蒸馏
1、Distilling Object Detectors with Fine-grained Feature Imitation(CVPR19)
https://github.com/twangnh/Distilling-Object-Detectors
剪枝
1、soft-filter-pruning(FPGM)
https://github.com/he-y/filter-pruning-geometric-median?utm_source=catalyzex.com
Filter Pruning via Geometric Median for Deep Convolutional Neural Networks Acceleration(CVPR19 Oral)
量化
1、EasyQuant: Post-training Quantization via Scale Optimization
https://github.com/deepglint/EasyQuant
EasyQuant(EQ) is an efficient and simple post-training quantization method via effectively optimizing the scales of weights and activations
2、dnn-gating(PACT)
https://github.com/cornell-zhang/dnn-gating?utm_source=catalyzex.com
PACT: PARAMETERIZED CLIPPING ACTIVATION FOR QUANTIZED NEURAL NETWORKS
3、scale-adjusted-training
https://github.com/jakc4103/scale-adjusted-training?utm_source=catalyzex.com
Towards Efficient Training for Neural Network Quantization
