The official implementation of paper PreNAS: Preferred One-Shot Learning Towards Efficient Neural Architecture Search
-
Updated
Sep 5, 2023 - Python
The official implementation of paper PreNAS: Preferred One-Shot Learning Towards Efficient Neural Architecture Search
Zero-cost trainability signals for hyperparameter prediction: a real finding, a headline number we walked back, and a mathematical bug we couldn't fix.
A list of Training-Free NAS papers.
LightMix: Multi-Objective Search for Lightweight Mixed-Scale Convolutional Neural Networks
Efficient Perturbation-Aware Distinguishing Score for Zero-Shot Neural Architecture Search
In this repository, you'll find the application of NAS algorithm on NATS-Bench together with some variations to have better result.
To associate your repository with the zero-cost-proxies topic, visit your repo's landing page and select "manage topics."