# HBPU-AI **Repository Path**: CoreTheGreat/HBPU-AI ## Basic Information - **Project Name**: HBPU-AI - **Description**: 湖北理工学院计算机科学与技术人工智能课程 - **Primary Language**: Python - **License**: MIT - **Default Branch**: main - **Homepage**: None - **GVP Project**: No ## Statistics - **Stars**: 10 - **Forks**: 15 - **Created**: 2024-07-04 - **Last Updated**: 2026-04-17 ## Categories & Tags **Categories**: Uncategorized **Tags**: None ## README # HBPU-AI This repository is built for students of Hubei Polytechnic Unersity to learning artificial intelligence. Metierials link of Machine Learning 2024: https://pan.baidu.com/s/1Tv_VCZxnoBmxC-z9qimyiQ?pwd=2493 Metierials link of Artificial Intelligence 2025: https://pan.baidu.com/s/1IxueuU_ioVRQVJQhvtFxOg?pwd=c8nj Metierials link of Artificial Intelligence 2026: https://pan.baidu.com/s/1gYJPvpX0N9OsuMgOXRK04A?pwd=dwt8 # 工程文件夹说明 - AI_2025: 2025年课件 - AI_2026: 2026年课件 - Machine_Learning_2024: 2024年《机器学习》课件及实验 - data:实验数据 - fonts:实验所需字体 - images:工程图片 # 参考资料与致敬 本课程参考以下文献以及多媒体资源,向原作者表达崇高敬意。 ## 📚 参考教材 - 张奇,桂涛,郑锐,黄萱菁. 大语言模型从理论到实践(第二版).电子工业出版社,2025. - 教材代码仓库: - (国际)[https://github.com/intro-llm/intro-llm-code](https://github.com/intro-llm/intro-llm-code) - (国内)[https://gitee.com/CoreTheGreat/intro-llm-code](https://gitee.com/CoreTheGreat/intro-llm-code) ## 🎥 参考视频资源 - 李宏毅.智能体2026系列课程:[BV1o3wvzUEDD(哔哩哔哩)](https://www.bilibili.com/video/BV1o3wvzUEDD) - 李宏毅.人工智能2025系列课程:[BV1oduvzPEk4(哔哩哔哩)](https://www.bilibili.com/video/BV1oduvzPEk4) - 李宏毅.生成式AI导论2024:[BV1XS411w7qr(哔哩哔哩)](https://www.bilibili.com/video/BV1XS411w7qr) - 李宏毅YouTube官网:[https://www.youtube.com/@HungyiLeeNTU](https://www.youtube.com/@HungyiLeeNTU) - 吴恩达. 2026智能体系列课程:[BV1DfrdByE2H(哔哩哔哩)](https://www.bilibili.com/video/BV1DfrdByE2H) - 吴恩达. 2025生成式人工智能:[BV1sMEyzhEM3(哔哩哔哩)](https://www.bilibili.com/video/BV1sMEyzhEM3) - 吴恩达Agentic AI官网:[https://www.deeplearning.ai/courses/agentic-ai](https://www.deeplearning.ai/courses/agentic-ai) ## 💻 实验参考代码仓库 - **LLM & Agent 核心技术实战教程** - Greebie/LLM-Agent-Core_Concept_Code: [https://github.com/Greebbie/LLM-Agent-Core_Concept_Code](https://github.com/Greebbie/LLM-Agent-Core_Concept_Code) - (Gitee Mirror): [https://gitee.com/CoreTheGreat/LLM-Agent-Core_Concept_Code](https://gitee.com/CoreTheGreat/LLM-Agent-Core_Concept_Code) - **LLM Course** - Mlabonne/llm-course: [https://github.com/mlabonne/llm-course](https://github.com/mlabonne/llm-course) - (Gitee Mirror): [https://gitee.com/CoreTheGreat/llm-course](https://gitee.com/CoreTheGreat/llm-course) - **从0开始训练自己的Phi2中文小模型** - charent/Phi2-mini-Chinese: [https://github.com/CoreTheGreat/Phi2-mini-Chinese](https://github.com/CoreTheGreat/Phi2-mini-Chinese) - (Gitee Mirror): [https://gitee.com/CoreTheGreat/Phi2-mini-Chinese](https://gitee.com/CoreTheGreat/Phi2-mini-Chinese) - **中文对话0.2B小模型** - charent/ChatLM-mini-Chinese: [https://github.com/charent/ChatLM-mini-Chinese](https://github.com/charent/ChatLM-mini-Chinese) - (Gitee Mirror): [https://gitee.com/CoreTheGreat/ChatLM-mini-Chinese](https://gitee.com/CoreTheGreat/ChatLM-mini-Chinese) - **OpenManus** - FoundationAgents/OpenManus: [https://github.com/FoundationAgents/OpenManus](https://github.com/FoundationAgents/OpenManus) - (Gitee Mirror): [https://gitee.com/CoreTheGreat/OpenManus](https://gitee.com/CoreTheGreat/OpenManus) - **System Prompts Leaks** - asgeirtj/system_prompts_leaks: [https://github.com/asgeirtj/system_prompts_leaks](https://github.com/asgeirtj/system_prompts_leaks) - (Gitee Mirror): [https://gitee.com/CoreTheGreat/system_prompts_leaks](https://gitee.com/CoreTheGreat/system_prompts_leaks) # 实验环境(Anaconda + VS Code)安装参考博客 博客园:https://www.cnblogs.com/superhui/category/2492198.html CSDN:https://blog.csdn.net/south_rosefinch/category_13120133.html > 参考来自肖彭辉 ## 🛠️ 环境准备(Conda + CPU 版本) 默认前提: 1. 已安装 Anaconda / Miniconda 2. 已安装VS Code,或Cursor、Antigravity、TRAE(这个IDE需同学们自行测试,如果可行记得告诉我) 3. `conda venv` 还未创建 4. 相关包尚未安装 ### 1. 创建 Conda 虚拟环境并拉取代码 运行以下代码创建虚拟环境(环境中没有依赖包) ```bash conda create -n hbpu-ai python=3.11 -y conda activate hbpu-ai ``` 进入git工作目录拉取代码 ```bash # 码云(国内) git clone https://gitee.com/CoreTheGreat/HBPU-AI.git # 或者 # GitHub(国外) git clone https://github.com/CoreTheGreat/HBPU-AI.git ``` ### 2. 安装依赖包 ### 2.1 可选:配置国内镜像(更快) #### Conda 镜像 可先临时使用: ```bash conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/ conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/ conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/r/ conda config --set show_channel_urls yes ``` ### 2.2 验证国内镜像是否生效 - Conda 频道验证: - `conda config --show channels`(查看已添加镜像优先级) - `conda search numpy`(如果来自镜像源即成功) - pip 源验证: - `pip config get global.index-url`(如果配置了全局源) - `pip install -U pip -i https://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com`(看下载源是否为阿里) 或直接在安装时观察输出: - `Looking in indexes: https://mirrors.aliyun.com/pypi/simple/` 表示镜像成功。 ### 3. 安装依赖 ```bash pwd ``` 确保当前路径在"./HBPU-AI" ```bash pip install -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com # 或者清华源 pip install -r requirements.txt -i https://pypi.tuna.tsinghua.edu.cn/simple/ --trusted-host pypi.tuna.tsinghua.edu.cn ``` TIPS:如果你的电脑有独立显卡,可以安装GPU版本的PyTorch ### 4. 运行环境验证 1. 在环境中运行 `environment_validation.ipynb` 2. 或在终端中执行"python"进入python观景,并输出以下脚本检查版本 ```bash import torch, transformers, accelerate, sentence_transformers, datasets print('torch', torch.__version__) print('transformers', transformers.__version__) print('accelerate', accelerate.__version__) print('sentence_transformers', sentence_transformers.__version__) print('datasets', datasets.__version__) ``` ### 5. 工程拉取补充说明 models/uer_gpt2_chinese/model.safetensors等模型文件过大无法上传,请自行下载 1. 可运行.ipynb中的相关代码下载模型 2. 从网盘链接手动下载至models目录下 网盘链接: https://pan.baidu.com/s/1gYJPvpX0N9OsuMgOXRK04A?pwd=dwt8