# -*- coding: utf-8 -*-
# --------------------------------------------------------
# Fast/er R-CNN
# Licensed under The MIT License [see LICENSE for details]
# Written by Bharath Hariharan
# --------------------------------------------------------
import xml.etree.ElementTree as ET
import os
import pickle
import numpy as np
def parse_rec(filename):
    """ Parse a PASCAL VOC xml file """
    tree = ET.parse(filename)
    objects = []
    for obj in tree.findall('object'):
        obj_struct = {}
        obj_struct['name'] = (obj.find('name').text).replace(" ", "")
        obj_struct['pose'] = obj.find('pose').text
        obj_struct['truncated'] = int(obj.find('truncated').text)
        obj_struct['difficult'] = int(obj.find('difficult').text)
        bbox = obj.find('bndbox')
        obj_struct['bbox'] = [int(bbox.find('xmin').text),
                              int(bbox.find('ymin').text),
                              int(bbox.find('xmax').text),
                              int(bbox.find('ymax').text)]
        objects.append(obj_struct)
    return objects
def voc_ap(rec, prec, use_07_metric=False):  # voc2007的计算方式和voc2012的计算方式不同，目前一般采用第二种
    """ ap = voc_ap(rec, prec, [use_07_metric])
    Compute VOC AP given precision and recall.
    If use_07_metric is true, uses the
    VOC 07 11 point method (default:False).
    if use_07_metric:
        # 11 point metric
        ap = 0.
        for t in np.arange(0., 1.1, 0.1):
            if np.sum(rec >= t) == 0:
                p = 0
            else:
                p = np.max(prec[rec >= t])
            ap = ap + p / 11.
    else:
        # correct AP calculation
        # first append sentinel values at the end
        mrec = np.concatenate(([0.], rec, [1.]))
        mpre = np.concatenate(([0.], prec, [0.]))
        # compute the precision envelope
        for i in range(mpre.size - 1, 0, -1):
            mpre[i - 1] = np.maximum(mpre[i - 1], mpre[i])
        # to calculate area under PR curve, look for points
        # where X axis (recall) changes value
        i = np.where(mrec[1:] != mrec[:-1])[0]
        # and sum (\Delta recall) * prec
        ap = np.sum((mrec[i + 1] - mrec[i]) * mpre[i + 1])
    return ap
## 程序入口
def yolov5_eval(detpath,  # 保存检测到的目标框的文件路径，每一类的目标框单独保存在一个文件
                annopath,  # Annotations的路径
                imagesetfile,  # 测试图片名字列表
                classname,  # 类别名称
                cachedir,  # 缓存文件夹
                ovthresh=0.5,  # IoU阈值
                use_07_metric=False):  # mAP计算方法
    """rec, prec, ap = voc_eval(eval_classtxt_path,
                                annopath,
                                imagesetfile,
                                classname,
                                [ovthresh],
                                [use_07_metric])
    Top level function that does the PASCAL VOC evaluation.
    eval_classtxt_path: Path to detections
        eval_classtxt_path.format(classname) should produce the detection results file.
    annopath: Path to annotations
        annopath.format(imagename) should be the xml annotations file.
    imagesetfile: Text file containing the list of images, one image per line.
    classname: Category name (duh)
    cachedir: Directory for caching the annotations
    [ovthresh]: Overlap threshold (default = 0.5)
    [use_07_metric]: Whether to use VOC07's 11 point AP computation
        (default False)
    # assumes detections are in eval_classtxt_path.format(classname)
    # assumes annotations are in annopath.format(imagename)
    # assumes imagesetfile is a text file with each line an image name
    # cachedir caches the annotations in a pickle file
    # first load gt   获取真实目标框
    # 当程序第一次运行时，会读取Annotations下的xml文件获取每张图片中真实的目标框
    # 然后把获取的结果保存在annotations_cache文件夹中
    # 以后再次运行时直接从缓存文件夹中读取真实目标
    if not os.path.isdir(cachedir):
        os.mkdir(cachedir)
    cachefile = os.path.join(cachedir, 'annots.pkl')
    # read list of images
    with open(imagesetfile, 'r') as f:
        lines = f.readlines()
    imagenames = [x.strip() for x in lines]
    if not os.path.isfile(cachefile):
        # load annots
        recs = {}
        for i, imagename in enumerate(imagenames):
            recs[imagename] = parse_rec(annopath.format(imagename))
            if i % 100 == 0:
                print('Reading annotation for {:d}/{:d}'.format(i + 1, len(imagenames)))
        # save
        print('Saving cached annotations to {:s}'.format(cachefile))
        # with open(cachefile, 'w') as cls:
        #     pickle.dump(recs, cls)
        with open(cachefile, 'wb') as f:
            pickle.dump(recs, f)
    else:
        # load
        with open(cachefile, 'rb') as f:
            recs = pickle.load(f)
    # extract gt objects for this class 提取该类的真实目标
    class_recs = {}
    npos = 0  # 保存该类一共有多少真实目标
    for imagename in imagenames:
        R = [obj for obj in recs[imagename] if obj['name'] == classname]  # 保存名字为imagename的图片中，类别为classname的目标框的信息
        bbox = np.array([x['bbox'] for x in R])  # 目标框的坐标
        difficult = np.array([x['difficult'] for x in R]).astype(np.bool)  # 是否是难以识别的目标
        det = [False] * len(R)  # 每一个目标框对应一个det[i]，用来判断该目标框是否已经处理过
        npos = npos + sum(~difficult)  # 计算总的目标个数
        class_recs[imagename] = {'bbox': bbox,  # 把每一张图像中的目标框信息放到class_recs中
                                 'difficult': difficult,
                                 'det': det}
    # read dets
    detfile = detpath.format(classname)  # 打开classname类别检测到的目标框文件
    with open(detfile, 'r') as f:
        lines = f.readlines()
    splitlines = [x.strip().split(' ') for x in lines]
    image_ids = [x[0] for x in splitlines]  # 图像名字
    confidence = np.array([float(x[1]) for x in splitlines])  # 置信度
    BB = np.array([[float(z) for z in x[2:]] for x in splitlines])  # 目标框坐标
    # sort by confidence  按照置信度排序
    sorted_ind = np.argsort(-confidence)
    sorted_scores = np.sort(-confidence)
    BB = BB[sorted_ind, :]
    image_ids = [image_ids[x] for x in sorted_ind]
    # go down dets and mark TPs and FPs
    nd = len(image_ids)  # 统计检测到的目标框个数
    tp = np.zeros(nd)  # 创建tp列表，列表长度为目标框个数
    fp = np.zeros(nd)  # 创建fp列表，列表长度为目标框个数
    for d in range(nd):
        R = class_recs[image_ids[d]]  # 得到图像名字为image_ids[d]真实的目标框信息
        bb = BB[d, :].astype(float)  # 得到图像名字为image_ids[d]检测的目标框坐标
        ovmax = -np.inf
        BBGT = R['bbox'].astype(float)  # 得到图像名字为image_ids[d]真实的目标框坐标
        if BBGT.size > 0:
            # compute overlaps  计算IoU
            # intersection
            ixmin = np.maximum(BBGT[:, 0], bb[0])
            iymin = np.maximum(BBGT[:, 1], bb[1])
            ixmax = np.minimum(BBGT[:, 2], bb[2])
            iymax = np.minimum(BBGT[:, 3], bb[3])
            iw = np.maximum(ixmax - ixmin + 1., 0.)
            ih = np.maximum(iymax - iymin + 1., 0.)
            inters = iw * ih
            # union
            uni = ((bb[2] - bb[0] + 1.) * (bb[3] - bb[1] + 1.) +
                   (BBGT[:, 2] - BBGT[:, 0] + 1.) *
                   (BBGT[:, 3] - BBGT[:, 1] + 1.) - inters)
            overlaps = inters / uni
            ovmax = np.max(overlaps)  # 检测到的目标框可能预若干个真实目标框都有交集，选择其中交集最大的
            jmax = np.argmax(overlaps)
        if ovmax > ovthresh:  # IoU是否大于阈值
            if not R['difficult'][jmax]:  # 真实目标框是否难以识别
                if not R['det'][jmax]:  # 该真实目标框是否已经统计过
                    tp[d] = 1.  # 将tp对应第d个位置变成1
                    R['det'][jmax] = 1  # 将该真实目标框做标记
                else:
                    fp[d] = 1.  # 否则将fp对应的位置变为1
        else:
            fp[d] = 1.  # 否则将fp对应的位置变为1
    # compute precision recall
    fp = np.cumsum(fp)  # 按列累加，最大值即为tp数量
    tp = np.cumsum(tp)  # 按列累加，最大值即为fp数量
    rec = tp / float(npos)  # 计算recall
    # avoid divide by zero in case the first detection matches a difficult
    # ground truth
    prec = tp / np.maximum(tp + fp, np.finfo(np.float64).eps)  # 计算精度
    ap = voc_ap(rec, prec, use_07_metric)  # 计算ap
    return rec, prec, ap
首先运行detect_eval_class_txt.py
 在yolov5\data_test\predictions_manual文件夹中保存了测试结果图片，在imgs_name_manual.txt文件中记录测试图片名称，在yolov5\data_test\class_txt_manual文件夹中保存了每一类的结果信息。
 
然后运行compute_mAP.py
 输出各类的AP、recall、precision 以及 mAP.
 
"D:\Program Files\Python38\python.exe" E:/tpz/yolov5/mAP/compute_mAP.py
Reading annotation for 1/712
Reading annotation for 101/712
Reading annotation for 201/712
Reading annotation for 301/712
Reading annotation for 401/712
Reading annotation for 501/712
Reading annotation for 601/712
Reading annotation for 701/712
Saving cached annotations to ../data_test/cachedir_manual/annots.pkl
AP for combustion_lining = 0.9992
recall for combustion_lining = 1.0000
precision for combustion_lining = 0.9951
AP for fan = 0.9968
recall for fan = 0.9968
precision for fan = 1.0000
AP for fan_stator_casing_and_support = 0.9995
recall for fan_stator_casing_and_support = 1.0000
precision for fan_stator_casing_and_support = 0.9918
AP for hpc_spool = 1.0000
recall for hpc_spool = 1.0000
precision for hpc_spool = 0.9950
AP for hpc_stage_5 = 0.9967
recall for hpc_stage_5 = 0.9967
precision for hpc_stage_5 = 0.9918
AP for hp_core_casing = 0.9951
recall for hp_core_casing = 0.9967
precision for hp_core_casing = 0.9870
AP for mixer = 0.9992
recall for mixer = 1.0000
precision for mixer = 0.9967
AP for nozzle = 0.9953
recall for nozzle = 0.9953
precision for nozzle = 0.9953
AP for nozzle_cone = 0.9984
recall for nozzle_cone = 0.9984
precision for nozzle_cone = 0.9967
AP for stand = 1.0000
recall for stand = 1.0000
precision for stand = 0.9985
Mean AP = 0.9980
~~~~~~~~
Results:
0.999
0.997
0.999
1.000
0.997
0.995
0.999
0.995
0.998
1.000
~~~~~~~~
0.998
~~~~~~~~
运行mAP_line.py，可以绘制AP曲线。绘制其他曲线的代码类似。
 
                    目录参考文献代码和权重下载准备工作data中新建几个文件夹makeTxt.pyvoc_label.py文件修改数据集方面的yaml文件修改网络参数方面的yaml文件修改train.py中的一些参数修改训练测试新建data_test新建几个py文件cfg_mAP.py：detect_eval_class_txt.py :compute_mAP.py：mAP_line.py :utils_mAP.py:yolov5_eval.py :本人前段时间用yolov5进行目标检测研究，记录一下流程方便查看，也希望能帮
或者使用命令行
python train.py --cfg yolov5l,yaml --batch-size 64 --weights './runs/train/exp4/weight/last.pt' 
				
有时候i利用yolov5进行检测的时候，除了指定的分类外，还会将一些干扰项整理为一些类，以减少yolov5的检测误判。但在评估网路性能时，因为干扰项是标签以外的类，所以计算出的mAP会比实际上的低一些。
解决方案： 很简单，就是删除掉检测为干扰项的部分结果，但难点在于如何删除。
yolov5官方的评估脚本：val.py
我们只需要在该代码内run()函数中增加几行代码即可：
添加位置：在检测框经历NMS去除重复框后，即：
        # Run NMS
        targets[:, 2:] *=
				
在 YOLOv5 训练过程中，验证集用于评估模型在未见过的数据上的性能表现。下面是使用验证集进行训练的一般步骤：
1. 准备验证集数据：与训练集类似，您需要准备一个包含验证图像和相应标签的数据集。确保验证集数据与训练集数据具有相同的目录结构和标注格式。
2. 配置训练参数：在 YOLOv5 的配置文件中，您可以找到用于指定训练参数和超参数的部分。请确保以下参数设置正确：
   - `val: [path]`：指定验证集数据的路径，可以是单个文件、文件夹或包含多个文件夹的列表。
   - `val_interval`：设置多少个训练批次后进行一次验证。例如，`val_interval=10` 表示每隔 10 个批次进行一次验证。
3. 开始训练：使用命令行或脚本启动 YOLOv5 的训练过程。训练过程中，模型将会周期性地在验证集上进行评估。
4. 查看验证结果：在每次验证后，您可以在训练日志中查看模型在验证集上的性能指标，如 mAP（平均精度均值）。您还可以使用可视化工具来查看模型在验证集上的预测结果。
请注意，确保验证集数据与训练集数据是相互独立的，避免数据泄漏。同时，验证集应该具有足够的样本以确保对模型的性能进行准确评估。
希望这些步骤对您有所帮助！如果您有其他问题，请随时提问。