TencentYoutuResearch / CrowdCounting-P2PNet

The official codes for the ICCV2021 Oral presentation "Rethinking Counting and Localization in Crowds: A Purely Point-Based Framework"
Other
423 stars 114 forks source link

提供的权重测试指标无法达到论文报告的水平(Deviation from expected performance on SHTechA dataset) #65

Open ocean146 opened 11 months ago

ocean146 commented 11 months ago

我使用weights文件夹下的SHTechA.pth测试SHA数据的mae和mse分别为58.3和96.5,没有达到论文报告的指标(mae52.8 mse 85.1)


I have used the weights file "SHTechA.pth" from the "weights" folder to test the SHA dataset. However, the mean absolute error (MAE) and mean squared error (MSE) I obtained were 58.3 and 96.5 respectively. These results do not align with the metrics reported in the associated paper (MAE 52.8, MSE 85.1).

DragonBoyL commented 8 months ago

请问你训了多少轮达到的最优指标