processing_qwen2_mm.py

# coding=utf-8
# Copyright 2024 The Qwen team, Alibaba Group 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.
"""
Processor class for Qwen2-VL.
"""

from typing import List, Union

from transformers.image_processing_utils import BatchFeature
from transformers.image_utils import ImageInput, VideoInput
from transformers.processing_utils import ProcessorMixin, ProcessingKwargs, Unpack
from transformers.tokenization_utils_base import PaddingStrategy, PreTokenizedInput, TextInput, TruncationStrategy, AudioInput
from transformers.utils import TensorType, requires_backends, is_torch_dtype, is_torch_device, logging

import torch
logger = logging.get_logger(__name__)


class Qwen2VLProcessorKwargs(ProcessingKwargs, total=False):
    _defaults = {
        "text_kwargs": {
            "padding": False,
        },
    }


class Qwen2MMProcessor(ProcessorMixin):
    r"""
    Constructs a Qwen2-VL processor which wraps a Qwen2-VL image processor and a Qwen2 tokenizer into a single processor.
    [`Qwen2VLProcessor`] offers all the functionalities of [`Qwen2VLImageProcessor`] and [`Qwen2TokenizerFast`]. See the
    [`~Qwen2VLProcessor.__call__`] and [`~Qwen2VLProcessor.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", "audio_processor"]
    valid_kwargs = ["chat_template"]
    image_processor_class = "Qwen2VLImageProcessor"
    audio_processor_class = "SeamlessM4TFeatureExtractor"
    tokenizer_class = ("Qwen2Tokenizer", "Qwen2TokenizerFast")

    def __init__(self, image_processor=None, tokenizer=None, audio_processor=None, chat_template=None, **kwargs):
        super().__init__(image_processor, tokenizer, audio_processor, chat_template=chat_template)
        #print("image_processor: ", image_processor)
        #print("audio_processor: ", audio_processor)
        

    def __call__(
        self,
        images: ImageInput = None,
        text: Union[TextInput, PreTokenizedInput, List[TextInput], List[PreTokenizedInput]] = None,
        videos: VideoInput = None,
        audios: AudioInput = None,
        **kwargs: Unpack[Qwen2VLProcessorKwargs],
    ) -> 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`.
        """
        output_kwargs = self._merge_kwargs(
            Qwen2VLProcessorKwargs,
            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["videos_kwargs"])
            video_grid_thw = videos_inputs["video_grid_thw"]
        else:
            videos_inputs = {}
            video_grid_thw = None

        if audios is not None:
            print("audios: ", audios)
            audio_inputs = self.audio_processor(audios, sampling_rate=16000, return_tensors="pt")
            audio_grid_thw = torch.tensor([torch.sum(attention_mask == 1).item() // 8 + 1 for attention_mask in audio_inputs["attention_mask"]])
            audio_inputs = {"audio_values": audio_inputs["input_features"], "audio_attention_mask": audio_inputs["attention_mask"], "audio_grid_thw": audio_grid_thw}

        else:
            audio_inputs = {}
            audio_grid_thw = None

        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 "<|image_pad|>" in text[i]:
                    text[i] = text[i].replace(
                        "<|image_pad|>", "<|placeholder|>" * (image_grid_thw[index].prod() // merge_length), 1
                    )
                    index += 1
                text[i] = text[i].replace("<|placeholder|>", "<|image_pad|>")

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

        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)

    @property
    def model_input_names(self):
        tokenizer_input_names = self.tokenizer.model_input_names
        image_processor_input_names = self.image_processor.model_input_names
        audio_processor_input_names = self.audio_processor.model_input_names
        return list(dict.fromkeys(tokenizer_input_names + image_processor_input_names))