Pruning with yolov8

[duong0411]

hi, i just oruning and finetuning yolov8x, prune_rate is 0.8. I converted model pt to onnx include Efficient NMS , and continue convert to TensorRT,but im deploy model on jetson nano, memory CPU is increase compare with model no pruning and ram on gpu is decrease. Can u explain for me ? and i want to decrease memory on CPU