Floating Point and Fixed Point 32-bits Quantizers for Quantization of Weights of Neural Networks | IEEE Conference Publication | IEEE Xplore