Can Yang

@jarro2783 The current help print like this ``` --log_level arg Log level (default: 3) --network arg Network file name --id arg Network id name (default: id) ``` When the user...

@nineKnight Thanks. The reason for me to come up with this question is that I want my program to also print out help if there is no option provided. I...

I had the same issue  I think one case is when it tries to split a long edge into shorter ones according to the intersections. The...

@zhangxiaoyang520 roi的输入来自于rpn的输出，它的输入在训练阶段，你检查该文件第198行可以知道它训练的时候采用的输入为 ``` loss_class = model_classifier.train_on_batch([X, X2[:, sel_samples, :]],[Y1[:, sel_samples, :], Y2[:, sel_samples, :]]) ``` 在对应到145到153行 ``` loss_rpn = model_rpn.train_on_batch(X, Y) P_rpn = model_rpn.predict_on_batch(X) result = roi_helpers.rpn_to_roi(P_rpn[0], P_rpn[1], cfg, K.image_dim_ordering(),...