Precision, recall and mAP are 0

WongKinYiu / yolor

implementation of paper - You Only Learn One Representation: Unified Network for Multiple Tasks (https://arxiv.org/abs/2105.04206)

GNU General Public License v3.0

1.98k stars 524 forks source link

Precision, recall and mAP are 0 #243

Closed JYSKKwithm closed 2 years ago

JYSKKwithm commented 2 years ago

I tried training on different datasets, all got the same result.

garyhsu29 commented 2 years ago

How many classes are there in your dataset? Did you update classes and filter size in config file correspondingly?

garyhsu29 commented 2 years ago

You could follow the part 3 of this repository (https://github.com/DataXujing/YOLOR-) to modify your config file. If you have any question or you need translate please let me know.

JYSKKwithm commented 2 years ago

你的数据集中有多少类？您是否相应地更新了配置文件中的类和过滤器大小？ my dataset only one classes, I update classes -1 and filter size -(1+5)*3

garyhsu29 commented 2 years ago

Are you still getting mAP = 0 after updating the config file?

JYSKKwithm commented 2 years ago

您可以按照此存储库的第 3 部分 ( https://github.com/DataXujiing/YOLOR- ) 修改您的配置文件。如果您有任何问题或需要翻译，请告诉我。 Thank you for your advice. Its work!!!