{"id":14989249,"url":"https://github.com/orvillex/machinelearning","last_synced_at":"2025-04-09T14:13:03.412Z","repository":{"id":184853644,"uuid":"293269249","full_name":"OrvilleX/MachineLearning","owner":"OrvilleX","description":"本项目以应用为主出发，结合了从基础的机器学习、深度学习到目标检测以及目前最新的大模型，采用目前成熟的 第三方库、开源预训练模型以及相关论文的最新技术，目的是记录学习的过程同时也进行分享以供更多人可以直接进行使用。","archived":false,"fork":false,"pushed_at":"2025-03-02T14:11:22.000Z","size":11965,"stargazers_count":66,"open_issues_count":1,"forks_count":22,"subscribers_count":3,"default_branch":"master","last_synced_at":"2025-04-09T14:12:55.088Z","etag":null,"topics":["knn","llm","machine-learning","mllm","numpy","scipy","siglip","sklearn","spark-mllib","svm","tensorflow"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/OrvilleX.png","metadata":{"files":{"readme":"ReadMe.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2020-09-06T12:10:44.000Z","updated_at":"2025-03-17T07:57:28.000Z","dependencies_parsed_at":"2023-07-30T16:37:18.882Z","dependency_job_id":"1ee29816-f35a-42b5-8a6c-e65c9caa747f","html_url":"https://github.com/OrvilleX/MachineLearning","commit_stats":{"total_commits":98,"total_committers":3,"mean_commits":"32.666666666666664","dds":"0.15306122448979587","last_synced_commit":"2e15444aa9ad15731706bb73006cc5994ab17fce"},"previous_names":["orvillex/machinelearning"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/OrvilleX%2FMachineLearning","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/OrvilleX%2FMachineLearning/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/OrvilleX%2FMachineLearning/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/OrvilleX%2FMachineLearning/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/OrvilleX","download_url":"https://codeload.github.com/OrvilleX/MachineLearning/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248054193,"owners_count":21039952,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["knn","llm","machine-learning","mllm","numpy","scipy","siglip","sklearn","spark-mllib","svm","tensorflow"],"created_at":"2024-09-24T14:17:56.304Z","updated_at":"2025-04-09T14:13:03.393Z","avatar_url":"https://github.com/OrvilleX.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 人工智能集合 \n\n本项目以应用为主出发，结合了从基础的机器学习、深度学习到目标检测以及目前最新的大模型，采用目前成熟的\n第三方库、开源预训练模型以及相关论文的最新技术，目的是记录学习的过程同时也进行分享以供更多人可以直接\n进行使用。  \n\n\u003e 本人自己目前属于自己创业，目前要时围绕各类算法场景的应用开发，目前主要的领域为船舶、教育以及企业定制的开发  \n\n## 一、目录  \n\n对应每个案例将采用独立的文件夹的方式进行管理，非源码的可以参考对应的文档进行相关依赖的安装，部分存在源码的则可以\n通过源码中对应的requirements.txt安装对应的依赖。  \n\n\n* [TTS解决方案](#tts解决方案)\n* [ASR解决方案](#asr解决方案)\n* [图片特征提取](#图片特征提取)\n\n### 机器学习基础\n\n* [基于numpy实现的机器学习算法](./numpy/ReadMe.md): 主要是讲述底层的算法的逻辑，实际使用中往往采用第三方库来实现  \n* [基于sklearn的机器学习算法](./sklearn/ReadMe.md): 主要是讲述如何使用第三方类库快速使用成熟的算法  \n* [预处理技术](./preprocessing/ReadMe.md): 其主要包含针对机器学习工程中针对数据的预处理的部分的算法  \n* [特征工程](./featureengineering/ReadMe.md): 主要是围绕各类数据分析场景下针对数据的特征表示的算法  \n\n### 数据挖掘\n\n* [挖掘频繁项集](./frequentItemsets/ReadMe.md): 主要是采用numpy与sklearn的方式实现这类算法    \n\n### TTS解决方案\n\n* [Kokore适合边缘设备的TTS解决方案](./kokore/ReadMe.md)  \n\n### ASR解决方案\n\n* [基于faster-whisper的实时音频处理](./voice/whisper/README.md)\n\n### 图片特征提取\n\n* [SigLIP 图文对照模型](./siglip/ReadMe.md): 大量的多模态模型的图像特种提取必使用的模型，本文档基于目前主流的`siglip-so400m-patch14-384`模型进行编写，开发多模态大模型必须掌握的图像特征提取库\n\n* [InternVideo2 多模态视频理解模型](./internvideo/ReadMe.md): 由于上海人工智能实验室（General Vision Team of Shanghai AI Laboratory）推出的针对视频理解的模型，目前针对视频理解的论文逐渐将其作为融合siglip来实现针对视频\u0026图片场景的多模态大模型的基础组件  \n\n### 目标检测技术\n\n* [基于yolo目标检测系列](./yolo/ReadMe.md)  \n* [DETR技术的应用方式](./detr/ReadMe.md)  \n* [face_recognition人脸识别应用方式](./facerecognition/ReadMe.md)  \n\n### 其他技术\n\n* [Spark ML的使用方式](./spark/ReadMe.md): 目前该技术的应用场景逐步减少，本教程也是基于较老的版本进行编写，读者需要根据自己的使用\n以及目前最新的文档结合进行对应的API调整。  \n\n\n—————— 以下为未重构的老版本 ————————\n\n## 二、文档目录\n\n### 2.1 目标检测相关 (cnn)  \n\n* [相关基本术语介绍](./docs/cnn/Basic.md)  \n* [介绍关于各类NMS相关的概念以及对应的实现方式](./docs/cnn/NMS.md)  \n* [关于Yolo模型中输入图片尺寸的影响分析](./docs/cnn/yolo/InputSize.md)  \n* [针对Yolo训练结果的评估验证](./docs/cnn/yolo/Evaluation.md)  \n* [数据增强技术的分析](./docs/cnn/DataAugmentation.md)  \n* [边缘检测图像增强技术](./docs/cnn/Vague.md)  \n* [yolo网络层剖析](./docs/cnn/yolo/Network.md)  \n* [yolo各个版本的使用方式](./docs/cnn/yolo/Usage.md)\n\n### 2.3 LLM大模型相关  \n\n* [Transformer模型基础知识](./docs/llm/Transformer.md)  \n\n### 2.4 机器学习基础\n\n* [机器学习中的学习方式](./docs/ml/Learning.md)  \n\n### 2.5 机器人基础\n\n* [基础知识内容](./docs/ml/Robotics.md)\n\n### 数据基础知识  \n\n* [统计计算基础知识](https://www.math.pku.edu.cn/teachers/lidf/docs/statcomp/html/_statcompbook/index.html)\n\n#### 正态分布\n\n* [正态分布含义](https://www.zhihu.com/question/56891433)  \n* [高斯分布](https://baijiahao.baidu.com/s?id=1621087027738177317\u0026wfr=spider\u0026for=pc)  \n\n可使用`numpy.random中的randn、standard_normal和normal`返回随机正态分布的数组，其\n中`normal`是[普遍使用](./normal/numpyTest.py)的方法。  \n\n\n* [泊松分布](https://www.matongxue.com/madocs/858)  \n* [伯努利分布](https://www.cnblogs.com/jmilkfan-fanguiju/p/10589773.html)  \n\n\n## 其他算法与工具  \n\n### 扩展算法  \n\n1. [黎曼和估算与面积法](https://zhuanlan.zhihu.com/p/76304788)  \n\n### 指标  \n\n即衡量目标的单位或方法，这里我们列举几个在互联网中比较常见的指标进行说明：  \n\n1. PV：页面浏览树数，即每天的点击数。\n2. UV：独立用户数，即每天每个用户的浏览数。\n3. DAU：日活跃用户数，即每天活跃的用户数量。  \n\n当然指标不仅仅只有上面还有`MAU`、`LTV`和`ARPU`等，每个指标都要满足以下几点：\n\n* 数字化\n* 易衡量\n* 意义清晰  \n* 周期适当  \n* 尽量客观  \n\n### 依赖工具  \n\n1. [matplotlib可视化](https://www.matplotlib.org.cn/)  \n2. [训练模型持久化](https://github.com/joblib/joblib)  \n3. [Sklearn中文文档](https://sklearn.apachecn.org/)  \n4. [将模型持久化为PMML供Java应用运行](https://github.com/jpmml/sklearn2pmml)  \n5. [Java运行PMML模型算法](https://github.com/jpmml/jpmml-evaluator)  \n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Forvillex%2Fmachinelearning","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Forvillex%2Fmachinelearning","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Forvillex%2Fmachinelearning/lists"}