重要声明

当时尝试微调模型是为了了解【特定领域模型微调的实际工作流程】，在进行实际的微调工作时，为了方便和网友交流，因此 jupyter notebook 中均使用英文。但遣词造句均比较基础，如果阅读实在困难，可以使用翻译插件。就改回来了，实在实在厌恶重复工作。

准备工作

首先声明一些重要且必须要做的准备基础工作如下。且需要注意，在安装软件时，需要严格注意【软件版本】，之前试过由于软件版本不同导致的计算数值差异：

Nvidia Driver and Cuda toolkit: choose a proper driver for your Nvidia GPU and installcuda 12.4 toolkit
Anaconda: we need this to create virtual environment with python3.10, and install PyTorch and Unsloth
git and cmake: when we save finetuned model into gguf format, we need to callsave_pretrained_gguf()method and the method will clonellama.cppfrom github and compile it. you can download it from thegit offical siteandcmake offical siteand append their bin folders after windows'path
ollama and Page Assist: we use ollama to run finetuned model and Page Assist for a friendly UI. You can download ollama ->hereand install Page Assist from ->here
Triton wheel for python 3.10: if you are using a windows pc, you will need to download thetriton-windows-buildsand install it later viapip install triton-3.0.0-cp310-cp310-win_amd64.whl