<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[Abibulla的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/Abibulla</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; Abibulla]]></copyright><item><title><![CDATA[Anaconda3 目录迁移教程]]></title><link>https://blog.youkuaiyun.com/Abibulla/article/details/129669567</link><guid>https://blog.youkuaiyun.com/Abibulla/article/details/129669567</guid><author>Abibulla</author><pubDate>Mon, 20 Mar 2023 16:09:47 +0800</pubDate><description><![CDATA[由于服务器home目录存储空间已满，发现anaconda3文件占的空间最大。]]></description><category></category></item><item><title><![CDATA[ubuntu conda使用报错：An unexpected error has occurred. Conda has prepared the above report.]]></title><link>https://blog.youkuaiyun.com/Abibulla/article/details/118789399</link><guid>https://blog.youkuaiyun.com/Abibulla/article/details/118789399</guid><author>Abibulla</author><pubDate>Fri, 16 Jul 2021 12:03:39 +0800</pubDate><description><![CDATA[ubuntu conda使用报错：An unexpected error has occurred. Conda has prepared the above report.

非常简单：删除用户目录下的~~.condarc~~ 文件。

]]></description><category></category></item><item><title><![CDATA[pytorch之squeeze和unsqueeze的用法及注意事项]]></title><link>https://blog.youkuaiyun.com/Abibulla/article/details/117931154</link><guid>https://blog.youkuaiyun.com/Abibulla/article/details/117931154</guid><author>Abibulla</author><pubDate>Tue, 15 Jun 2021 19:28:36 +0800</pubDate><description><![CDATA[pytorch之squeeze和unsqueeze的用法及注意事项用法注意事项
用法

squeeze：对数据的维度进行压缩，去掉维数为1的的维度，比如是一行或者一列这种，一个一行三列(1, 10)的数去掉第一个维数为一的维度之后就变成(10)行。用法两种:


b= a.squeeze(dim=index) 输入需要降维的维度index
b = torch.squeeze(a, dim=index) 同上

a = torch.randn([1, 10], )
b= a.squeeze(0)
print(]]></description><category></category></item><item><title><![CDATA[pytorch中contiguous()的作用及使用]]></title><link>https://blog.youkuaiyun.com/Abibulla/article/details/111347744</link><guid>https://blog.youkuaiyun.com/Abibulla/article/details/111347744</guid><author>Abibulla</author><pubDate>Thu, 17 Dec 2020 22:58:17 +0800</pubDate><description><![CDATA[方法介绍
touch.view()方法对张量改变“形状”其实并没有改变张量在内存中真正的形状。
简单地说，view方法没有拷贝新的张量，没有开辟新内存，与原张量共享内存，只是重新定义了访问张量的规则，使得取出的张量按照我们希望的形状展现。
举例说，如下代码：
t = torch.tensor([[0, 1, 2, 3], [4, 5, 6, 8], [9, 10, 11, 12]])
t2 = t.transpose(0, 1)
print(t2)
tensor([[ 0,  4,  9],
       ]]></description><category></category></item><item><title><![CDATA[Attention Is All You Need----Transformer]]></title><link>https://blog.youkuaiyun.com/Abibulla/article/details/110450526</link><guid>https://blog.youkuaiyun.com/Abibulla/article/details/110450526</guid><author>Abibulla</author><pubDate>Tue, 01 Dec 2020 20:16:51 +0800</pubDate><description><![CDATA[转发：Attention Is All You Need----Transformer写的很详细。

]]></description><category></category></item><item><title><![CDATA[Seq2Seq模型及Attention机制]]></title><link>https://blog.youkuaiyun.com/Abibulla/article/details/110429859</link><guid>https://blog.youkuaiyun.com/Abibulla/article/details/110429859</guid><author>Abibulla</author><pubDate>Tue, 01 Dec 2020 13:49:55 +0800</pubDate><description><![CDATA[注意力机制讲的最完整的博客
转发：Seq2Seq模型及Attention机制

]]></description><category></category></item><item><title><![CDATA[latex学习：texlive+texstudio安装]]></title><link>https://blog.youkuaiyun.com/Abibulla/article/details/110240591</link><guid>https://blog.youkuaiyun.com/Abibulla/article/details/110240591</guid><author>Abibulla</author><pubDate>Fri, 27 Nov 2020 21:46:58 +0800</pubDate><description><![CDATA[latex学习：texlive+texstudio安装前言一、资源分享二、安装步骤1.texstudio安装2.texlive安装

前言
最近写了一篇开题前的综述类文章，所以顺便学习了Latex。他确实很方便，作者不需要关注排版格式等问题，只需要关心文章内容。本篇记录texlive+texstudio安装

一、资源分享
链接：latex资源包：https://pan.baidu.com/s/1VRGEAfdWz0D2_wmJPvTLDg 提取码：aibi
二、安装步骤
1.texstudio安装
按其它]]></description><category></category></item><item><title><![CDATA[科研那些事儿：开题总结]]></title><link>https://blog.youkuaiyun.com/Abibulla/article/details/110233457</link><guid>https://blog.youkuaiyun.com/Abibulla/article/details/110233457</guid><author>Abibulla</author><pubDate>Fri, 27 Nov 2020 18:32:44 +0800</pubDate><description><![CDATA[硕士研究生开题注意事项（总结）
一.研究内容方面
1. 首先大部分研究需要落地到应用上来的，所以选题必须考虑得考虑实际应用问题。
2. 对工作量进行大概的评估。刚上研究生自己很有激情，但是实际上研究生研一几乎是基础课程，研二开始才算是进入初步研究阶段。所以对这个阶段，学生（比如本人）会有很大的期望，所以在研究领域想做一些新的事情或者创新。其实研究应该就这么做，但是硕士阶段不是很长，所以需要考虑时间的问题，当然了转博或者继续攻读博士的话，是有比较长的做研究的时间。所以需要对工作量做一个评估，其实评估就是一个难]]></description><category></category></item><item><title><![CDATA[Latex笔记]]></title><link>https://blog.youkuaiyun.com/Abibulla/article/details/109738931</link><guid>https://blog.youkuaiyun.com/Abibulla/article/details/109738931</guid><author>Abibulla</author><pubDate>Tue, 17 Nov 2020 19:51:13 +0800</pubDate><description><![CDATA[Latex学习
1.累乘输入：
\begin{equation}
    p(Y|X,B)=\prod_{t=1}^n \ p(y_{t},|y_{1},y_{2},...,y_{t-1},X,B)
\end{equation} 

结果：

2.连续来那行公式输入*\begin{gather}*
	p_{intent}(d|x_{1},x_{2},...,x_{n})	\\
	p_{s-v}(y_{1},y_{2},...,y_{n}|x_{1},x_{2},...,x_{n})	
\end{gather]]></description><category></category></item><item><title><![CDATA[Resource xxx not found.Please use the NLTK Downloader to obtain the resource:]]></title><link>https://blog.youkuaiyun.com/Abibulla/article/details/109721438</link><guid>https://blog.youkuaiyun.com/Abibulla/article/details/109721438</guid><author>Abibulla</author><pubDate>Mon, 16 Nov 2020 15:22:00 +0800</pubDate><description><![CDATA[场景：
使用nltk库的时候找不到资源，即：Resource xxx  not found.Please use the NLTK Downloader to obtain the resource:

描述：
问题如下：
**********************************************************************
  Resource punkt not found.
  Please use the NLTK Downloader to obtain th]]></description><category></category></item><item><title><![CDATA[PermissionError: [Errno 13] Permission denied: ‘./models/‘]]></title><link>https://blog.youkuaiyun.com/Abibulla/article/details/109696678</link><guid>https://blog.youkuaiyun.com/Abibulla/article/details/109696678</guid><author>Abibulla</author><pubDate>Sat, 14 Nov 2020 21:18:32 +0800</pubDate><description><![CDATA[问题记录：PermissionError: [Errno 13] Permission denied: './models/'场景：原因分析：解决方案：参考
场景：
ubuntu18.04 模型训练，没有权限的文件。

# 描述：
不需要root权限的文件不能访问：
错误提示：   PermissionError: [Errno 13] Permission denied: ‘./models/’

原因分析：
可能是系统默认这是一个高风险目录，Ubuntu linux ls 看到目录文件夹绿色高亮问题

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