<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[YELLOWAUG的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/YELLOWAUG</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; YELLOWAUG]]></copyright><item><title><![CDATA[GLM3-6B微调实践-从设计数据集开始针对一个下游业务进行微调]]></title><link>https://blog.youkuaiyun.com/YELLOWAUG/article/details/140135381</link><guid>https://blog.youkuaiyun.com/YELLOWAUG/article/details/140135381</guid><author>YELLOWAUG</author><pubDate>Tue, 02 Jul 2024 21:57:35 +0800</pubDate><description><![CDATA[记录一次微调下游任务的核心思路，使用glm3-6b模型以及Llama-factory进行微调。]]></description><category></category></item><item><title><![CDATA[本地容器化快速部署GLM-4-9B-chat教程]]></title><link>https://blog.youkuaiyun.com/YELLOWAUG/article/details/139526214</link><guid>https://blog.youkuaiyun.com/YELLOWAUG/article/details/139526214</guid><author>YELLOWAUG</author><pubDate>Fri, 07 Jun 2024 14:38:32 +0800</pubDate><description><![CDATA[glm-4-9b-chat 容器化快速部署]]></description><category></category></item><item><title><![CDATA[多少数据微调LLM模型才有意义？记录一次微调ChatGLM3-6B模型]]></title><link>https://blog.youkuaiyun.com/YELLOWAUG/article/details/139029001</link><guid>https://blog.youkuaiyun.com/YELLOWAUG/article/details/139029001</guid><author>YELLOWAUG</author><pubDate>Sat, 18 May 2024 17:52:08 +0800</pubDate><description><![CDATA[到底需要在什么样的场景下对大模型进行微调]]></description><category></category></item><item><title><![CDATA[GLM3-6B微调实践-从0开始，搭建Web页面的训练平台]]></title><link>https://blog.youkuaiyun.com/YELLOWAUG/article/details/138367165</link><guid>https://blog.youkuaiyun.com/YELLOWAUG/article/details/138367165</guid><author>YELLOWAUG</author><pubDate>Wed, 01 May 2024 10:23:23 +0800</pubDate><description><![CDATA[本文介绍的是使用docker 的方式快速部署LLaMA-Factory,并进行初步体验]]></description><category></category></item><item><title><![CDATA[GLM4实践-GLM4智能体的本地化实现及部署]]></title><link>https://blog.youkuaiyun.com/YELLOWAUG/article/details/137336638</link><guid>https://blog.youkuaiyun.com/YELLOWAUG/article/details/137336638</guid><author>YELLOWAUG</author><pubDate>Wed, 03 Apr 2024 11:23:39 +0800</pubDate><description><![CDATA[实现GLM4智能体的本地化部署及调用，本文较长，涉及的技术点也较多]]></description><category></category></item><item><title><![CDATA[Grok模型初步评测-部署Grok模型的硬件配置]]></title><link>https://blog.youkuaiyun.com/YELLOWAUG/article/details/136835736</link><guid>https://blog.youkuaiyun.com/YELLOWAUG/article/details/136835736</guid><author>YELLOWAUG</author><pubDate>Tue, 19 Mar 2024 11:00:39 +0800</pubDate><description><![CDATA[马斯克最新模型grok本地化部署需要的配置以及预算的估算]]></description><category></category></item><item><title><![CDATA[GLM4智能体实践-通过GLM4 API构建线上智能体]]></title><link>https://blog.youkuaiyun.com/YELLOWAUG/article/details/136572956</link><guid>https://blog.youkuaiyun.com/YELLOWAUG/article/details/136572956</guid><author>YELLOWAUG</author><pubDate>Fri, 08 Mar 2024 22:20:50 +0800</pubDate><description><![CDATA[使用GLM4 API 构建自定义知识库及智能体。图文教程，简单易懂]]></description><category></category></item><item><title><![CDATA[GLM4智能体创建实践]]></title><link>https://blog.youkuaiyun.com/YELLOWAUG/article/details/136355159</link><guid>https://blog.youkuaiyun.com/YELLOWAUG/article/details/136355159</guid><author>YELLOWAUG</author><pubDate>Thu, 29 Feb 2024 20:31:11 +0800</pubDate><description><![CDATA[GLM4智能体创建实践以及指南，GLM4是模型国内唯一支持定制智能体的大预言模型，百度的文心，阿里千问，这些公司都是不支持的。]]></description><category></category></item><item><title><![CDATA[Langchain最小化部署（调用线上大模型API，无本地知识库）]]></title><link>https://blog.youkuaiyun.com/YELLOWAUG/article/details/136241731</link><guid>https://blog.youkuaiyun.com/YELLOWAUG/article/details/136241731</guid><author>YELLOWAUG</author><pubDate>Fri, 23 Feb 2024 15:33:49 +0800</pubDate><description><![CDATA[langchat-chatchat 最小化部署，支持使用多种线上大模型的API，适合无显卡又想体验多种大模型的用户]]></description><category></category></item><item><title><![CDATA[Langchain架构原理浅谈]]></title><link>https://blog.youkuaiyun.com/YELLOWAUG/article/details/136219542</link><guid>https://blog.youkuaiyun.com/YELLOWAUG/article/details/136219542</guid><author>YELLOWAUG</author><pubDate>Wed, 21 Feb 2024 20:02:25 +0800</pubDate><description><![CDATA[所以Embedding model模型的选择很重要，用户输入查询或者问题信息，通过嵌入器转换成向量，把查询信息的向量与知识库的向量做比较，并通过提示词的形式输入LLM模型，最终通过LLM模型得到大胆。在使用Langchain的过程中发现，Langchain的原理图使用的是英文图片，流程的描述用词过于专业，所以这里我就用自己的理解来简单的表述Langchain的工作过程，希望对大家在使用Langchain开发应用的过程中有所帮助。在处理文本时，每个单词或短语都被转换为一个固定大小的向量。]]></description><category></category></item><item><title><![CDATA[Docker compose 容器编排实战系列-一键部署redis]]></title><link>https://blog.youkuaiyun.com/YELLOWAUG/article/details/135961567</link><guid>https://blog.youkuaiyun.com/YELLOWAUG/article/details/135961567</guid><author>YELLOWAUG</author><pubDate>Wed, 31 Jan 2024 21:04:51 +0800</pubDate><description><![CDATA[一键部署redis 单节点， 集群模式（主从复制）]]></description><category></category></item><item><title><![CDATA[docker compose 官方文档翻译（20240129更新）]]></title><link>https://blog.youkuaiyun.com/YELLOWAUG/article/details/135853631</link><guid>https://blog.youkuaiyun.com/YELLOWAUG/article/details/135853631</guid><author>YELLOWAUG</author><pubDate>Thu, 25 Jan 2024 21:08:14 +0800</pubDate><description><![CDATA[docker compose 中文文档。]]></description><category></category></item><item><title><![CDATA[Docker compose 容器编排实战系列-部署Wordpress]]></title><link>https://blog.youkuaiyun.com/YELLOWAUG/article/details/135787117</link><guid>https://blog.youkuaiyun.com/YELLOWAUG/article/details/135787117</guid><author>YELLOWAUG</author><pubDate>Tue, 23 Jan 2024 20:34:45 +0800</pubDate><description><![CDATA[文件内容如下，在linux主机上创建wordpress.yml 文件，例如vim /home/wordpress.yml。由于我们不是使用传统的docker-compose.yml 文件命名，而是自定义命名，所以需要使用以下命令启动。]]></description><category></category></item></channel></rss>