<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[code_fighter   博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/code_fighter</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; code_fighter]]></copyright><item><title><![CDATA[linux下安装MongoDB]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/99433252</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/99433252</guid><author>code_fighter</author><pubDate>Tue, 13 Aug 2019 14:32:28 +0800</pubDate><description><![CDATA[下载安装包：wget http://fastdl.mongodb.org/linux/mongodb-linux-x86_64-2.4.6.tgz
解压：tar -xvzf mongodb-linux-x86_64-3.2.10.tgz
将解压后的文件移动到指定目录并改名:mv mongodb-linux-x86_64-3.2.10 /usr/local/mongodb
切换到mongodb:c...]]></description><category></category></item><item><title><![CDATA[pytorch学习笔记7--循环神经网络、GAN]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/97106983</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/97106983</guid><author>code_fighter</author><pubDate>Wed, 24 Jul 2019 10:32:30 +0800</pubDate><description><![CDATA[文章目录循环神经网络序列的表示方法RNN Layer的使用nn.RNNnn.RNNCell预测sin(x)曲线trainpredict梯度弥散和梯度爆炸gradient clippinggradient vanishingLSTM 的使用LSTMCell 的使用情感分类实战GAN损失纳什均衡transposed convolutionWGAN 使用wassertein Distance 代替了Di...]]></description><category></category></item><item><title><![CDATA[pytorch学习笔记6--过拟合，交叉验证，正则化]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/97102333</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/97102333</guid><author>code_fighter</author><pubDate>Wed, 24 Jul 2019 09:31:43 +0800</pubDate><description><![CDATA[文章目录过拟合过拟合、欠拟合交叉验证regularization动量与学习率衰减dropout ，early stop卷积神经网络resnet  densenetnn.Module数据增强实战lenet5resnet
过拟合
过拟合、欠拟合
交叉验证
regularization

occam’s razor

more things should not be used than are nec...]]></description><category></category></item><item><title><![CDATA[pytorch学习笔记5--pytorch基本运算2]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/97101892</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/97101892</guid><author>code_fighter</author><pubDate>Wed, 24 Jul 2019 09:27:59 +0800</pubDate><description><![CDATA[文章目录感知机的梯度推导单输出单层感知机多输出感知机梯度推导链式法则多层感知机反向传播Logistic regressionQ1: why not maximize accuracy?Q2 : why call logistic regressionsoftmax交叉熵EntropyCross Entropy对于分类问题为什么不用MSEsoftmax 和 sigmoid？多分类全连接层nn.ReL...]]></description><category></category></item><item><title><![CDATA[Git & Github]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/97098827</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/97098827</guid><author>code_fighter</author><pubDate>Wed, 24 Jul 2019 08:43:23 +0800</pubDate><description><![CDATA[文章目录git本地库初始化 ： git init设置签名命令分支管理 git branc []Git 基本原理基本操作总结：githubGitLab
git

本地库初始化 ： git init
设置签名


形式

用户名：niklaus
email:niklaus.z.lee@gmail.com



作用:区分不同的开发人员身份


辨析：这里设置的签名和登陆远程库(代码托管中心)的账号、密...]]></description><category></category></item><item><title><![CDATA[pytorch学习笔记5--pytorch基本运算]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/91613400</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/91613400</guid><author>code_fighter</author><pubDate>Wed, 12 Jun 2019 20:26:28 +0800</pubDate><description><![CDATA[文章目录基本运算add/minus/multiply/dividematmul基本函数统计属性normmean,sum,min,max,prod,argmin,argmaxkthvalue,topk\&gt;,&gt;=,&lt;,&lt;=,!=,==高阶操作(GPU)whereGather梯度lossMean Squared Error(MSE)：自动求导：torch.autograd.gra...]]></description><category></category></item><item><title><![CDATA[pytorch学习笔记4--pytorch数据类型]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/91578381</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/91578381</guid><author>code_fighter</author><pubDate>Wed, 12 Jun 2019 17:49:35 +0800</pubDate><description><![CDATA[文章目录pytorch 数据类型创建tensor将numpy数据转化为Tensor：从python list中导入数据：uninitialized设置默认类型：随机初始化，rand/rand_like,randint正态分布：randn全部赋值为相同的元素：torch.arange(start,end,step)等分：torch.eye(d1,d2) or torch.eye(num)randpe...]]></description><category></category></item><item><title><![CDATA[pytorch学习笔记3--回归问题]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/91577440</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/91577440</guid><author>code_fighter</author><pubDate>Wed, 12 Jun 2019 17:45:25 +0800</pubDate><description><![CDATA[回归问题

linear regression

y∈[−∞,+∞]y \in [-\infty, +\infty]y∈[−∞,+∞]


logistics regression

把y值压缩到[0,1]，变为概率问题


代码：

loss=(WX+b−y)2loss = (WX+b-y)^2loss=(WX+b−y)2
代码实现如下：  def compute_error_for_line_...]]></description><category></category></item><item><title><![CDATA[pytorch学习笔记2--梯度下降的介绍]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/91572770</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/91572770</guid><author>code_fighter</author><pubDate>Wed, 12 Jun 2019 17:25:40 +0800</pubDate><description><![CDATA[梯度下降




给定y=x2∗sinxy=x^2*sinxy=x2∗sinx它的导数：y′=2xsinx+x2cosxy&amp;#x27;=2xsinx + x^2cosxy′=2xsinx+x2cosx则梯度下降对x进行更新：x′=x−lr∗Δxx&amp;#x27;=x-lr * \Delta xx′=x−lr∗Δx



噪声





example





optimize



...]]></description><category></category></item><item><title><![CDATA[pytorch学习笔记1--开发环境安装（windows）]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/91571115</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/91571115</guid><author>code_fighter</author><pubDate>Wed, 12 Jun 2019 17:19:23 +0800</pubDate><description><![CDATA[先附两个api
文章目录开发环境安装(windows)
开发环境安装(windows)

安装anaconda

勾选 Add Anaconda to my PATH environment variable


安装CUDA10

到CUDA官网下载，选择local 版本
安装过程中如果没有CUDA Visual Studio Integration 可以勾选然后跳过
如果在cuda&gt;v1...]]></description><category></category></item><item><title><![CDATA[python中append 和extend的区别]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/91569859</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/91569859</guid><author>code_fighter</author><pubDate>Wed, 12 Jun 2019 17:14:40 +0800</pubDate><description><![CDATA[append
a = [1,2,3]
b = [4,5,6]
a.append(b)
print a
结果：
[1, 2, 3, [4, 5, 6]]


extend
a = [1,2,3]
b = [4,5,6]
a.extend(b)
print a
结果：
[1, 2, 3, 4, 5, 6]


]]></description><category></category></item><item><title><![CDATA[scrapy 教程]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/91569072</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/91569072</guid><author>code_fighter</author><pubDate>Wed, 12 Jun 2019 17:11:53 +0800</pubDate><description><![CDATA[中文文档：
scrapy 4 步：


新建项目：（scrapy startproject projectname）:新建爬虫项目


创建爬虫：scrapy genspider  spidername “http://www.xxx.cn/”


明确目标：（编写items.py）：明确想要抓取的目标


制作爬虫：（spiders/xxspider.py）:制作爬虫开始爬取的网页


编写p...]]></description><category></category></item><item><title><![CDATA[自定义安装office教程]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/91567529</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/91567529</guid><author>code_fighter</author><pubDate>Wed, 12 Jun 2019 17:07:01 +0800</pubDate><description><![CDATA[文章目录卸载旧版本office下载office Deployment Tool（ODT）创建/修改配置文件
卸载旧版本office

控制面板–卸载–office–卸载

下载office Deployment Tool（ODT）

微软官方下载地址
执行下载程序，执行后，会输出4个文件：configuration-Office365-x64.xml、configuration-Office365...]]></description><category></category></item><item><title><![CDATA[latex语法]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/91565892</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/91565892</guid><author>code_fighter</author><pubDate>Wed, 12 Jun 2019 17:01:49 +0800</pubDate><description><![CDATA[文章目录上下标括号分数开方对数向量三角运算省略号需要转义的字符普通符号希腊字母集合运算逻辑运算微积分箭头求和、连乘和积分号数字模式重音符连线符号矩阵
上下标




命令
显示




x_{2}
x2x_{2}x2​


x^2
x2x^{2}x2


括号

() [] 直接写就行，而 {} 则需要转义

y \in {3, 4, 5} : y∈{3,4,5}y \in \{3, 4, 5\}...]]></description><category></category></item><item><title><![CDATA[Large-Scale Long-Tailed Recognition in an OpenWorld]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/90582730</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/90582730</guid><author>code_fighter</author><pubDate>Sun, 26 May 2019 23:42:37 +0800</pubDate><description><![CDATA[Open Long-Tailed Recognition(OLTR):从自然分布的数据中学习并在包括头、尾和开放类的平衡测试集上优化分类精度。也就是对三个问题的综合

imbalanced classification
few-shot learning
open-set recognition.


论文针对的问题：

OLTR不仅要在封闭环境中处理不平衡分类和few-shot learning...]]></description><category></category></item><item><title><![CDATA[vim常用命令]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/90579691</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/90579691</guid><author>code_fighter</author><pubDate>Sun, 26 May 2019 19:59:42 +0800</pubDate><description><![CDATA[文章目录vim命令模式可视模式删除命令撤销命令重复命令文本移动复制粘贴查找命令替换操作替换命令输入模式末行模式补充
vim
命令模式

h：左移
j：下移
k：上移
l(L)：右移
M：光标移动到中间行
L：光标移动到屏幕最后一行行首
G：移动到指定行，行号 -G，只有G的话，到文件最末尾
gg：文件第一个字符
w:向后一次移动一个字
b：向前一次移动一个字
{：按段移动，上移
}：按段下移，下移...]]></description><category></category></item><item><title><![CDATA[linux的文件系统]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/90578146</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/90578146</guid><author>code_fighter</author><pubDate>Sun, 26 May 2019 17:23:18 +0800</pubDate><description><![CDATA[/boot:系统启动相关的文件，如内核、initrd、以及grub（bootloader）
/dev:设备文件：

块设备：随机访问，数据块
字符设备：线性访问，按字符为单位
设备号：主设备号（major）和次设备号（minor）


/etc：配置文件
/home：用户的家目录，每一个用户的家目录通常默认为/home/username
/root：管理员的家目录；
/lib:库文件

静态库：...]]></description><category></category></item><item><title><![CDATA[SEMI-SUPERVISED CLASSIFICATION WITH GRAPH CONVOLUTIONAL NETWORKS]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/84779949</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/84779949</guid><author>code_fighter</author><pubDate>Mon, 03 Dec 2018 21:29:46 +0800</pubDate><description><![CDATA[


震

]]></description><category></category></item><item><title><![CDATA[A Generative Adversarial Approach for Zero-Shot Learning from Noisy Texts]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/83990592</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/83990592</guid><author>code_fighter</author><pubDate>Mon, 12 Nov 2018 13:39:35 +0800</pubDate><description><![CDATA[




震

]]></description><category></category></item><item><title><![CDATA[Discriminative Learning of Latent Features for Zero-Shot Recognition]]></title><link>https://blog.youkuaiyun.com/code_fighter/article/details/83655451</link><guid>https://blog.youkuaiyun.com/code_fighter/article/details/83655451</guid><author>code_fighter</author><pubDate>Fri, 02 Nov 2018 14:56:29 +0800</pubDate><description><![CDATA[



震

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