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。

Energy consumption comparison in machine learning platforms

11 点作者 adrianomartins超过 2 年前

2 条评论

sva_超过 2 年前
Alright, so the more I&#x27;m looking into this website, the weirder it becomes. Never mind even talking about the pricing model.<p>Clicking on their blog, the first entry is<p>&gt; &quot;How are variables in the dataset for machine learning?&quot;[0]<p>That doesn&#x27;t even seem like a valid English sentence to me.<p>Searching the sentences from the text will send you to various sources from which they were taken without being given credit.<p>In fact you can find plenty of sites who are seemingly recycling the the same sentences used in this blog. It&#x27;s pretty bizarre.<p>[0] <a href="https:&#x2F;&#x2F;www.neuraldesigner.com&#x2F;blog&#x2F;type-uses-variables" rel="nofollow">https:&#x2F;&#x2F;www.neuraldesigner.com&#x2F;blog&#x2F;type-uses-variables</a>
sva_超过 2 年前
&gt; TF [...] The final mean squared error is 0.0003.<p>&gt; Neural Designer [...] reaches a mean squared error of 0.023.<p>&gt; The following table summarizes the the[sic] <i>most important metrics</i> that the two machine learning platforms yielded .<p>[omits MSE]<p>They should train both to the same loss and then compare.
评论 #33348997 未加载