LDoubleZhi comments

Results 7 comments of


                                            LDoubleZhi

Hi, have you solve this problem yet?@penguinshin

@chenyuntc

@mahavird

now I'm using endernewton's tf-faster-rcnn..... @cmstudyscode @rxqy

When I am using cv2, the time consuming reduce to 0.007s. But the `output = net(data)` cost 0.1s. I don't konw why.

In fact, When I am runing the code. sometime the net cost 0.01s, and sometime it cost 0.1s. It's ridiculous. I am using single k80

　是否应该先按一个尺度把自己的数据集聚类得到的anchor_size作为参照，然后在random_input_scale时按参照的尺度放缩anchor_size来训练