TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Engineering is the bottleneck in Deep Learning research

2 pointsby pramodbiligiriover 8 years ago

1 comment

saipover 8 years ago
Agreed. The tooling around deep learning is not as mature as the tooling around software development. There is a fair amount of engineering and grunt work needed to even get started, let alone build on others&#x27; research. A few problems from top of mind:<p>- Setup: Installing DL frameworks, Nvidia drivers and CUDA is an exercise in dependency hell. Trying to run someone&#x27;s project, which has different dependencies than what you have is difficult to get right. Docker images [1] and nvidia-docker make this simple, but are still not the norm.<p>- Reproducibility: This is big as Denny mentions. Folks still use Github for sharing code. But DL pipelines need versioning of more than just code. It&#x27;s code, environment, parameters, data and results.<p>- Sharing and collaboration: I&#x27;ve noticed that most collaboration on deep learning research, unlike software, happens only when the folks are co-located (e.g. part of the same school or company). This likely links back to reproducibility, but there are not many good tools for effective collaboration currently IMHO.<p>[1] <a href="https:&#x2F;&#x2F;github.com&#x2F;floydhub&#x2F;dl-docker" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;floydhub&#x2F;dl-docker</a> (Disclaimer: I created this)