TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Ask HN: How to get some practical experience with ML?

5 点作者 colobas超过 8 年前
I recently did my finished a ML course at my university and I&#x27;ve been also reading up a lot on the subject. However, I need to materialize what I&#x27;ve learned into some practical work, i.e., projects. My ultimate goal would be to become a ML engineer, but I feel like even to get an internship in the area I&#x27;d already need to have some projects to show.<p>What do you recommend? Project-oriented books&#x2F;tutorials?<p>Thanks!

3 条评论

codelitt超过 8 年前
One of our guys started a tutorial series on writing an AI to play the game, DOOM. It assumes very little knowledge and is a fun exercise to do when taking a break from some of the great courses posted here. It&#x27;s no replacement for the excellent course work, but it is fun.: <a href="https:&#x2F;&#x2F;www.codelitt.com&#x2F;blog&#x2F;doom-ai&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.codelitt.com&#x2F;blog&#x2F;doom-ai&#x2F;</a><p>My favourite course for grasping the foundations of the concepts was Andrew NG&#x27;s course (although it seems like you&#x27;re beyond this now): <a href="https:&#x2F;&#x2F;www.coursera.org&#x2F;learn&#x2F;machine-learning" rel="nofollow">https:&#x2F;&#x2F;www.coursera.org&#x2F;learn&#x2F;machine-learning</a><p>I think the best way to learn, is to build things though. Have you checked out the Kaggle challenges? <a href="https:&#x2F;&#x2F;www.kaggle.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.kaggle.com&#x2F;</a> Those will give you great practical skills.
评论 #13674221 未加载
deepaksurti超过 8 年前
Have a look at: <a href="http:&#x2F;&#x2F;www.cs.cmu.edu&#x2F;~10701&#x2F;projects.html" rel="nofollow">http:&#x2F;&#x2F;www.cs.cmu.edu&#x2F;~10701&#x2F;projects.html</a> for project ideas. Project A1, A2, B etc listed on the page. You can then apply the toolkit you used in the ML course and solve these and make a portfolio of your ML projects.<p>Kaggle could be another option.<p>Else a well defined problem that you may want to solve in a domain: example graphics: See some papers here: <a href="http:&#x2F;&#x2F;www.cs.ubc.ca&#x2F;~van&#x2F;research&#x2F;machlearn.html" rel="nofollow">http:&#x2F;&#x2F;www.cs.ubc.ca&#x2F;~van&#x2F;research&#x2F;machlearn.html</a> which you could again solve in software.
评论 #13668575 未加载
colobas超过 8 年前
*I recently finished.<p>Sorry for the typo.