推理时，训练集的mIOU总是比验证集低很多？

PaddlePaddle / PaddleSeg

Easy-to-use image segmentation library with awesome pre-trained model zoo, supporting wide-range of practical tasks in Semantic Segmentation, Interactive Segmentation, Panoptic Segmentation, Image Matting, 3D Segmentation, etc.

https://arxiv.org/abs/2101.06175

Apache License 2.0

8.62k stars 1.68k forks source link

推理时，训练集的mIOU总是比验证集低很多？ #3060

Closed m828 closed 3 months ago

m828 commented 1 year ago

问题描述 Please describe your issue

我尝试了ppliteseg在Cityscapes和CamVid数据集上的测试结果，val比test高，很多论文中这两个的结果差不多，我在自己的模型上尝试也是这样，是什么问题？是eval代码块的原因吗

juncaipeng commented 1 year ago

收到我们跟踪看下是否和cityscapes官网评估有差别

m828 commented 1 year ago

收到我们跟踪看下是否和cityscapes官网评估有差别不是和cityscapes官网评估有无差别，我的意思是模型的验证和测试集结果相差较大，

djl00 commented 1 year ago

Hello. I'm also running this model. Can you communicate with me?