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。

Friendly Machine Learning for the Web

1 点作者 tdevito超过 4 年前

2 条评论

Jugurtha超过 4 年前
The content is behind an overlay containing a statement and a visitor cannot get to the machine learning content unless they click a button labeled &quot;No Justice, No Peace&quot;, or remove the overlay using browser dev tools or other schemes.<p>How does that work? Do you make requests to deployed models somewhere, and is there a rate limit?<p>I&#x27;m asking because one of the reasons we&#x27;re building our ML platform[0] is that we built many models for clients, but when we wanted to use these models, we had to find them again, bother with dependencies, etc. Now, we can just deploy them and have an endpoint we could hit to get results, which makes it easier for developers to work with HTTP requests, without bothering with the &quot;model&quot; itself.<p>- [0]: <a href="https:&#x2F;&#x2F;iko.ai" rel="nofollow">https:&#x2F;&#x2F;iko.ai</a>
talolard超过 4 年前
Ive been using tfjs and am actually quite impressed with its ease of use, though I do have a lot of experience with ml in python. Curious what the difference is between this and tf , website wasn’t super clear about it