上一条: Simultaneously optimizing weight and quantizer of ternary neural network using truncated gaussian approximation
下一条: MRIMA: An MRAM-based In-Memory Accelerator