search

V2AI / Det3D

World's first general purpose 3D object detection codebse.
https://arxiv.org/abs/1908.09492
Apache License 2.0
1.48k stars 299 forks source link

problems when training pointpillars #126

Closed majaya closed 3 years ago

majaya commented 3 years ago

when I train pointpillars(using Kitti), train is normal, but when val, it will goes wrong like this, I guess that self.batch_processsor() function has some bug in val, please check it.

Traceback (most recent call last): File "/home/ljt/pycharm-community-2018.2.8/helpers/pydev/pydevd.py", line 1664, in main() File "/home/ljt/pycharm-community-2018.2.8/helpers/pydev/pydevd.py", line 1658, in main globals = debugger.run(setup['file'], None, None, is_module) File "/home/ljt/pycharm-community-2018.2.8/helpers/pydev/pydevd.py", line 1068, in run pydev_imports.execfile(file, globals, locals) # execute the script File "/home/ljt/pycharm-community-2018.2.8/helpers/pydev/_pydev_imps/_pydev_execfile.py", line 18, in execfile exec(compile(contents+"\n", file, 'exec'), glob, loc) File "/home/ljt/Desktop/Det3D_local/tools/train.py", line 137, in main() File "/home/ljt/Desktop/Det3D_local/tools/train.py", line 132, in main logger=logger, File "/home/ljt/Desktop/Det3D_local/det3d/torchie/apis/train.py", line 327, in train_detector trainer.run(data_loaders, cfg.workflow, cfg.total_epochs, local_rank=cfg.local_rank) File "/home/ljt/Desktop/Det3D_local/det3d/torchie/trainer/trainer.py", line 536, in run epoch_runner(data_loaders[i], self.epoch, kwargs) File "/home/ljt/Desktop/Det3D_local/det3d/torchie/trainer/trainer.py", line 403, in train self.model, data_batch, train_mode=False, kwargs File "/home/ljt/Desktop/Det3D_local/det3d/torchie/trainer/trainer.py", line 374, in batch_processor_inline return model(example, return_loss=False) File "/home/ljt/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/nn/modules/module.py", line 493, in call result = self.forward(*input, **kwargs) File "/home/ljt/Desktop/Det3D_local/det3d/models/detectors/point_pillars.py", line 54, in forward return self.bbox_head.predict(example, preds, self.test_cfg) File "/home/ljt/Desktop/Det3D_local/det3d/models/bbox_heads/mg_head.py", line 775, in predict meta_list, File "/home/ljt/Desktop/Det3D_local/det3d/models/bbox_heads/mg_head.py", line 1046, in get_task_detections ) ^ dir_labels.bool() AttributeError: 'Tensor' object has no attribute 'bool'

tianweiy commented 3 years ago

just change bool() to byte(). some pytorch version problem.

majaya commented 3 years ago

thanks !