python中darks_【YOLOv3】基于darknet的训练过程（简洁版）

准备工作

下载darknet

修改Makefile

下载yolov3.weights

下载darknet53.conv.74(后面训练数据集会用到，先下载好了，留着备用)

测试

#1. 下载`darknet`

git clone https://github.com/pjreddie/darknet

#2. 修改`Makefile`

cd darknet

make

#3. 下载`yolov3.weights`

wget https://pjreddie.com/media/files/yolov3.weights

#4. 下载`darknet53.conv.74`

wget https://pjreddie.com/media/files/darknet53.conv.74

#5. 测试图片

./darknet detect cfg/yolov3.cfg yolov3.weights data/dog.jpg

./darknet detector test cfg/coco.data cfg/yolov3.cfg yolov3.weights data/dog.jpg

#5. 测试视频

./darknet detector demo cfg/coco.data cfg/yolov3.cfg yolov3.weights

#5. 加载摄像头

./darknet detector demo cfg/coco.data cfg/yolov3.cfg yolov3.weights

总结：可以将上述下载的文件备份一下，以后每次自己训练数据集的时候就不用再次下载了，直接改用。

训练自己的数据集

在darknet文件夹下新建一个myData文件夹。将标注好的图片和xml文件放到对应目录下。运行test.py生成train.txt/val.txt/test.txt/trainval.txt文件。myData包含如下文件(夹)：

myData

......JPEGImages #存放图像

......Annotations #存放图像对应的xml文件

......ImageSets/Main #存放训练/存放train.txt/val.txt/test.txt/trainval.txt文件

......test.py #生成train.txt/val.txt/test.txt/trainval.txt文件

test.py代码如下：

import os

import random

trainval_percent = 0.1

train_percent = 0.9

xmlfilepath = 'Annotations'

txtsavepath = 'ImageSets\Main'

total_xml = os.listdir(xmlfilepath)

num = len(total_xml)

list = range(num)

tv = int(num * trainval_percent)

tr = int(tv * train_percent)

trainval = random.sample(list, tv)

train = random.sample(trainval, tr)

ftrainval = open('ImageSets/Main/trainval.txt', 'w')

ftest = open('ImageSets/Main/test.txt', 'w')

ftrain = open('ImageSets/Main/train.txt', 'w')

fval = open('ImageSets/Main/val.txt', 'w')

for i in list:

name = total_xml[i][:-4] + '\n'

if i in trainval:

ftrainval.write(name)

if i in train:

ftest.write(name)

else:

fval.write(name)

else:

ftrain.write(name)

ftrainval.close()

ftrain.close()

fval.close()

ftest.close()

如果按照上述文件结构，则test.py文件不需要修改，直接运行，即可生成txt文件。

将darknet文件夹下的scripts/voc_label.py拷贝出来，修改成my_labels.py放在darknet文件夹中。【注意修改类别和路径】

运行该脚本my_lables.py会在./myData目录下生成一个labels文件夹一个txt文件(myData_train.txt)(内容是: 类别的编码和目标的相对位置)。

my_labels.py代码如下：

import xml.etree.ElementTree as ET

import pickle

import os

from os import listdir, getcwd

from os.path import join

sets=[('myData', 'train'), ('myData', 'val'), ('myData', 'train'), ('myData', 'val'), ('myData', 'test')]

classes = ["person", "foot", "face"] # 改成自己的类别

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(year, image_id):

in_file = open('myData/Annotations/%s.xml'%(image_id))

out_file = open('myData/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'):

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))

bb = convert((w,h), b)

out_file.write(str(cls_id) + " " + " ".join([str(a) for a in bb]) + '\n')

wd = getcwd()

for year, image_set in sets:

if not os.path.exists('myData/labels/'): # 改成自己建立的myData

os.makedirs('myData/labels/')

image_ids = open('myData/ImageSets/Main/%s.txt'%(image_set)).read().strip().split()

list_file = open('myData/%s_%s.txt'%(year, image_set), 'w')

for image_id in image_ids:

list_file.write('%s/myData/JPEGImages/%s.jpg\n'%(wd, image_id))

convert_annotation(year, image_id)

list_file.close()

注意：这里面如果采用上述文件结构，只需要将classes改成自己的类别即可，其他内容不需要修改。

在myData文件夹下新建myData.names文件。

在myData文件夹下新建weights文件，用于保存生成的权重文件。

修改darknet/cfg下的voc.data和yolov3-voc.cfg文件。

复制这两个文件，并分别重命名为my_data.data和my_yolov3.cfg

(1)修改my_data.data：

classes= 4 #改为自己的分类个数

##下面都改为自己的路径

train = /home/zhan/darknet/myData/myData_train.txt

valid =/home/zhan/darknet/myData/myData_test.txt

names = /home/zhan/darknet/myData/myData.names

backup = /home/zhan/darknet/myData/weights

(2)修改my_yolov3.cfg：

Ctrl+F，搜出3个含有yolo的地方。每个地方都必须要改2处，filters 、classes

filters：3*(5+len(classes))

可修改：random = 1：原来是1，显存小改为0。(是否要多尺度输出。)

一般地，max_batches修改成合适的数值。

开始训练

训练

首先，将cfg/my_yolov3.cfg文件中改成Training模式。

训练命令：

./darknet detector train cfg/my_data.data cfg/my_yolov3.cfg darknet53.conv.74

# 指定gpu训练，默认使用gpu0(查看GPU情况，`nvidia-smi`)

./darknet detector train cfg/my_data.data cfg/my_yolov3.cfg darknet53.conv.74 -gups 0,1,2,3

# 训练过程中保存训练日志xxx.log

./darknet detector train cfg/my_data.data cfg/my_yolov3.cfg darknet53.conv.74 | tee train_yolov3.log

# 断点继续训练

./darknet detector train cfg/my_data.data cfg/my_yolov3.cfg myData/weights/my_yolov3.backup | tee new_train_yolov3.log

训练日志可视化

vis_yolov3_log.py代码如下：

# -*- coding: utf-8 -*-

import pandas as pd

import matplotlib.pyplot as plt

import os

# ==================可能需要修改的地方=====================================#

g_log_path = "train_yolov3.log" # 此处修改为自己的训练日志文件名

# ==========================================================================#

def extract_log(log_file, new_log_file, key_word):

'''

:param log_file:日志文件

:param new_log_file:挑选出可用信息的日志文件

:param key_word:根据关键词提取日志信息

:return:

'''

with open(log_file, "r") as f:

with open(new_log_file, "w") as train_log:

for line in f:

# 去除多gpu的同步log

if "Syncing" in line:

continue

# 去除nan log

if "nan" in line:

continue

if key_word in line:

train_log.write(line)

f.close()

train_log.close()

def drawAvgLoss(loss_log_path):

'''

:param loss_log_path: 提取到的loss日志信息文件

:return: 画loss曲线图

'''

line_cnt = 0

for count, line in enumerate(open(loss_log_path, "rU")):

line_cnt += 1

result = pd.read_csv(loss_log_path, skiprows=[iter_num for iter_num in range(line_cnt) if ((iter_num < 500))],

error_bad_lines=False,

names=["loss", "avg", "rate", "seconds", "images"])

result["avg"] = result["avg"].str.split(" ").str.get(1)

result["avg"] = pd.to_numeric(result["avg"])

fig = plt.figure(1, figsize=(6, 4))

ax = fig.add_subplot(1, 1, 1)

ax.plot(result["avg"].values, label="Avg Loss", color="#ff7043")

ax.legend(loc="best")

ax.set_title("Avg Loss Curve")

ax.set_xlabel("Batches")

ax.set_ylabel("Avg Loss")

def drawIOU(iou_log_path):

'''

:param iou_log_path: 提取到的iou日志信息文件

:return: 画iou曲线图

'''

line_cnt = 0

for count, line in enumerate(open(iou_log_path, "rU")):

line_cnt += 1

result = pd.read_csv(iou_log_path, skiprows=[x for x in range(line_cnt) if (x % 39 != 0 | (x < 5000))],

error_bad_lines=False,

names=["Region Avg IOU", "Class", "Obj", "No Obj", "Avg Recall", "count"])

result["Region Avg IOU"] = result["Region Avg IOU"].str.split(": ").str.get(1)

result["Region Avg IOU"] = pd.to_numeric(result["Region Avg IOU"])

result_iou = result["Region Avg IOU"].values

# 平滑iou曲线

for i in range(len(result_iou) - 1):

iou = result_iou[i]

iou_next = result_iou[i + 1]

if abs(iou - iou_next) > 0.2:

result_iou[i] = (iou + iou_next) / 2

fig = plt.figure(2, figsize=(6, 4))

ax = fig.add_subplot(1, 1, 1)

ax.plot(result_iou, label="Region Avg IOU", color="#ff7043")

ax.legend(loc="best")

ax.set_title("Avg IOU Curve")

ax.set_xlabel("Batches")

ax.set_ylabel("Avg IOU")

if __name__ == "__main__":

loss_log_path = "train_log_loss.txt"

iou_log_path = "train_log_iou.txt"

if os.path.exists(g_log_path) is False:

exit(-1)

if os.path.exists(loss_log_path) is False:

extract_log(g_log_path, loss_log_path, "images")

if os.path.exists(iou_log_path) is False:

extract_log(g_log_path, iou_log_path, "IOU")

drawAvgLoss(loss_log_path)

drawIOU(iou_log_path)

plt.show()

可视化这部分除了需要将训练日志文件名修改成自己的，还要特别注意skiprows=[iter_num for iter_num in range(line_cnt) if ((iter_num < 500))]和skiprows=[x for x in range(line_cnt) if (x % 39 != 0 | (x < 5000))]这两部分，需要根据自己的训练次数来设定的。

分别表示，迭代次数小于500次的跳过，画图不用，从501开始画图；每隔39个数或者前5000个数跳过，说白了就是，前5000个数值舍弃，从第5001个数开始，每隔39个数取一个数值参与画图。

附录两张非常令人糟心的图(因为不知怎么地，它就失败了～～)