search

ICEORY / PMF

Perception-aware multi-sensor fusion for 3D LiDAR semantic segmentation (ICCV 2021)
MIT License
191 stars 37 forks source link

在验证时报错：stack expects each tensor to be equal size, but got [19961] at entry 0 and [123389] at entry 1 #29

Open myself0816 opened 1 year ago

myself0816 commented 1 year ago

具体情况如下; ===init env success=== Traceback (most recent call last): File "/home/lcy/桌面/PMF-master/tasks/pmf_eval_semantickitti/infer.py", line 369, in exp.run() File "/home/lcy/桌面/PMF-master/tasks/pmf_eval_semantickitti/infer.py", line 354, in run self.inference.run() File "/home/lcy/桌面/PMF-master/tasks/pmf_eval_semantickitti/infer.py", line 134, in run self.evaluator.addBatch( File "/home/lcy/桌面/PMF-master/tasks/pmf_eval_semantickitti/../../pc_processor/metrics/iou_eval.py", line 44, in addBatch idxs = torch.stack([x_row, y_row], dim=0) RuntimeError: stack expects each tensor to be equal size, but got [19961] at entry 0 and [123389] at entry 1

Process finished with exit code 1

请问该如何修改呢？

myself0816 commented 1 year ago

已经解决了，原因是没有做fov的数据预处理。

birdsheng commented 8 months ago

已经解决了，原因是没有做fov的数据预处理。

你好，遇到了和你同样的问题，请问这个fov数据预处理在哪里改啊，应该如何修改程序啊？

dafengtian commented 8 months ago

已经解决了，原因是没有做fov的数据费用。

你好，遇到了和你同样的问题，请问这个数据丢失在哪里改啊，应该如何修改程序啊？

你好，这个地方错误是因为没有使用PMF-master > tasks > process_semantickitti_fov制作fov数据集。具体如下：

1.进入tasks/process_semantickitti_fov目录，修改create_fov_dataset.pu文件中数据集路径src_root为semantic-kitti实际路径 image

2.运行如下指令执行构建semantic-kitti-fov数据集 python create_fov_dataset.py

数据集生成成功，即可正常运行