paper-with-me

Image Classification 벤치마크

Image Classification on CINIC-10

9개 결과 · ⬇ CSV · JSON

Accuracy

87.77 89.78 91.78 93.79 95.8 2018-10 2026-11 ResNeXt29_2x64d — 91.45 (2018-10-02) DenseNet-121 — 91.26 (2018-10-02) ResNet-18 — 90.27 (2018-10-02) VGG-16 — 87.77 (2018-10-02) NAT-M3 — 94.3 (2020-05-12) NAT-M2 — 94.1 (2020-05-12) NAT-M1 — 93.4 (2020-05-12) efficient adaptive ensembling — 95.064 (2022-06-15) VIT-L/16 (Spinal FC, Background) — 95.8 (2023-05-05) ResNeXt29_2x64d — 91.45 (2018-10-02) NAT-M3 — 94.3 (2020-05-12) efficient adaptive ensembling — 95.064 (2022-06-15) VIT-L/16 (Spinal FC, Background) — 95.8 (2023-05-05)
RankModel AccuracyFLOPSPARAMS PaperCodeYear
1 VIT-L/16 (Spinal FC, Background) 95.80–– Reduction of Class Activation Uncertainty with Background Information dipuk0506/SpinalNet · dipuk0506/uq 2023
2 efficient adaptive ensembling 95.064–– Efficient Adaptive Ensembling for Image Classification 2022
3 NAT-M3 94.3501M8.1M Neural Architecture Transfer human-analysis/neural-architecture-transfer · awesomelemon/encas 2020
4 NAT-M2 94.1411M6.2M Neural Architecture Transfer human-analysis/neural-architecture-transfer · awesomelemon/encas 2020
5 NAT-M1 93.4317M4.6M Neural Architecture Transfer human-analysis/neural-architecture-transfer · awesomelemon/encas 2020
6 ResNeXt29_2x64d 91.45–– CINIC-10 is not ImageNet or CIFAR-10 BayesWatch/cinic-10 · 2024-MindSpore-1/Code6 2018
7 DenseNet-121 91.26–– CINIC-10 is not ImageNet or CIFAR-10 BayesWatch/cinic-10 · 2024-MindSpore-1/Code6 2018
8 ResNet-18 90.27–– CINIC-10 is not ImageNet or CIFAR-10 BayesWatch/cinic-10 · 2024-MindSpore-1/Code6 2018
9 VGG-16 87.77–– CINIC-10 is not ImageNet or CIFAR-10 BayesWatch/cinic-10 · 2024-MindSpore-1/Code6 2018
1–9 / 9 페이지당 10 20 50 100