一、問(wèn)題解決過(guò)程

為了實(shí)現(xiàn)目標(biāo)檢測(cè)模型fast_rcnn，準(zhǔn)備使用coco數(shù)據(jù)集進(jìn)行模型的訓(xùn)練，這里需要導(dǎo)入coco數(shù)據(jù)集，并且進(jìn)行數(shù)據(jù)集處理。

import torch
import torchvision
import torch.utils.data
from pycocotools.coco import COCO


def convert_to_coco_api(ds):
    coco_ds = COCO()
    # annotation IDs need to start at 1, not 0
    ann_id = 1
    dataset = {'images': [], 'categories': [], 'annotations': []}
    categories = set()
    for img_idx in range(len(ds)):
        # find better way to get target
        hw, targets = ds.coco_index(img_idx)
        image_id = targets["image_id"].item()
        img_dict = {}
        img_dict['id'] = image_id
        img_dict['height'] = hw[0]
        img_dict['width'] = hw[1]
        dataset['images'].append(img_dict)
        bboxes = targets["boxes"]
        bboxes[:, 2:] -= bboxes[:, :2]
        bboxes = bboxes.tolist()
        labels = targets['labels'].tolist()
        areas = targets['area'].tolist()
        iscrowd = targets['iscrowd'].tolist()
        num_objs = len(bboxes)
        for i in range(num_objs):
            ann = {}
            ann['image_id'] = image_id
            ann['bbox'] = bboxes[i]
            ann['category_id'] = labels[i]
            categories.add(labels[i])
            ann['area'] = areas[i]
            ann['iscrowd'] = iscrowd[i]
            ann['id'] = ann_id
            dataset['annotations'].append(ann)
            ann_id += 1
    dataset['categories'] = [{'id': i} for i in sorted(categories)]
    coco_ds.dataset = dataset
    coco_ds.createIndex()
    return coco_ds


def get_coco_api_from_dataset(dataset):
    for _ in range(10):
        if isinstance(dataset, torchvision.datasets.CocoDetection):
            break
        if isinstance(dataset, torch.utils.data.Subset):
            dataset = dataset.dataset
    if isinstance(dataset, torchvision.datasets.CocoDetection):
        return dataset.coco
    return convert_to_coco_api(dataset)

運(yùn)行出現(xiàn)
可以得出當(dāng)前環(huán)境沒(méi)有 pycocotools 模塊 故需要

pip install pycocotools

------報(bào)錯(cuò)了-------

然后查詢有一種從gitee上下載的方式

pip install git+https://gitee.com/pursuit_zhangyu/cocoapi.git#subdirectory=PythonAPI

------報(bào)錯(cuò)了-------

上面報(bào)錯(cuò)信息發(fā)現(xiàn)是缺少Cython模塊，那就下載Cython模塊

pip install Cython

然后在進(jìn)行上一步的操作

pip install git+https://gitee.com/pursuit_zhangyu/cocoapi.git#subdirectory=PythonAPI

------還是不行-------

最后發(fā)現(xiàn)linux下載是上面的操作，因?yàn)槲沂莣indous應(yīng)該按照下面的指令

pip install pycocotools-windows

成功?。?/code>