<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[m0_62053105的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/m0_62053105</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; m0_62053105]]></copyright><item><title><![CDATA[学习Transformer必看]]></title><link>https://blog.youkuaiyun.com/m0_62053105/article/details/140592806</link><guid>https://blog.youkuaiyun.com/m0_62053105/article/details/140592806</guid><author>m0_62053105</author><pubDate>Sun, 21 Jul 2024 19:56:16 +0800</pubDate><description><![CDATA[这是一个搬运，发现了一篇关于学习transformer很好的教程：The Annotated Transformer]]></description><category></category></item><item><title><![CDATA[什么是ABC类（abstract base class）]]></title><link>https://blog.youkuaiyun.com/m0_62053105/article/details/140387691</link><guid>https://blog.youkuaiyun.com/m0_62053105/article/details/140387691</guid><author>m0_62053105</author><pubDate>Fri, 12 Jul 2024 20:04:38 +0800</pubDate><description><![CDATA[这个ABC其实全称就是abstract base class，也叫做抽象基类。那么这个ABC有什么用呢？大家肯定经常在一些代码中看到下面这样一行东西。这样应该就清楚一些了。]]></description><category></category></item><item><title><![CDATA[sliding window attention]]></title><link>https://blog.youkuaiyun.com/m0_62053105/article/details/140154494</link><guid>https://blog.youkuaiyun.com/m0_62053105/article/details/140154494</guid><author>m0_62053105</author><pubDate>Wed, 03 Jul 2024 18:16:30 +0800</pubDate><description><![CDATA[同时，这种方式并不是意味着当前token只能获取到前window_size个token的信息，因为当前token前面的window_size个token也都是能够获取到前面的信息的，因此只要网络达到一定的深度，这样的sliding window attention是可行的，并不会损失太多信息。sliding window attention是为了解决在输出序列长度sequence length很大的时候attention计算量爆炸增长的问题。]]></description><category></category></item><item><title><![CDATA[搬运工合集]]></title><link>https://blog.youkuaiyun.com/m0_62053105/article/details/138428863</link><guid>https://blog.youkuaiyun.com/m0_62053105/article/details/138428863</guid><author>m0_62053105</author><pubDate>Sat, 04 May 2024 00:36:09 +0800</pubDate><description><![CDATA[当个搬运工：https://www.cnblogs.com/luzhuangzhi/p/17869210.html]]></description><category></category></item><item><title><![CDATA[langchain学习记录]]></title><link>https://blog.youkuaiyun.com/m0_62053105/article/details/137798860</link><guid>https://blog.youkuaiyun.com/m0_62053105/article/details/137798860</guid><author>m0_62053105</author><pubDate>Mon, 15 Apr 2024 22:50:07 +0800</pubDate><description><![CDATA[开一个帖子记录自己学习langchain工具使用的经历，先开贴占个坑，也督促自己好好学习一下。]]></description><category></category></item><item><title><![CDATA[Sequence to Sequence Learningwith Neural Networks论文阅读]]></title><link>https://blog.youkuaiyun.com/m0_62053105/article/details/137149513</link><guid>https://blog.youkuaiyun.com/m0_62053105/article/details/137149513</guid><author>m0_62053105</author><pubDate>Fri, 29 Mar 2024 17:05:44 +0800</pubDate><description><![CDATA[简要概述阅读论文Sequence to Sequence Learning with Neural Networks]]></description><category></category></item><item><title><![CDATA[cocos加载图片]]></title><link>https://blog.youkuaiyun.com/m0_62053105/article/details/135937059</link><guid>https://blog.youkuaiyun.com/m0_62053105/article/details/135937059</guid><author>m0_62053105</author><pubDate>Tue, 30 Jan 2024 17:02:31 +0800</pubDate><description><![CDATA[cocos  resources.load加载图片]]></description><category></category></item><item><title><![CDATA[transformer库generate函数参数测试]]></title><link>https://blog.youkuaiyun.com/m0_62053105/article/details/134519983</link><guid>https://blog.youkuaiyun.com/m0_62053105/article/details/134519983</guid><author>m0_62053105</author><pubDate>Mon, 20 Nov 2023 23:03:08 +0800</pubDate><description><![CDATA[第一篇文章想要记录一下自己在玩transformer库中模型"gpt2"进行生成文字过程中调整generate函数参数对生成的文字的影响。]]></description><category></category></item></channel></rss>