tensor`, `List[np.ndarray]`, `List[torch.Tensor]`):
                The image or batch of videos to be prepared. Each video can be a 4D NumPy array or PyTorch
                tensor, or a nested list of 3D frames. Both channels-first and channels-last formats are supported.
            return_tensors (`str` or [`~utils.TensorType`], *optional*):
                If set, will return tensors of a particular framework. Acceptable values are:
                - `'tf'`: Return TensorFlow `tf.constant` objects.
                - `'pt'`: Return PyTorch `torch.Tensor` objects.
                - `'np'`: Return NumPy `np.ndarray` objects.
                - `'jax'`: Return JAX `jnp.ndarray` objects.

        Returns:
            [`BatchFeature`]: A [`BatchFeature`] with the following fields:

            - **input_ids** -- List of token ids to be fed to a model. Returned when `text` is not `None`.
            - **attention_mask** -- List of indices specifying which tokens should be attended to by the model (when
              `return_attention_mask=True` or if *"attention_mask"* is in `self.model_input_names` and if `text` is not
              `None`).
            - **pixel_values** -- Pixel values to be fed to a model. Returned when `images` is not `None`.
            - **pixel_values_videos** -- Pixel values of videos to be fed to a model. Returned when `videos` is not `None`.
            - **image_grid_thw** -- List of image 3D grid in LLM. Returned when `images` is not `None`.
            - **video_grid_thw** -- List of video 3D grid in LLM. Returned when `videos` is not `None`.
            - **second_per_grid_ts** -- List of video seconds per time grid. Returned when `videos` is not `None`.
        """
        output_kwargs = self._merge_kwargs(
            Qwen2_5_VLProcessorKwargs,
            tokenizer_init_kwargs=self.tokenizer.init_kwargs,
            **kwargs,
        )
        if images is not None:
            image_inputs = self.image_processor(images=images, videos=None, **output_kwargs["images_kwargs"])
            image_grid_thw = image_inputs["image_grid_thw"]
        else:
            image_inputs = {}
            image_grid_thw = None

        if videos is not None:
            videos_inputs = self.image_processor(images=None, videos=videos, **output_kwargs["images_kwargs"])
            video_grid_thw = videos_inputs["video_grid_thw"]

            fps = output_kwargs["videos_kwargs"].pop("fps", 2.0)
            if isinstance(fps, (int, float)):
                second_per_grid_ts = [self.image_processor.temporal_patch_size / fps] * len(video_grid_thw)
            elif hasattr(fps, "__len__") and len(fps) == len(video_grid_thw):
                second_per_grid_ts = [self.image_processor.temporal_patch_size / tmp for tmp in fps]
            else:
                raise ValueError(
                    f"The length of fps ({len(fps) if hasattr(fps, '__len__') else fps}) must be equal to the length of video_grid_thw ({len(video_grid_thw)}) or fps should be a single number."
                )
            videos_inputs.update({"second_per_grid_ts": second_per_grid_ts})

        else:
            videos_inputs = {}
            video_grid_thw = None

        if audios is not None:
            new_kwargs = {k: kwargs[k] for k in  kwargs if k not in ['padding', 'truncation','max_length']}

            audio_inputs = self.feature_extractor(
                audios, sampling_rate=sampling_rate, return_attention_mask=True, padding="max_length", **new_kwargs
            )

            audio_inputs["feature_attention_mask"] = audio_inputs.pop(
                "attention_mask"
            )  # rename attention_mask to prevent conflicts later on

            expanded_text = []
            audio_lengths = audio_inputs["feature_attention_mask"].sum(-1).tolist()

            for sample in text:
                replace_str = []
                while self.audio_token in sample:
                    audio_length = audio_lengths.pop(0)
                    input_length = (audio_length - 1) // 2 + 1
                    num_audio_tokens = (input_length - 2) // 2 + 1

                    expanded_audio_token = self.audio_token * num_audio_tokens

                    audio_token_start_idx = sample.find(self.audio_token)
                    audio_token_end_idx = audio_token_start_idx + len(self.audio_token)

                    has_bos = (
                        sample[audio_token_start_idx - len(self.audio_bos_token) : audio_token_start_idx]
                        == self.audio_bos_token
                    )
                    has_eos = (
                        sample[audio_token_end_idx : audio_token_end_idx + len(self.audio_eos_token)]
                        == self.audio_eos_token
                    )

                    # Check if this audio token is surrounded by bos/eos tokens
                    if not has_bos and not has_eos:
                        expanded_audio_token = self.audio_bos_token + expanded_audio_token + self.audio_eos_token

                    replace_str.append(expanded_audio_token)
                    sample = sample.replace(self.audio_token, "<placeholder>", 1)

                while "<placeholder>" in sample:
                    sample = sample.replace("<placeholder>", replace_str.pop(0), 1)
                expanded_text.append(sample)
            text = expanded_text
        else:
            audio_inputs = {}

        if not isinstance(text, list):
            text = [text]

        if image_grid_thw is not None:
            merge_length = self.image_processor.merge_size**2
            index = 0
            for i in range(len(text)):
                while self.image_token in text[i]:
                    text[i] = text[i].replace(
                        self.image_token,
                        "<|placeholder|>" * (image_grid_thw[index].prod() // merge_length),
                        1,
                    )
                    index += 1
                text[i] = text[i].replace("<|placeholder|>", self.image_token)

        if video_grid_thw is not None:
            merge_length = self.image_processor.merge_size**2
            index = 0
            for i in range(len(text)):
                while self.video_token in text[i]:
                    text[i] = text[i].replace(
                        self.video_token,
                        "<|placeholder|>" * (video_grid_thw[index].prod() // merge_length),
                        1,
                    )
                    index += 1
                text[i] = text[i].replace("<|placeholder|>", self.video_token)

        text_inputs = self.tokenizer(text, **output_kwargs["text_kwargs"])

        return BatchFeature(data={**text_inputs, **image_inputs, **videos_inputs, **audio_inputs})

    def batch_decode(self, *args, **kwargs):
        """
        This method forwards all its arguments to Qwen2TokenizerFast's [`~PreTrainedTokenizer.batch_decode`]. Please
        refer to the docstring of this method for more information.
        """
        return self.tokenizer.batch_decode(*args, **kwargs)

    def decode(self, *args, **kwargs):
        """
        This method forwards all its arguments to Qwen2TokenizerFast's [`~PreTrainedTokenizer.decode`]. Please refer to
        the docstring of this method for more information.
        """
        return self.tokenizer.decode(*args, **kwargs)

    def post_process_image_text_to_text(
        self, generated_outputs, skip_special_tokens=True, clean_up_tokenization_spaces=False, **kwargs
    ):
        """
        Post-process the output of the model to decode the text.

        Args:
            generated_outputs (`torch.Tensor` or `np.ndarray`):
                The output of the model `generate` function. The output is expected to be a tensor of shape `(batch_size, sequence_length)`
                or `(sequence_length,)`.
            skip_special_tokens (`bool`, *optional*, defaults to `True`):
                Whether or not to remove special tokens in the output. Argument passed to the tokenizer's `batch_decode` method.
            Clean_up_tokenization_spaces (`bool`, *optional*, defaults to `False`):
                Whether or not to clean up the tokenization spaces. Argument passed to the tokenizer's `batch_decode` method.
            **kwargs:
                Additional arguments to be passed to the tokenizer's `batch_decode method`.

        Returns:
            `List[str]`: The decoded text.
        """
        return self.tokenizer.batch_decode(
            generated_outputs,
            skip_special_tokens=skip_special_tokens,
            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
            **kwargs,
        )

    @property
    def model_input_names(self):
        tokenizer_input_names = self.tokenizer.model_input_names
        image_processor_input_names = self.image_processor.model_input_names
        names_from_processor = list(dict.fromkeys(tokenizer_input_names + image_processor_input_names))
        return names_from_processor + ["second_per_grid_ts"]


__all__ = ["Qwen2_5_VL_Audio_Processor"]