在github官网上搜索yolov5,链接
注意:yolov5有好几个版本,本文主要采用yolov5-V2.0的版本作为此次的使用,因此最好不要下错了,注意选取版本。
根据代码中自带的Dockerfile文件,在装有Docker环境的系统中运行:docker build -t 镜像名:版本号 .
注意:这里的镜像名和版本号是自己起的。
注意: 在训练之前我们需要去查看自己的算法运行环境是否正确,避免出现后续的环境调试问题。
在本地的环境调试中(即pycharm中),我们直接运行detect.py文件就行了,如果环境和权重有无误的话会出现如下内容:
在项目路径的 inference/output中生成如下内容:
在linux的docker环境下,直接运行python detect.py文件即可(需要先挂载容器哦,详细见下一个步骤)
挂载容器,无非是需要挂载两个目录路径:
运行:
sudo docker run --runtime=nvidia --name=yolov5-v2.0_test -v 数据集目录/:/yolov5/data -v 代码路径/:/yolov5 -it 镜像id /bin/bash
解释说明: 通过-v 将数据集目录挂载到容器根目录下的:yolov5/data的目录中
通过-v将代码挂载到容器根目录下得yolov5目录中。
由于yolov5和传统的VOC数据集存在差异,yolov5有自己的标签存储方式,所以针对标准的VOC数据集来说需要转换成yolo格式的标签存储。
# -*- coding: utf-8 -*-
import xml.etree.ElementTree as ET
import os
from os import getcwd
sets = ['train', 'val'] #ImageSets目录下得txt文件
classes = ["10"] # 改成自己的类别
root_dir = r"D:\Dataset\one_pointer_sum\VOC2007" #VOC数据集的根目录
def convert(size, box):
dw = 1. / (size[0])
dh = 1. / (size[1])
x = (box[0] + box[1]) / 2.0 - 1
y = (box[2] + box[3]) / 2.0 - 1
w = box[1] - box[0]
h = box[3] - box[2]
x = x * dw
w = w * dw
y = y * dh
h = h * dh
return x, y, w, h
def convert_annotation(image_id):
in_file = open(root_dir + '/Annotations/%s.xml' % (image_id), encoding='UTF-8')
out_file = open(root_dir + '/labels/%s.txt' % (image_id), 'w')
tree = ET.parse(in_file)
root = tree.getroot()
size = root.find('size')
w = int(size.find('width').text)
h = int(size.find('height').text)
for obj in root.iter('object'):
if obj.find('difficult'):
difficult = float(obj.find('difficult').text)
else:
difficult = 0
# difficult = obj.find('Difficult').text
cls = obj.find('name').text
if cls not in classes or int(difficult) == 1:
continue
cls_id = classes.index(cls)
xmlbox = obj.find('bndbox')
b = (float(xmlbox.find('xmin').text), float(xmlbox.find('xmax').text), float(xmlbox.find('ymin').text),
float(xmlbox.find('ymax').text))
b1, b2, b3, b4 = b
# 标注越界修正
if b2 > w:
b2 = w
if b4 > h:
b4 = h
b = (b1, b2, b3, b4)
bb = convert((w, h), b)
out_file.write(str(cls_id) + " " + " ".join([str(a) for a in bb]) + '\n')
wd = getcwd()
for image_set in sets:
if not os.path.exists(root_dir + '/labels/'):
os.makedirs(root_dir + '/labels/')
image_ids = open(root_dir + '/ImageSets/Main/%s.txt' % (image_set)).read().strip().split()
if not os.path.exists(root_dir + '/dataSet_path/'):
os.makedirs(root_dir + '/dataSet_path/')
list_file = open(root_dir + '/dataSet_path/%s.txt' % (image_set), 'w')
for image_id in image_ids:
list_file.write(root_dir + '/images/%s.jpg\n' % (image_id))
convert_annotation(image_id)
list_file.close()
运行上述代码后,VOC的根目录格式变为如下所示。
其中labels中存放的是VOC数据集的XML标签转换为yolo格式标签。
dataSet_path中存放的是train和val,区别于VOC数据的是VOC中存放的train和val是文件名,而yolo格式中的train和val是每个文件的路径
熟悉目标检测算法的都知道,运行github上的代码主要做配置文件的修改一般只有两个地方
在项目根目录下得data/voc.yaml或coco.yaml中需要修改的地方如下所示:
其中需要修改的地方有四处。
注意:在改完后要注意冒号":"后面要有一个空格,不然会出错,如"train:"后面不能直接写路径,需要添加一个空格,然后写路径。
在项目根目录下的model/目录下有可以选取的网络模型,本文选取yolov5s.yaml做测试。
需要更改的地方主要有两处,其中第二处可改可不改.
新建一个py文件,并命名为clauculate.py,文件的内容如下所示:
# -*- coding: utf-8 -*-
# 根据标签文件求先验框
import os
import numpy as np
import xml.etree.cElementTree as et
from kmeans import kmeans, avg_iou
FILE_ROOT = r"E:\2021年无人机缺陷识别比赛数据\大金具/" # 根路径
ANNOTATION_PATH = FILE_ROOT + "Annotations"
ANCHORS_TXT_PATH = FILE_ROOT + "/anchors.txt" #anchors文件保存位置
CLUSTERS = 10
CLASS_NAMES = ['xcxjzc', 'fzczc', 'fzchy', 'fzcxs', 'zczc', 'xcxjpy', 'fzcpy', 'pbhxs', 'fzctl', 'zcxs'] #类别名称
def load_data(anno_dir, class_names):
xml_names = os.listdir(anno_dir)
boxes = []
for xml_name in xml_names:
xml_pth = os.path.join(anno_dir, xml_name)
tree = et.parse(xml_pth)
width = float(tree.findtext("./size/width"))
height = float(tree.findtext("./size/height"))
for obj in tree.findall("./object"):
cls_name = obj.findtext("name")
if cls_name in class_names:
xmin = float(obj.findtext("bndbox/xmin")) / width
ymin = float(obj.findtext("bndbox/ymin")) / height
xmax = float(obj.findtext("bndbox/xmax")) / width
ymax = float(obj.findtext("bndbox/ymax")) / height
box = [xmax - xmin, ymax - ymin]
boxes.append(box)
else:
continue
return np.array(boxes)
if __name__ == '__main__':
anchors_txt = open(ANCHORS_TXT_PATH, "w")
train_boxes = load_data(ANNOTATION_PATH, CLASS_NAMES)
count = 1
best_accuracy = 0
best_anchors = []
best_ratios = []
for i in range(10): ##### 可以修改,不要太大,否则时间很长
anchors_tmp = []
clusters = kmeans(train_boxes, k=CLUSTERS)
idx = clusters[:, 0].argsort()
clusters = clusters[idx]
# print(clusters)
for j in range(CLUSTERS):
anchor = [round(clusters[j][0] * 640, 2), round(clusters[j][1] * 640, 2)]
anchors_tmp.append(anchor)
print(f"Anchors:{anchor}")
temp_accuracy = avg_iou(train_boxes, clusters) * 100
print("Train_Accuracy:{:.2f}%".format(temp_accuracy))
ratios = np.around(clusters[:, 0] / clusters[:, 1], decimals=2).tolist()
ratios.sort()
print("Ratios:{}".format(ratios))
print(20 * "*" + " {} ".format(count) + 20 * "*")
count += 1
if temp_accuracy > best_accuracy:
best_accuracy = temp_accuracy
best_anchors = anchors_tmp
best_ratios = ratios
anchors_txt.write("Best Accuracy = " + str(round(best_accuracy, 2)) + '%' + "\r\n")
anchors_txt.write("Best Anchors = " + str(best_anchors) + "\r\n")
anchors_txt.write("Best Ratios = " + str(best_ratios))
anchors_txt.close()
新建一个py文件,并命名为kemans.py,文件的内容如下所示:
import numpy as np
def iou(box, clusters):
"""
Calculates the Intersection over Union (IoU) between a box and k clusters.
:param box: tuple or array, shifted to the origin (i. e. width and height)
:param clusters: numpy array of shape (k, 2) where k is the number of clusters
:return: numpy array of shape (k, 0) where k is the number of clusters
"""
x = np.minimum(clusters[:, 0], box[0])
y = np.minimum(clusters[:, 1], box[1])
if np.count_nonzero(x == 0) > 0 or np.count_nonzero(y == 0) > 0:
raise ValueError("Box has no area") # 如果报这个错,可以把这行改成pass即可
intersection = x * y
box_area = box[0] * box[1]
cluster_area = clusters[:, 0] * clusters[:, 1]
iou_ = intersection / (box_area + cluster_area - intersection)
return iou_
def avg_iou(boxes, clusters):
"""
Calculates the average Intersection over Union (IoU) between a numpy array of boxes and k clusters.
:param boxes: numpy array of shape (r, 2), where r is the number of rows
:param clusters: numpy array of shape (k, 2) where k is the number of clusters
:return: average IoU as a single float
"""
return np.mean([np.max(iou(boxes[i], clusters)) for i in range(boxes.shape[0])])
def translate_boxes(boxes):
"""
Translates all the boxes to the origin.
:param boxes: numpy array of shape (r, 4)
:return: numpy array of shape (r, 2)
"""
new_boxes = boxes.copy()
for row in range(new_boxes.shape[0]):
new_boxes[row][2] = np.abs(new_boxes[row][2] - new_boxes[row][0])
new_boxes[row][3] = np.abs(new_boxes[row][3] - new_boxes[row][1])
return np.delete(new_boxes, [0, 1], axis=1)
def kmeans(boxes, k, dist=np.median):
"""
Calculates k-means clustering with the Intersection over Union (IoU) metric.
:param boxes: numpy array of shape (r, 2), where r is the number of rows
:param k: number of clusters
:param dist: distance function
:return: numpy array of shape (k, 2)
"""
rows = boxes.shape[0]
distances = np.empty((rows, k))
last_clusters = np.zeros((rows,))
np.random.seed()
# the Forgy method will fail if the whole array contains the same rows
clusters = boxes[np.random.choice(rows, k, replace=False)]
while True:
for row in range(rows):
distances[row] = 1 - iou(boxes[row], clusters)
nearest_clusters = np.argmin(distances, axis=1)
if (last_clusters == nearest_clusters).all():
break
for cluster in range(k):
clusters[cluster] = dist(boxes[nearest_clusters == cluster], axis=0)
last_clusters = nearest_clusters
return clusters
if __name__ == '__main__':
a = np.array([[1, 2, 3, 4], [5, 7, 6, 8]])
print(translate_boxes(a))
运行clauculate.py后,会生成一个txt文件,该文件中写入的是anchors的建议框大小,只需要将建议框大小取整写入即可。
在训练之前通过观看,train.py文件,可以看到,需要注意的几个地方如下所示:
多GPU训练:
Python -m torch.distributed.launch --nporc_per_node 训练卡数 train.py --img 训练图片大小 --batch 训练的batchsize
--epoch 训练的轮次 --data 数据集的yaml文件路径 --cfg 训练的模型yaml文件路径,--weights 预训练权重路径
--device gpu设备的id号
同理,当你采用单卡训练时,只需要将训练卡数设置为1,并且将gpu 的id号写对就行。
CPU训练的话,不需要给训练的卡数就行。