Time sequence processing accelerator based on one-dimensional convolutional neural network

The invention discloses a time sequence processing accelerator based on a one-dimensional convolutional neural network, and belongs to the field of artificial intelligence and integrated circuit design, and the accelerator comprises an input processing module which comprises N rows of register groups, the number of registers in the first row of register group is N, and the number of registers in each row of register group is subtracted by one row by row; under the cooperative control of the global control module, reasoning data are sequentially input through a register reg1N, transversely flow in a first row of register groups and then are output through a register reg11, data in the first row of register groups longitudinally flow in each row of register groups and then are output through a register regnn, and n = 2, 3,-N; the convolution operation array carries out convolution operation and activation on data output by the input processing module, the pooling processing module pools an activation result and