{"id":17182261,"url":"https://github.com/geektutu/interview-questions","last_synced_at":"2025-05-16T05:04:10.477Z","repository":{"id":51240569,"uuid":"132635545","full_name":"geektutu/interview-questions","owner":"geektutu","description":"机器学习/深度学习/Python/Go语言面试题笔试题(Machine learning Deep Learning Python and Golang Interview Questions)","archived":false,"fork":false,"pushed_at":"2024-08-06T18:06:40.000Z","size":3086,"stargazers_count":1095,"open_issues_count":3,"forks_count":207,"subscribers_count":22,"default_branch":"master","last_synced_at":"2025-05-16T05:03:22.269Z","etag":null,"topics":["deep-learning","golang","interview-questions","machine-learning","python"],"latest_commit_sha":null,"homepage":"https://geektutu.com/post/qa-ml.html","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/geektutu.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":"2018-05-08T16:23:08.000Z","updated_at":"2025-05-11T05:00:38.000Z","dependencies_parsed_at":"2024-10-30T03:27:04.104Z","dependency_job_id":null,"html_url":"https://github.com/geektutu/interview-questions","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/geektutu%2Finterview-questions","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/geektutu%2Finterview-questions/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/geektutu%2Finterview-questions/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/geektutu%2Finterview-questions/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/geektutu","download_url":"https://codeload.github.com/geektutu/interview-questions/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":254471061,"owners_count":22076585,"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":["deep-learning","golang","interview-questions","machine-learning","python"],"created_at":"2024-10-15T00:36:28.861Z","updated_at":"2025-05-16T05:04:10.433Z","avatar_url":"https://github.com/geektutu.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 机器学习\u0026编程面试笔试题\n\n## 序言\n\n持续整理、更新Python、机器学习(Machine Learning)、深度学习(Deep Learning)领域的面试笔试题(interview questions)。\n\n声明：所有习题系博主花费大量精力整理，请尊重劳动成果，未经许可，禁止原文转载。\n\n## 目录\n\n### 机器学习\n\n- [机器学习面试题 01-10](https://geektutu.com/post/qa-ml-1.html)，[md](qa-ml/qa-ml-1.md)\n- [机器学习面试题 11-20](https://geektutu.com/post/qa-ml-2.html)，[md](qa-ml/qa-ml-2.md)\n\n### Go 语言\n\n- [Go 语言笔试面试题(基础语法)](https://geektutu.com/post/qa-golang-1.html)\n- [Go 语言笔试面试题(实现原理)](https://geektutu.com/post/qa-golang-2.html)\n- [Go 语言笔试面试题(并发编程)](https://geektutu.com/post/qa-golang-3.html)\n- [Go 语言笔试面试题(代码输出)](https://geektutu.com/post/qa-golang-c1.html)\n\n## 相关链接\n\n- [知乎专栏](https://zhuanlan.zhihu.com/geektutu)\n- [Go 语言笔试面试题](https://geektutu.com/post/qa-golang.html)\n- [机器学习笔试面试题](https://geektutu.com/post/qa-ml.html)，[Github](https://github.com/geektutu/interview-questions)\n- [TensorFlow 2.0 中文文档](https://geektutu.com/post/tf2doc.html)，[Github](https://github.com/geektutu/tensorflow2-docs-zh)\n- [TensorFlow 2.0 图像识别\u0026强化学习实战](https://geektutu.com/post/tensorflow2-mnist-cnn.html)，[Github](https://github.com/geektutu/tensorflow-tutorial-samples)\n\n## 选择题示例\n\n使用决策树分类时，如果输入的某个特征的值是连续的，通常使用二分法对连续属性离散化，即根据是否大于/小于某个阈值进行划分。如果采用多路划分，每个出现的值都划分为一个分支，这种方式的最大问题是：\n\n- A 计算量太大\n- B 训练集和测试集表现都很差\n- C 训练集表现良好，测试集表现差\n- D 训练集表现差，测试集表现良好\n\n\u003cdetails\u003e\n\u003csummary\u003e答案\u003c/summary\u003e\n\u003cdiv\u003e\n\n**C** 连续值通常采用二分法，离散特征通常采用多路划分的方法，但分支数不宜过多。\n连续特征每个值都划分为一个分支，容易过拟合，泛化能力差，导致训练集表现好，测试集表现差。\n\u003c/div\u003e\n\u003c/details\u003e\n\n\n对神经网络(neural network)而言，哪一项对过拟合(overfitting)和欠拟合(underfitting)影响最大。\n\n- A 隐藏层节点(hidden nodes)数量\n- B 学习速率(learning rate)\n- C 初始权重\n- D 每一次训练的输入个数固定\n\n\u003cdetails\u003e\n\u003csummary\u003e答案\u003c/summary\u003e\n\u003cdiv\u003e\n\n**A** 过拟合和欠拟合与神经网络的复杂程度有关，模型越大越容易过拟合。隐藏层节点数量直接决定了模型的大小与复杂程度。\n\u003c/div\u003e\n\u003c/details\u003e\n\n## 问答题示例\n\n经验误差(empirical error)与泛化误差(generalization error)分别指？\n\n\u003cdetails\u003e\n\u003csummary\u003e答案\u003c/summary\u003e\n\u003cdiv\u003e\n经验误差: 也叫训练误差(training error)，模型在训练集上的误差。\n泛化误差: 模型在新样本集(测试集)上的误差。\n\u003c/div\u003e\n\u003c/details\u003e\n\n简述 K折交叉验证(k-fold crossValidation)。\n\n\u003cdetails\u003e\n\u003csummary\u003e答案\u003c/summary\u003e\n\u003cdiv\u003e\n- 数据集大小为N，分成K份，则每份含有样本N/K个。每次选择其中1份作为测试集，另外K-1份作为训练集，共K种情况。\n- 在每种情况中，训练集训练模型，用测试集测试模型，计算模型的泛化误差。\n- 将K种情况下，模型的泛化误差取均值，得到模型最终的泛化误差。\n\u003c/div\u003e\n\u003c/details\u003e\n\n## 附：题目主要来源\n\n- [Machine Learning exam - CMU](http://www.cs.cmu.edu/~tom/10701_sp11/prev.shtml)\n- [Andrew Ng - coursera](https://www.coursera.org/learn/machine-learning)","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgeektutu%2Finterview-questions","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fgeektutu%2Finterview-questions","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgeektutu%2Finterview-questions/lists"}