EfficientNet-W4A4

Reported on 3 benchmarks across 1 task · 1 paper

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

Methodology3 results

QuantizationonImageNet
Activation bits· 2020-04-20
4
best: 8 (FQ-ViT (ViT-L))
LSQ+: Improving low-bit quantization through learnable offsets and better initialization arXiv:2004.09576
QuantizationonImageNet
Top-1 Accuracy (%)· 2020-04-20
73.8
best: 85.03 (FQ-ViT (ViT-L))
LSQ+: Improving low-bit quantization through learnable offsets and better initialization arXiv:2004.09576
QuantizationonImageNet
Weight bits· 2020-04-20
4
best: 8 (FQ-ViT (ViT-L))
LSQ+: Improving low-bit quantization through learnable offsets and better initialization arXiv:2004.09576