zhulf0804 / PointPillars

A Simple PointPillars PyTorch Implementation for 3D LiDAR(KITTI) Detection.
MIT License
511 stars 124 forks source link

哪里能得到每一轮训练日志文件? #62

Open Hichengdong opened 1 year ago

Hichengdong commented 1 year ago

不是要损失、学习率曲线,而是需要每一轮的训练情况,类似下面这样的: ----------- AP40 Results ------------

Pedestrian AP40@0.50, 0.50, 0.50: bbox AP40:37.4608, 33.8643, 32.5023 bev AP40:37.5405, 33.7067, 31.7938 3d AP40:30.3489, 27.1389, 24.6025 aos AP40:15.51, 14.54, 14.02 Pedestrian AP40@0.50, 0.25, 0.25: bbox AP40:37.4608, 33.8643, 32.5023 bev AP40:46.6569, 42.9096, 41.9350 3d AP40:46.6306, 42.7739, 41.7899 aos AP40:15.51, 14.54, 14.02 Cyclist AP40@0.50, 0.50, 0.50: bbox AP40:60.1731, 46.2010, 44.4464 bev AP40:56.5488, 41.7519, 39.5119 3d AP40:54.3483, 38.4813, 36.1145 aos AP40:53.23, 38.66, 36.49 Cyclist AP40@0.50, 0.25, 0.25: bbox AP40:60.1731, 46.2010, 44.4464 bev AP40:59.8210, 45.5637, 43.2625 3d AP40:59.8200, 45.4073, 43.1876 aos AP40:53.23, 38.66, 36.49 Car AP40@0.70, 0.70, 0.70: bbox AP40:88.2406, 78.1348, 75.7258 bev AP40:90.9977, 79.6726, 76.7591 3d AP40:54.1200, 47.8741, 45.0490 aos AP40:87.85, 77.14, 74.07 Car AP40@0.70, 0.50, 0.50: bbox AP40:88.2406, 78.1348, 75.7258 bev AP40:95.9082, 91.5705, 89.0958 3d AP40:95.6189, 90.4408, 86.5347 aos AP40:87.85, 77.14, 74.07

Overall AP40@easy, moderate, hard: bbox AP40:61.9582, 52.7334, 50.8915 bev AP40:61.6957, 51.7104, 49.3549 3d AP40:46.2724, 37.8314, 35.2553 aos AP40:52.20, 43.45, 41.53