This paper proposes a distribution aware quantization which chooses between recentralized and shift quantizations based on weight distributions in the kernels. The proposed methods is novel, and provides a new general framework to quantize sparse CNNs. Experimental results are extensive and solid, and show the effectiveness of the proposed approach by comparing with the state-of-the-art on well known neural networks. There is also good ablation study. Moreover, the paper is well-written, except some figures are confusing.