Design of FPGA-Based Accelerator for Convolutional Neural Network under Heterogeneous Computing Framework with OpenCL

Table 6

Calculating time with different number of compute units.

Compute unit	1	2	3	4	5	6	7	8	9	10
Computing time	27.9	14.5	10.8	7.4	7.3	5.9	5.8	4.2	3.9	3.5