Thursday, 8 September 2022

New Show Hacker News story: latest news

Show HN: How simple (but clever) algorithms can find label issues in datasets
2 by calebchiam | 0 comments on Hacker News.
I built Vizzy as a hackathon project, with the goal of explaining how relatively simple (but clever) algorithms can be a powerful tool to automatically find issues in datasets, including label errors and out-of-distribution data. Vizzy uses a JavaScript port of (a part of) https://ift.tt/436VWxO , which implements the algorithms described in https://ift.tt/GrAu8O7 . There are other neat technical nuggets in the implementation of Vizzy as well, including ML model training in the browser (using features from a pretrained ResNet-18, performing truncated SVD, and using an SVM model for speed). If you’re interested in the details of how Vizzy works, check out this blog post: https://ift.tt/HByjt3S I’m happy to answer any questions related to Vizzy, cleanlab, or confident learning and data-centric AI in general!

New Show Hacker News story: latest news

Show HN: I'm building a mind and this is how I plan to do it
3 by lehmannerich | 1 comments on Hacker News.


New Show Hacker News story: latest news

Show HN: Shared – Triptych for data exchange and persistence
2 by alexrustic | 1 comments on Hacker News.


New Show Hacker News story: latest news

Show HN: Twofer Goofer: daily rhyme game built with Midjourney/Stable Diffusion
12 by cowllin | 1 comments on Hacker News.
Twofer Goofers are pairs of rhyming words that form a nonsensical phrase. Each day, there’s a roundabout prompt that describes a new Twofer Goofer. Players can use up to 4 hints. Inspired by a game back from my high school English teacher, I built this in the past few weeks with a couple friends. We’ve been using Midjourney and Stable Diffusion (been on waitlist for Dalle for ages) to generate each day’s art. Extremely fun and simple way to leverage these super-powerful generative art tools! We “playtested” on Google Forms for about a month while building the app w/ Vercel and Supabase. We also use Figma, Midjourney, Stable Diffusion, Sendgrid. Lots of usual suspects!

New Show Hacker News story: latest news

Show HN: 6 Steps to build a usage-based billing system with Lago (YC S21)
3 by Rafsark | 0 comments on Hacker News.


New Show Hacker News story: latest news

Show HN: Kubernetes Native AutoML System
2 by gaocegege | 0 comments on Hacker News.
Katib is a Kubernetes-native project for automated machine learning (AutoML). Katib supports Hyperparameter Tuning, Early Stopping and Neural Architecture Search. Katib is the project which is agnostic to machine learning (ML) frameworks. It can tune hyperparameters of applications written in any language of the users’ choice and natively supports many ML frameworks, such as TensorFlow, Apache MXNet, PyTorch, XGBoost, and others.