【Pytorch】自定义DataLoader相关分析

秒速五厘米 2022-09-11 08:24 378阅读 0赞

[1.三步法写自定义Torch的DataLoader - 知乎][1._Torch_DataLoader -]

[2.pytorch Dataset, DataLoader产生自定义的训练数据\_pan\_jinquan的博客-CSDN博客][2.pytorch Dataset_ DataLoader_pan_jinquan_-CSDN]

[3.GitHub - JDAI-CV/fast-reid: SOTA Re-identification Methods and Toolbox][3.GitHub - JDAI-CV_fast-reid_ SOTA Re-identification Methods and Toolbox]

[4.https://github.com/chenyuntc/simple-faster-rcnn-pytorch][4.https_github.com_chenyuntc_simple-faster-rcnn-pytorch]

第一个链接，会给出一个大体的框架思路，第二个链接，是一个写得比较好的且完整的例子，第三个链接，是reid的自定义dataloader的实例，第四个链接，是目标检测自定义的VOC dataloader的实例。

# 1.以三步走框架分析Market1501自定义的dataloader #

## 1）loaddata函数 ##

从知乎链接1上的来看，分类任务中这个函数，主要是返回图像的路径、训练所需要的label信息等。

[https://github.com/JDAI-CV/fast-reid/blob/master/fastreid/data/datasets/market1501.py![icon-default.png_t_L892][]https://github.com/JDAI-CV/fast-reid/blob/master/fastreid/data/datasets/market1501.py][https_github.com_JDAI-CV_fast-reid_blob_master_fastreid_data_datasets_market1501.py_icon-default.png_t_L892_https_github.com_JDAI-CV_fast-reid_blob_master_fastreid_data_datasets_market1501.py]

分析下面的代码，由于ReID训练过程中，除了图像路径信息外，还需要对应的id与camera信息。类Market1501中的方法，初始化后，就会返回train、query、gallery等的图像路径信息，对应的id与camera信息。

# encoding: utf-8
    """
    @author:  sherlock
    @contact: sherlockliao01@gmail.com
    """
    
    import glob
    import os.path as osp
    import re
    import warnings
    
    from .bases import ImageDataset
    from ..datasets import DATASET_REGISTRY
    
    
    @DATASET_REGISTRY.register()
    class Market1501(ImageDataset):
        """Market1501.
        Reference:
            Zheng et al. Scalable Person Re-identification: A Benchmark. ICCV 2015.
        URL: `<http://www.liangzheng.org/Project/project_reid.html>`_
        Dataset statistics:
            - identities: 1501 (+1 for background).
            - images: 12936 (train) + 3368 (query) + 15913 (gallery).
        """
        _junk_pids = [0, -1]
        dataset_dir = ''
        dataset_url = 'http://188.138.127.15:81/Datasets/Market-1501-v15.09.15.zip'
        dataset_name = "market1501"
    
        def __init__(self, root='datasets', market1501_500k=False, **kwargs):
            # self.root = osp.abspath(osp.expanduser(root))
            self.root = root
            self.dataset_dir = osp.join(self.root, self.dataset_dir)
    
            # allow alternative directory structure
            self.data_dir = self.dataset_dir
            data_dir = osp.join(self.data_dir, 'Market-1501-v15.09.15')
            if osp.isdir(data_dir):
                self.data_dir = data_dir
            else:
                warnings.warn('The current data structure is deprecated. Please '
                              'put data folders such as "bounding_box_train" under '
                              '"Market-1501-v15.09.15".')
    
            self.train_dir = osp.join(self.data_dir, 'bounding_box_train')
            self.query_dir = osp.join(self.data_dir, 'query')
            self.gallery_dir = osp.join(self.data_dir, 'bounding_box_test')
            self.extra_gallery_dir = osp.join(self.data_dir, 'images')
            self.market1501_500k = market1501_500k
    
            required_files = [
                self.data_dir,
                self.train_dir,
                self.query_dir,
                self.gallery_dir,
            ]
            if self.market1501_500k:
                required_files.append(self.extra_gallery_dir)
            self.check_before_run(required_files)
    
            train = lambda: self.process_dir(self.train_dir)
            query = lambda: self.process_dir(self.query_dir, is_train=False)
            gallery = lambda: self.process_dir(self.gallery_dir, is_train=False) + \
                              (self.process_dir(self.extra_gallery_dir, is_train=False) if self.market1501_500k else [])
    
            super(Market1501, self).__init__(train, query, gallery, **kwargs)
    
        def process_dir(self, dir_path, is_train=True):
            img_paths = glob.glob(osp.join(dir_path, '*.jpg'))
            pattern = re.compile(r'([-\d]+)_c(\d)')
    
            data = []
            for img_path in img_paths:
                pid, camid = map(int, pattern.search(img_path).groups())
                if pid == -1:
                    continue  # junk images are just ignored
                assert 0 <= pid <= 1501  # pid == 0 means background
                assert 1 <= camid <= 6
                camid -= 1  # index starts from 0
                if is_train:
                    pid = self.dataset_name + "_" + str(pid)
                    camid = self.dataset_name + "_" + str(camid)
                data.append((img_path, pid, camid))
    
            return data

## 2）继承Dataset，重写初始化函数，定义自己的Dataset ##

从知乎链接1上的来看，可以看到这步会读取图片，归一化，并将图片转化为NCHW（pytorch的默认输入数据结构顺序）。

\*\*可能需要分析的是[\_\_getitem\_\_函数][getitem]：在python中\_\_getitem\_\_(self,key)方法被称为魔法方法，这个方法返回所给键对应的值。当对象是序列时，键是整数。当对象是映射时（字典），键是任意值。

ReID中，也类似，这个地方加了一个transformer。img = [read\_image][read_image](img\_path)。\_\_getitem\_\_只需要给输入图像的index就可以得到：处理好的图像、id、camera信息、图像路径。

[https://github.com/JDAI-CV/fast-reid/blob/master/fastreid/data/common.py![icon-default.png_t_L892][]https://github.com/JDAI-CV/fast-reid/blob/master/fastreid/data/common.py][https_github.com_JDAI-CV_fast-reid_blob_master_fastreid_data_common.py_icon-default.png_t_L892_https_github.com_JDAI-CV_fast-reid_blob_master_fastreid_data_common.py]

# encoding: utf-8
    """
    @author:  liaoxingyu
    @contact: sherlockliao01@gmail.com
    """
    
    from torch.utils.data import Dataset
    
    from .data_utils import read_image
    
    
    class CommDataset(Dataset):
        """Image Person ReID Dataset"""
    
        def __init__(self, img_items, transform=None, relabel=True):
            self.img_items = img_items
            self.transform = transform
            self.relabel = relabel
    
            pid_set = set()
            cam_set = set()
            for i in img_items:
                pid_set.add(i[1])
                cam_set.add(i[2])
    
            self.pids = sorted(list(pid_set))
            self.cams = sorted(list(cam_set))
            if relabel:
                self.pid_dict = dict([(p, i) for i, p in enumerate(self.pids)])
                self.cam_dict = dict([(p, i) for i, p in enumerate(self.cams)])
    
        def __len__(self):
            return len(self.img_items)
    
        def __getitem__(self, index):
            img_item = self.img_items[index]
            img_path = img_item[0]
            pid = img_item[1]
            camid = img_item[2]
            img = read_image(img_path)
            if self.transform is not None: img = self.transform(img)
            if self.relabel:
                pid = self.pid_dict[pid]
                camid = self.cam_dict[camid]
            return {
                "images": img,
                "targets": pid,
                "camids": camid,
                "img_paths": img_path,
            }
    
        @property
        def num_classes(self):
            return len(self.pids)
    
        @property
        def num_cameras(self):
            return len(self.cams)

## 3）DataLoader把自定义Dataset包起来 ##

创建一个CommDataset的实例，使用DataLoader进行调用。此处代码高度集成，所以参考着看看：

[fast-reid/build.py at f4551a128ba17ef201301ccf3c986edae014cabd · JDAI-CV/fast-reid · GitHub![icon-default.png_t_L892][]https://github.com/JDAI-CV/fast-reid/blob/f4551a128ba17ef201301ccf3c986edae014cabd/fastreid/data/build.py\#L73][fast-reid_build.py at f4551a128ba17ef201301ccf3c986edae014cabd _ JDAI-CV_fast-reid _ GitHub_icon-default.png_t_L892_https_github.com_JDAI-CV_fast-reid_blob_f4551a128ba17ef201301ccf3c986edae014cabd_fastreid_data_build.py_L73]

# line 41
    train_set = CommDataset(train_items, transforms, relabel=True)
    
    
    # line 72-
    @configurable(from_config=_train_loader_from_config)
    def build_reid_train_loader(
            train_set, *, sampler=None, total_batch_size, num_workers=0,
    ):
        """
        Build a dataloader for object re-identification with some default features.
        This interface is experimental.
        Returns:
            torch.utils.data.DataLoader: a dataloader.
        """
    
        mini_batch_size = total_batch_size // comm.get_world_size()
    
        batch_sampler = torch.utils.data.sampler.BatchSampler(sampler, mini_batch_size, True)
    
        train_loader = DataLoaderX(
            comm.get_local_rank(),
            dataset=train_set,
            num_workers=num_workers,
            batch_sampler=batch_sampler,
            collate_fn=fast_batch_collator,
            pin_memory=True,
        )
    
        return train_loader

# 2.VOC自定义的dataloader #

参考三步走来说明下。

[https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/data/voc\_dataset.py![icon-default.png_t_L892][]https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/data/voc\_dataset.py][https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_data_voc_dataset.py_icon-default.png_t_L892_https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_data_voc_dataset.py]

[https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/data/util.py\#:~:text=def%20read\_image(path%2C%20dtype%3Dnp.float32%2C%20color%3DTrue)%3A![icon-default.png_t_L892][]https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/data/util.py\#:~:text=def%20read\_image(path%2C%20dtype%3Dnp.float32%2C%20color%3DTrue)%3A][https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_data_util.py_text_def_20read_image_path_2C_20dtype_3Dnp.float32_2C_20color_3DTrue_3A_icon-default.png_t_L892_https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_data_util.py_text_def_20read_image_path_2C_20dtype_3Dnp.float32_2C_20color_3DTrue_3A]

[https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/train.py][https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_train.py]

## 1）loaddata函数、定义自己的Dataset ##

此处把三步走的前两步都做了。

主要是解析voc的xml文件，然后读图，\_\_getitem\_\_将img, bbox, label, difficult几个结果返回。

import os
    import xml.etree.ElementTree as ET
    
    import numpy as np
    
    from .util import read_image
    
    
    class VOCBboxDataset:
        """Bounding box dataset for PASCAL `VOC`_.
        .. _`VOC`: http://host.robots.ox.ac.uk/pascal/VOC/voc2012/
        The index corresponds to each image.
        When queried by an index, if :obj:`return_difficult == False`,
        this dataset returns a corresponding
        :obj:`img, bbox, label`, a tuple of an image, bounding boxes and labels.
        This is the default behaviour.
        If :obj:`return_difficult == True`, this dataset returns corresponding
        :obj:`img, bbox, label, difficult`. :obj:`difficult` is a boolean array
        that indicates whether bounding boxes are labeled as difficult or not.
        The bounding boxes are packed into a two dimensional tensor of shape
        :math:`(R, 4)`, where :math:`R` is the number of bounding boxes in
        the image. The second axis represents attributes of the bounding box.
        They are :math:`(y_{min}, x_{min}, y_{max}, x_{max})`, where the
        four attributes are coordinates of the top left and the bottom right
        vertices.
        The labels are packed into a one dimensional tensor of shape :math:`(R,)`.
        :math:`R` is the number of bounding boxes in the image.
        The class name of the label :math:`l` is :math:`l` th element of
        :obj:`VOC_BBOX_LABEL_NAMES`.
        The array :obj:`difficult` is a one dimensional boolean array of shape
        :math:`(R,)`. :math:`R` is the number of bounding boxes in the image.
        If :obj:`use_difficult` is :obj:`False`, this array is
        a boolean array with all :obj:`False`.
        The type of the image, the bounding boxes and the labels are as follows.
        * :obj:`img.dtype == numpy.float32`
        * :obj:`bbox.dtype == numpy.float32`
        * :obj:`label.dtype == numpy.int32`
        * :obj:`difficult.dtype == numpy.bool`
        Args:
            data_dir (string): Path to the root of the training data. 
                i.e. "/data/image/voc/VOCdevkit/VOC2007/"
            split ({'train', 'val', 'trainval', 'test'}): Select a split of the
                dataset. :obj:`test` split is only available for
                2007 dataset.
            year ({'2007', '2012'}): Use a dataset prepared for a challenge
                held in :obj:`year`.
            use_difficult (bool): If :obj:`True`, use images that are labeled as
                difficult in the original annotation.
            return_difficult (bool): If :obj:`True`, this dataset returns
                a boolean array
                that indicates whether bounding boxes are labeled as difficult
                or not. The default value is :obj:`False`.
        """
    
        def __init__(self, data_dir, split='trainval',
                     use_difficult=False, return_difficult=False,
                     ):
    
            # if split not in ['train', 'trainval', 'val']:
            #     if not (split == 'test' and year == '2007'):
            #         warnings.warn(
            #             'please pick split from \'train\', \'trainval\', \'val\''
            #             'for 2012 dataset. For 2007 dataset, you can pick \'test\''
            #             ' in addition to the above mentioned splits.'
            #         )
            id_list_file = os.path.join(
                data_dir, 'ImageSets/Main/{0}.txt'.format(split))
    
            self.ids = [id_.strip() for id_ in open(id_list_file)]
            self.data_dir = data_dir
            self.use_difficult = use_difficult
            self.return_difficult = return_difficult
            self.label_names = VOC_BBOX_LABEL_NAMES
    
        def __len__(self):
            return len(self.ids)
    
        def get_example(self, i):
            """Returns the i-th example.
            Returns a color image and bounding boxes. The image is in CHW format.
            The returned image is RGB.
            Args:
                i (int): The index of the example.
            Returns:
                tuple of an image and bounding boxes
            """
            id_ = self.ids[i]
            anno = ET.parse(
                os.path.join(self.data_dir, 'Annotations', id_ + '.xml'))
            bbox = list()
            label = list()
            difficult = list()
            for obj in anno.findall('object'):
                # when in not using difficult split, and the object is
                # difficult, skipt it.
                if not self.use_difficult and int(obj.find('difficult').text) == 1:
                    continue
    
                difficult.append(int(obj.find('difficult').text))
                bndbox_anno = obj.find('bndbox')
                # subtract 1 to make pixel indexes 0-based
                bbox.append([
                    int(bndbox_anno.find(tag).text) - 1
                    for tag in ('ymin', 'xmin', 'ymax', 'xmax')])
                name = obj.find('name').text.lower().strip()
                label.append(VOC_BBOX_LABEL_NAMES.index(name))
            bbox = np.stack(bbox).astype(np.float32)
            label = np.stack(label).astype(np.int32)
            # When `use_difficult==False`, all elements in `difficult` are False.
            difficult = np.array(difficult, dtype=np.bool).astype(np.uint8)  # PyTorch don't support np.bool
    
            # Load a image
            img_file = os.path.join(self.data_dir, 'JPEGImages', id_ + '.jpg')
            img = read_image(img_file, color=True)
    
            # if self.return_difficult:
            #     return img, bbox, label, difficult
            return img, bbox, label, difficult
    
        __getitem__ = get_example
    
    
    VOC_BBOX_LABEL_NAMES = (
        'aeroplane',
        'bicycle',
        'bird',
        'boat',
        'bottle',
        'bus',
        'car',
        'cat',
        'chair',
        'cow',
        'diningtable',
        'dog',
        'horse',
        'motorbike',
        'person',
        'pottedplant',
        'sheep',
        'sofa',
        'train',
        'tvmonitor')

## 2） 封装train\_dataset与test\_dataset ##

[https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/data/dataset.py![icon-default.png_t_L892][]https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/data/dataset.py][https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_data_dataset.py_icon-default.png_t_L892_https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_data_dataset.py]

下面的代码主要是将训练与测试的数据经过不同的transformer封装起来。

from __future__ import  absolute_import
    from __future__ import  division
    import torch as t
    from data.voc_dataset import VOCBboxDataset
    from skimage import transform as sktsf
    from torchvision import transforms as tvtsf
    from data import util
    import numpy as np
    from utils.config import opt
    
    
    def inverse_normalize(img):
        if opt.caffe_pretrain:
            img = img + (np.array([122.7717, 115.9465, 102.9801]).reshape(3, 1, 1))
            return img[::-1, :, :]
        # approximate un-normalize for visualize
        return (img * 0.225 + 0.45).clip(min=0, max=1) * 255
    
    
    def pytorch_normalze(img):
        """
        https://github.com/pytorch/vision/issues/223
        return appr -1~1 RGB
        """
        normalize = tvtsf.Normalize(mean=[0.485, 0.456, 0.406],
                                    std=[0.229, 0.224, 0.225])
        img = normalize(t.from_numpy(img))
        return img.numpy()
    
    
    def caffe_normalize(img):
        """
        return appr -125-125 BGR
        """
        img = img[[2, 1, 0], :, :]  # RGB-BGR
        img = img * 255
        mean = np.array([122.7717, 115.9465, 102.9801]).reshape(3, 1, 1)
        img = (img - mean).astype(np.float32, copy=True)
        return img
    
    
    def preprocess(img, min_size=600, max_size=1000):
        """Preprocess an image for feature extraction.
        The length of the shorter edge is scaled to :obj:`self.min_size`.
        After the scaling, if the length of the longer edge is longer than
        :param min_size:
        :obj:`self.max_size`, the image is scaled to fit the longer edge
        to :obj:`self.max_size`.
        After resizing the image, the image is subtracted by a mean image value
        :obj:`self.mean`.
        Args:
            img (~numpy.ndarray): An image. This is in CHW and RGB format.
                The range of its value is :math:`[0, 255]`.
        Returns:
            ~numpy.ndarray: A preprocessed image.
        """
        C, H, W = img.shape
        scale1 = min_size / min(H, W)
        scale2 = max_size / max(H, W)
        scale = min(scale1, scale2)
        img = img / 255.
        img = sktsf.resize(img, (C, H * scale, W * scale), mode='reflect',anti_aliasing=False)
        # both the longer and shorter should be less than
        # max_size and min_size
        if opt.caffe_pretrain:
            normalize = caffe_normalize
        else:
            normalize = pytorch_normalze
        return normalize(img)
    
    
    class Transform(object):
    
        def __init__(self, min_size=600, max_size=1000):
            self.min_size = min_size
            self.max_size = max_size
    
        def __call__(self, in_data):
            img, bbox, label = in_data
            _, H, W = img.shape
            img = preprocess(img, self.min_size, self.max_size)
            _, o_H, o_W = img.shape
            scale = o_H / H
            bbox = util.resize_bbox(bbox, (H, W), (o_H, o_W))
    
            # horizontally flip
            img, params = util.random_flip(
                img, x_random=True, return_param=True)
            bbox = util.flip_bbox(
                bbox, (o_H, o_W), x_flip=params['x_flip'])
    
            return img, bbox, label, scale
    
    
    class Dataset:
        def __init__(self, opt):
            self.opt = opt
            self.db = VOCBboxDataset(opt.voc_data_dir)
            self.tsf = Transform(opt.min_size, opt.max_size)
    
        def __getitem__(self, idx):
            ori_img, bbox, label, difficult = self.db.get_example(idx)
    
            img, bbox, label, scale = self.tsf((ori_img, bbox, label))
            # TODO: check whose stride is negative to fix this instead copy all
            # some of the strides of a given numpy array are negative.
            return img.copy(), bbox.copy(), label.copy(), scale
    
        def __len__(self):
            return len(self.db)
    
    
    class TestDataset:
        def __init__(self, opt, split='test', use_difficult=True):
            self.opt = opt
            self.db = VOCBboxDataset(opt.voc_data_dir, split=split, use_difficult=use_difficult)
    
        def __getitem__(self, idx):
            ori_img, bbox, label, difficult = self.db.get_example(idx)
            img = preprocess(ori_img)
            return img, ori_img.shape[1:], bbox, label, difficult
    
        def __len__(self):
            return len(self.db)

## 3）DataLoader把自定义Dataset包起来 ##

[https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/train.py![icon-default.png_t_L892][]https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/train.py][https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_train.py]

下面没有什么需要说的了。

dataset = Dataset(opt)
        print('load data')
        dataloader = data_.DataLoader(dataset, \
                                      batch_size=1, \
                                      shuffle=True, \
                                      # pin_memory=True,
                                      num_workers=opt.num_workers)
        testset = TestDataset(opt)
        test_dataloader = data_.DataLoader(testset,
                                           batch_size=1,
                                           num_workers=opt.test_num_workers,
                                           shuffle=False, \
                                           pin_memory=True
                                           )

[1._Torch_DataLoader -]: https://zhuanlan.zhihu.com/p/346553758
[2.pytorch Dataset_ DataLoader_pan_jinquan_-CSDN]: https://blog.csdn.net/guyuealian/article/details/88343924
[3.GitHub - JDAI-CV_fast-reid_ SOTA Re-identification Methods and Toolbox]: https://github.com/JDAI-CV/fast-reid
[4.https_github.com_chenyuntc_simple-faster-rcnn-pytorch]: https://github.com/chenyuntc/simple-faster-rcnn-pytorch
[icon-default.png_t_L892]: /images/20220828/51ace59a9f8e43dfa9f2db8963fcf510.png
[https_github.com_JDAI-CV_fast-reid_blob_master_fastreid_data_datasets_market1501.py_icon-default.png_t_L892_https_github.com_JDAI-CV_fast-reid_blob_master_fastreid_data_datasets_market1501.py]: https://github.com/JDAI-CV/fast-reid/blob/master/fastreid/data/datasets/market1501.py
[getitem]: https://blog.csdn.net/Norsaa/article/details/105564931
[read_image]: https://github.com/JDAI-CV/fast-reid/blob/master/fastreid/data/data_utils.py
[https_github.com_JDAI-CV_fast-reid_blob_master_fastreid_data_common.py_icon-default.png_t_L892_https_github.com_JDAI-CV_fast-reid_blob_master_fastreid_data_common.py]: https://github.com/JDAI-CV/fast-reid/blob/master/fastreid/data/common.py
[fast-reid_build.py at f4551a128ba17ef201301ccf3c986edae014cabd _ JDAI-CV_fast-reid _ GitHub_icon-default.png_t_L892_https_github.com_JDAI-CV_fast-reid_blob_f4551a128ba17ef201301ccf3c986edae014cabd_fastreid_data_build.py_L73]: https://github.com/JDAI-CV/fast-reid/blob/f4551a128ba17ef201301ccf3c986edae014cabd/fastreid/data/build.py#L73
[https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_data_voc_dataset.py_icon-default.png_t_L892_https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_data_voc_dataset.py]: https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/data/voc_dataset.py
[https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_data_util.py_text_def_20read_image_path_2C_20dtype_3Dnp.float32_2C_20color_3DTrue_3A_icon-default.png_t_L892_https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_data_util.py_text_def_20read_image_path_2C_20dtype_3Dnp.float32_2C_20color_3DTrue_3A]: https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/data/util.py#:~:text=def%20read_image%28path%2C%20dtype%3Dnp.float32%2C%20color%3DTrue%29%3A
[https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_train.py]: https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/train.py
[https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_data_dataset.py_icon-default.png_t_L892_https_github.com_chenyuntc_simple-faster-rcnn-pytorch_blob_master_data_dataset.py]: https://github.com/chenyuntc/simple-faster-rcnn-pytorch/blob/master/data/dataset.py