Error while training the detector

[linzhi123]

Hello, I have this error when I am training the detector. What is this? File "alpr-unconstrained/train-detector.py", line 78, in YY = labels2output_map(llp,pts,dim,model_stride)

File "alpr-unconstrained/src/sampler.py", line 27, in labels2output_map iou = IOU_centre_and_dims(mn/MN,label.wh(),label.cc(),label.wh())

File "alpr-unconstrained/src/utils.py", line 112, in IOU_centre_and_dims return IOU(cc1-wh1/2.,cc1+wh1/2.,cc2-wh2/2.,cc2+wh2/2.)

File "alpr-unconstrained/src/utils.py", line 103, in IOU assert((wh1>=.0).all() and (wh2>=.0).all()) AssertionError

I do not understand

[sergiomsilva]

This means that you have wrong annotated labels, where the calculation of its height and width generated a negative number.

[linzhi123]

You mean, heights can't be bigger than widths? @sergiomsilva

[sergiomsilva]

It's not common but they can. Anyway, I think you are having problems related to the format of your annotations, as you posted in another issue. Take a look at #15 and download the new file containing the updated README.

[linzhi123]

alright, thank you very much

[linzhi123]

Excuse me, when my loss drops to 2.x, it will not drop anymore. What is the reason?

[li675599061]

@linzhi123 你好，请问第二个网络的标注要怎么做，怎么使用标注工具呢？