一般用ElementTree包
import xml.etree.ElementTree as ET
首先传入xml文件名，创建解析对象：
tree = ET.parse(xml_file)
然后获取根结点：
root = tree.getroot()
获取了根结点，也就是整个xml的根结点，就可以用 findall 来找子结点
比如找到’object’结点
root.findall("object") ，此时返回的是 所有object结点的list ,可以用index选择其中一个进行下一步操作
以下面这个xml为例

<annotation>
	<folder>jpg</folder>
	<filename>001801.jpg</filename>
	<path>D:\part4_Picture\2019-7-8\wuhan_3\jpg\001801.jpg</path>
	<source>
		<database>Unknown</database>
	</source>
		<width>4608</width>
		<height>2592</height>
		<depth>3</depth>
	</size>
	<segmented>0</segmented>
	<object>
		<name>lsqxz</name>
		<pose>Unspecified</pose>
		<truncated>0</truncated>
		<difficult>0</difficult>
		<bndbox>
			<xmin>2000</xmin>
			<ymin>1236</ymin>
			<xmax>2057</xmax>
			<ymax>1298</ymax>
		</bndbox>
	</object>
</annotation>
object这个结点下面还有子结点,name,poe,bndbox等等
 我们可以遍历所有的object结点，对每个Object结点进行分析
 
for obj in root.findall("object"):
	cls_name = obj.findall("name")[0].text
	bbox = obj.findall("bndbox")[0]
	xmin = bbox.findall("xmin")[0].text
	ymin = bbox.findall("ymin")[0].text
This method is used to find a specific single element in the document. It is essentially equivalent to calling the .find()method on the document's root element;
# encoding:utf-8
import xml.etree.ElementTree as ET
from xml.dom.minidom import parseString
import codecs
import sys
def get_xml_dict(source_file_name...
minixml常用函数接口：均需包含头文件#include mxml.h
mxml_node_t *mxmlFindElement(mxml_node_t *node, mxml_node_t *top,const char *name, const char *attr,const char *value, int descend);
desc:
    node节点为被查找的结点
    top为顶层结点
    name， attrr， value为NULL时表示任意匹配； 否则为
            MLP和Cross Attention的关系 (参考Memory-Space Visual Prompting for Efficient Vision-Language Fine-Tuning)
            Classifier guidance与Classifier free diffusion的简单理解
            Monkey: Image Resolution and Text Label Are Important Things for Large Multi-modal Models