mmpretrain_custom/datasets/transforms/loading.py

import warnings
from typing import Optional

import mmengine.fileio as fileio
import numpy as np
import cv2

import mmcv
from mmcv.transforms.base import BaseTransform
from mmpretrain.registry import TRANSFORMS

from mmcv.image.io import _jpegflag, imread_flags
from mmengine.utils import is_str


@TRANSFORMS.register_module()
class LoadImageFromFileFungi(BaseTransform):
    """Load an image from file.

    Required Keys:

    - img_path

    Modified Keys:

    - img
    - img_shape
    - ori_shape

    Args:
        to_float32 (bool): Whether to convert the loaded image to a float32
            numpy array. If set to False, the loaded image is an uint8 array.
            Defaults to False.
        color_type (str): The flag argument for :func:`mmcv.imfrombytes`.
            Defaults to 'color'.
        imdecode_backend (str): The image decoding backend type. The backend
            argument for :func:`mmcv.imfrombytes`.
            See :func:`mmcv.imfrombytes` for details.
            Defaults to 'cv2'.
        file_client_args (dict, optional): Arguments to instantiate a
            FileClient. See :class:`mmengine.fileio.FileClient` for details.
            Defaults to None. It will be deprecated in future. Please use
            ``backend_args`` instead.
            Deprecated in version 2.0.0rc4.
        ignore_empty (bool): Whether to allow loading empty image or file path
            not existent. Defaults to False.
        backend_args (dict, optional): Instantiates the corresponding file
            backend. It may contain `backend` key to specify the file
            backend. If it contains, the file backend corresponding to this
            value will be used and initialized with the remaining values,
            otherwise the corresponding file backend will be selected
            based on the prefix of the file path. Defaults to None.
            New in version 2.0.0rc4.
    """

    def __init__(self,
                 to_float32: bool = False,
                 color_type: str = 'color',
                 imdecode_backend: str = 'cv2',
                 file_client_args: Optional[dict] = None,
                 ignore_empty: bool = False,
                 *,
                 backend_args: Optional[dict] = None) -> None:
        print('Using LoadImageFromFile from mmpretrain_custom')
        self.ignore_empty = ignore_empty
        self.to_float32 = to_float32
        self.color_type = color_type
        self.imdecode_backend = imdecode_backend

        self.file_client_args: Optional[dict] = None
        self.backend_args: Optional[dict] = None
        if file_client_args is not None:
            warnings.warn(
                '"file_client_args" will be deprecated in future. '
                'Please use "backend_args" instead', DeprecationWarning)
            if backend_args is not None:
                raise ValueError(
                    '"file_client_args" and "backend_args" cannot be set '
                    'at the same time.')

            self.file_client_args = file_client_args.copy()
        if backend_args is not None:
            self.backend_args = backend_args.copy()

    def transform(self, results: dict) -> Optional[dict]:
        """Functions to load image.

        Args:
            results (dict): Result dict from
                :class:`mmengine.dataset.BaseDataset`.

        Returns:
            dict: The dict contains loaded image and meta information.
        """

        filename = results['img_path']
        try:
            if self.file_client_args is not None:
                file_client = fileio.FileClient.infer_client(
                    self.file_client_args, filename)
                img_bytes = file_client.get(filename)
            else:
                img_bytes = fileio.get(
                    filename, backend_args=self.backend_args)
            img = mmcv.imfrombytes(
                img_bytes, flag=self.color_type, backend=self.imdecode_backend)
            if img is None:
                flag = self.color_type
                flag = imread_flags[flag] if is_str(flag) else flag
                img = cv2.imread(filename, flag)
            assert img is not None, filename
        except Exception as e:
            if self.ignore_empty:
                if img is None:
                    flag = self.color_type
                    flag = imread_flags[flag] if is_str(flag) else flag
                    img = cv2.imread(filename, flag)
                assert img is not None, filename
                return img
            else:
                raise e
        # in some cases, images are not read successfully, the img would be
        # `None`, refer to https://github.com/open-mmlab/mmpretrain/issues/1427
        assert img is not None, f'failed to load image: {filename}'
        if self.to_float32:
            img = img.astype(np.float32)

        results['img'] = img
        results['img_shape'] = img.shape[:2]
        results['ori_shape'] = img.shape[:2]
        return results

    def __repr__(self):
        repr_str = (f'{self.__class__.__name__}('
                    f'ignore_empty={self.ignore_empty}, '
                    f'to_float32={self.to_float32}, '
                    f"color_type='{self.color_type}', "
                    f"imdecode_backend='{self.imdecode_backend}', ")

        if self.file_client_args is not None:
            repr_str += f'file_client_args={self.file_client_args})'
        else:
            repr_str += f'backend_args={self.backend_args})'

        return repr_str