Static Block Floating-Point Quantization for Convolutional Neural Networks on FPGA.
Hongxiang FanGang WangMartin FeriancXinyu NiuWayne LukPublished in: FPT (2019)
Keyphrases
- floating point
- convolutional neural networks
- fixed point
- convolutional network
- square root
- instruction set
- hardware implementation
- field programmable gate array
- transform coefficients
- fast fourier transform
- floating point arithmetic
- sparse matrices
- low cost
- data structure
- graphics processing units
- interval arithmetic