Show HN: All Vision – AutoML for Unsupervised Learning

9 points by chetan_v 4 years ago

Hi HN! We are Joe and Chetan, co-founders of All Vision Technologies ( https://all.vision/ ). We are building a suite of tools for unsupervised machine learning. We just launched our Unsupervised Model Zoo, which allows you to simultaneously run and compare 14 clustering and anomaly detection algorithms. You can also optimize their hyperparameters with no code. It's available both as a platform with a UI and an API.

Fun fact: we've reimplemented many of these algorithms from scratch, achieving up to 10x speedups in comparison with open-source implementations like scikit-learn.

This release is our first milestone in a larger journey of building powerful and easy to use unsupervised learning tools, aimed at both individual contributors and enterprise customers.

We have a 14-day free trial for anyone who signs up. To get started, make an account at https://all.vision/ and upload any unlabeled dataset to try out the model zoo!

hacker2477 4 years ago |

This looks super interesting! Will check out. Thanks for sharing

katie-k 4 years ago |

Looks super cool! What's the difference between your tool and using scikit learn? Don't they both let you use unsupervised learning?

chetan_v 4 years ago | |

Thanks for asking : So, first with scikit learn, it doesn't work on spark and other distributed frameworks. It's also a bit slower, doesn't provide gpu support for fast inference, can't support larger datasets easily, and runs fewer models than our tool.

The AllVision platform and dev tools tests and runs 10+ unsupervised models at once, so we automate the work needed to find the best model for your dataset. Under the hood, we redesigned many of those unsupervised models so they run significantly faster, so teams can optimize hyperparameters very quickly (hours instead of days), and we're building it on a series of API's so it's more easily deployed within a large scale system.