LPRNet_Pytorch icon indicating copy to clipboard operation
LPRNet_Pytorch copied to clipboard

Pytorch Implementation For LPRNet, A High Performance And Lightweight License Plate Recognition Framework.

Results 70 LPRNet_Pytorch issues
Sort by recently updated
recently updated
newest added

用预训练的模型测试所给的测试集,怎么只有0.9的正确率,再训练了自己的数据集之后就只有0.74了。。。

训练一开始还是挺正常的,就是中途test accurary为0.0了,而且最后的best模型也很低,只有0.44

hello, I dont have GPU right now but wants to try your example, is there any way to do so kind regard's

Hey, can you please help me? I was trying to train the model on custom dataset following instructions from [this issue](https://github.com/sirius-ai/LPRNet_Pytorch/issues/12) (_deleted LPRDataLoader.getitem and modified CHARS, though did not change...

您好,在进行测试的时候,当我把测试图片的名字里的中文去掉后再运行,就会出现 Acc = Tp * 1.0 / (Tp + Tn_1 + Tn_2) ZeroDivisionError: float division by zero 如果加上中文,则无法读取到图片,请问该如何解决

可提供一手车牌检测和车牌识别商用级别数据集,数据简介见链接:https://www.bilibili.com/video/bv1kU4y1e7Sb

训练数据集和测试数据集对文件名称有要求吗

您好,很抱歉打扰您,当我在用指定的test文件夹进行测试的时候,出现如下问题 AttributeError: 'NoneType' object has no attribute 'shape' 我仔细检查了路径,应该不是路径问题,不知道您了不了解这种情况?