Recent Docs
Recently Updated Pages
data type
ONNX OP statistics
alphabet order 3 Abs 4325 Add 20 And 2 ArgMax 101 AveragePool 34...
reference
量化技术背景 从CNN量化说起 在传统CNN网络中,为了加速网络的推理速度,一种非常有效的方法是INT8量化 ,即将权重与激活(feature map) 的浮点数值量化成8-bit整型表示...
Pruning
Knowledge Distillation
Model Quantization
Quantization Granularity Quantization is a magic spell to reduce the memory footprint of a model...