ResNeXt101 32x8d + IG-3.5B + RPL

Reported on 2 benchmarks across 2 tasks · 1 paper

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

Methodology1 result

Domain AdaptationonImageNet-C
mean Corruption Error (mCE)· uses extra data· 2021-04-27
40.9
best: 22 (EfficientNet-L2+RPL)
If your data distribution shifts, use self-learning arXiv:2104.12928

Other1 result

Unsupervised Domain AdaptationonImageNet-C
mean Corruption Error (mCE)· uses extra data· 2021-04-27
40.9
best: 22 (EfficientNet-L2+RPL)
If your data distribution shifts, use self-learning arXiv:2104.12928