<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[leewe7的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/leewe7</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; leewe7]]></copyright><item><title><![CDATA[CART回归树]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/134729253</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/134729253</guid><author>leewe7</author><pubDate>Fri, 01 Dec 2023 10:49:51 +0800</pubDate><description><![CDATA[2、针对集合S，遍历每一个feature的每一个value，用该value将原数据集S分裂成2个集合：左集合S_left(<=value的样本)、右集合S_right(>value的样本)，每一个集合也叫做一个结点。3、找到最佳分割feature以及最佳分割value之后，用该value将集合S分裂成2个集合：左集合S_left、右集合S_right，每一个集合也叫做一个结点。（ps：这个mean就是该结点的值，也就是落在该结点内的样本的预测值，同一个结点中的样本具有同一个预测值。]]></description><category></category></item><item><title><![CDATA[CART分类树原理]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/134728409</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/134728409</guid><author>leewe7</author><pubDate>Fri, 01 Dec 2023 10:23:28 +0800</pubDate><description><![CDATA[在计算出来的各个特征的各个值的基尼系数中，选择基尼系数最小的特征A及其对应的取值a作为最优特征和最优切分点。然后根据最优特征和最优切分点，将本节点的数据集划分成两部分  和  ，同时生成当前节点的两个子节点，左节点的数据集和右节点的数据集。m个样本的连续特征A有m个值，从小到大排列  ，则CART取相邻两样本值的平均数做划分点，一共有m-1个；CART分类树算法使用基尼系数选择特征，基尼系数代表了模型的不纯度，基尼系数越小，不纯度越低，特征越好。输入：训练集D，基尼系数的阈值，切分的最少样本个数阈值。]]></description><category></category></item><item><title><![CDATA[numpy中的lstsq（最小二乘法）函数]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/129375953</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/129375953</guid><author>leewe7</author><pubDate>Tue, 07 Mar 2023 09:49:50 +0800</pubDate><description><![CDATA[Numpy中的最小二乘法]]></description><category></category></item><item><title><![CDATA[numpy矩阵的常见操作]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/129283000</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/129283000</guid><author>leewe7</author><pubDate>Wed, 01 Mar 2023 17:00:31 +0800</pubDate><description><![CDATA[numpy矩阵的常用操作]]></description><category></category></item><item><title><![CDATA[模型验证处理]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/128497734</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/128497734</guid><author>leewe7</author><pubDate>Fri, 30 Dec 2022 18:02:05 +0800</pubDate><description><![CDATA[模型验证过程]]></description><category></category></item><item><title><![CDATA[Flink的UDF函数]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/128477251</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/128477251</guid><author>leewe7</author><pubDate>Thu, 29 Dec 2022 11:20:13 +0800</pubDate><description><![CDATA[udtf函数]]></description><category></category></item><item><title><![CDATA[热门服务号计算]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/128339407</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/128339407</guid><author>leewe7</author><pubDate>Fri, 16 Dec 2022 11:13:09 +0800</pubDate><description><![CDATA[热门服务号计算逻辑]]></description><category></category></item><item><title><![CDATA[用户资讯推荐]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/128316840</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/128316840</guid><author>leewe7</author><pubDate>Wed, 14 Dec 2022 15:47:36 +0800</pubDate><description><![CDATA[用户资讯相关推荐]]></description><category></category></item><item><title><![CDATA[item相似度]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/128297482</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/128297482</guid><author>leewe7</author><pubDate>Tue, 13 Dec 2022 11:53:19 +0800</pubDate><description><![CDATA[物品相似度]]></description><category></category></item><item><title><![CDATA[Pytorch自定义数据集]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/127784547</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/127784547</guid><author>leewe7</author><pubDate>Thu, 10 Nov 2022 18:05:06 +0800</pubDate><description><![CDATA[pytorch的Dataset与DataLoader]]></description><category></category></item><item><title><![CDATA[SQL case when 的用法]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/127785052</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/127785052</guid><author>leewe7</author><pubDate>Thu, 10 Nov 2022 11:32:11 +0800</pubDate><description><![CDATA[SQL case when 运用的场景]]></description><category></category></item><item><title><![CDATA[pytroch的nn.Linear]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/125292132</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/125292132</guid><author>leewe7</author><pubDate>Wed, 15 Jun 2022 10:32:27 +0800</pubDate><description><![CDATA[pytroch的linear]]></description><category></category></item><item><title><![CDATA[pythroch函数介绍]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/125290689</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/125290689</guid><author>leewe7</author><pubDate>Wed, 15 Jun 2022 09:26:27 +0800</pubDate><description><![CDATA[函数介绍]]></description><category></category></item><item><title><![CDATA[RocketMQ发送消息的三种方式]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/125000415</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/125000415</guid><author>leewe7</author><pubDate>Fri, 27 May 2022 11:24:41 +0800</pubDate><description><![CDATA[1.rocketQM有三种发送方式：
同步发送：等待返回结果
异步发送：立即返回，使用回调函数
一次性发送：只发送一次，不关心发送结果
2.DefaultMQProducer同步发送
DefaultMQProducer.send(msg);
public SendResult send(msg,CommunicationMode.SYNC,(SendCallback)null,timeout);

2.DefaultMQProducer异步发送
DefaultMQProudcer.send(msg,send]]></description><category></category></item><item><title><![CDATA[rocketMQ producer的异步发送]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/124999652</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/124999652</guid><author>leewe7</author><pubDate>Fri, 27 May 2022 10:54:45 +0800</pubDate><description><![CDATA[1.异步发送的回调函数
producer.send(msg,new SendCallback(){
	@Override
	public void onSuccess(SendResult sendResult){
		
	}
	@Override
	public void onException(Throwable e){
		
	}
});

2.消息发送成功 sendMessageAsync
sendResult = MQClientAPIImpl.this.processSendResponse(b]]></description><category></category></item><item><title><![CDATA[HIVE SQL 按照某字段先分区，分区内部排序，再取内部的第一条数据]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/123119691</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/123119691</guid><author>leewe7</author><pubDate>Thu, 24 Feb 2022 20:03:52 +0800</pubDate><description><![CDATA[with tmp as (
select *,ROW_NUMBER() OVER (PARTITION BY name ORDER BY level1,level2) as num from table 
)
select * from tmp where num=1


]]></description><category></category></item><item><title><![CDATA[python partial函数]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/122199922</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/122199922</guid><author>leewe7</author><pubDate>Tue, 28 Dec 2021 18:36:39 +0800</pubDate><description><![CDATA[partial的**kwargs应用场景：
def a(x,**kwargs):
	print(x)
	print(kwargs.get("a"))

from functools import partial
def hi(x,**kwargs):
	aa = partial(a,**kwargs)
	aa(x)

调用
hi("7",a="ok")



]]></description><category></category></item><item><title><![CDATA[如何在SQL中将某个算出来的值平摊到每一行中使用]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/118682911</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/118682911</guid><author>leewe7</author><pubDate>Mon, 12 Jul 2021 20:08:10 +0800</pubDate><description><![CDATA[构造一个关联字段，使用left join，例如求出总和，再求占比：
with a as (select name,count(1) as c,1 as j from t1 group by name),
b as (select count(1) as c,1 as j from t1)
select a.name,a.c as cn,b.c as total,(100*a.c/b.c) from a left join b on a.j=b.j


...]]></description><category></category></item><item><title><![CDATA[java调用PMML模型文件]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/117750128</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/117750128</guid><author>leewe7</author><pubDate>Wed, 09 Jun 2021 17:35:58 +0800</pubDate><description><![CDATA[生成模型
from sklearn2pmml import PMMLPipeline
from sklearn.datasets import load_iris
from sklearn import tree

iris = load_iris()
clf = tree.DecisionTreeClassifier()
pipeline = PMMLPipeline([("classifier", clf)])
pipeline.fit(iris.data, iris.target)

print(ir]]></description><category></category></item><item><title><![CDATA[IDEA控制台乱码问题]]></title><link>https://blog.youkuaiyun.com/leewe7/article/details/117126096</link><guid>https://blog.youkuaiyun.com/leewe7/article/details/117126096</guid><author>leewe7</author><pubDate>Fri, 21 May 2021 16:17:44 +0800</pubDate><description><![CDATA[IDEA文本为UTF-8编码格式，中文不乱码，但控制台乱码，解决方式：
HELP-&gt;Edit Custom VM OPtions中加 -Dfile.encoding=utf-8
重启idea

]]></description><category></category></item></channel></rss>