TResNet-L-V2

Reported on 2 benchmarks across 1 task · 2 papers · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision2 results

Image ClassificationonStanford Cars
Accuracy· 2021-04-22
96.32
best: 96.868 (efficient adaptive ensembling)
SOTA
ImageNet-21K Pretraining for the Masses arXiv:2104.10972
Image ClassificationonCIFAR-100
Percentage correct· uses extra data· 2020-03-30
92.6
best: 96.08 (EffNet-L2 (SAM))
TResNet: High Performance GPU-Dedicated Architecture arXiv:2003.13630