NASBench-101

The NASBench-101 dataset is a large-scale benchmark for neural architecture search. It contains nearly 423K unique convolutional neural network architectures with diverse structures and various number of convolutional operations.

Data and Resources

Cite this as

Kiran Seshadri, Berkin Akin, James Laudon, Ravi Narayanaswami, Amir Yazdanbakhsh (2024). Dataset: NASBench-101. https://doi.org/10.57702/lre1njq3

DOI retrieved: December 17, 2024

Additional Info

Field Value
Created December 17, 2024
Last update December 17, 2024
Defined In https://doi.org/10.48550/arXiv.2102.10423
Author Kiran Seshadri
More Authors
Berkin Akin
James Laudon
Ravi Narayanaswami
Amir Yazdanbakhsh
Homepage https://arxiv.org/abs/1907.04452