Closed paathelb closed 1 year ago
I have the same error!
Hi, I rerun the codes with the default configuration as in this file but cannot repeat this error. Have you made any modifications to the codes or the configuration?
Can you try printing out the min&max values of the 'crop_sub_cloud2d' when the error happens? It may be a rounding error during the image & 2D coordinates reshaping.
I get
out_shape: 112
crop_sub_cloud2d[:, 0].min(): 0.0
crop_sub_cloud2d[:, 0].max(): 111.2745253164557
crop_sub_cloud2d[:, 1].min(): 31.632940751087816
crop_sub_cloud2d[:, 1].max(): 79.43735475178006
Should >=
and <=
be used in assert np.logical_and.reduce([crop_sub_cloud2d[:, 0]>0, crop_sub_cloud2d[:,0]<out_shape, crop_sub_cloud2d[:, 1]>0, crop_sub_cloud2d[:,1] <out_shape]).all()
?
I think the problem should be caused by some 3D points being projected exactly on the 2D box border. The case should happen with a small probability but indeed will result in the assertion error.
Thanks @smueleg for the proposed solution.
We also provide another fix to the code that is more like our original intention. Please see the new commit, at Line215, kitti_detection.py. Specifically, we change >=
and <=
to >
and <
, so that points located on exactly the border will be omitted.
We have tested the above fix, and it will not harm the accuracy.
Thanks for your answer. Despite the fix, I'm still getting the same error.
Thanks for your answer. Despite the fix, I'm still getting the same error.
Hi, have you deleted the 'gt_base' and 'processed' directories under the dataset root? Because the dataset needs to be rebuilt, you will need to manually delete them, so that the code won't be skipped due to the existing files.
You're right, I didn't. Now it works, thank you!
Thanks! Already works for me!
Hello while training, I encountered this error.
===== START TRAINING ===== test
T-0 L:10.91, Seg:56.97, XYZ:3.73, IoU:17.34, R:0.10, Dr:51.20, Cf: 12.75: 100%|█████████████████████████████████████████████████████████████████████████████| 499/499 [02:40<00:00, 3.11it/s] T-1 L:7.31, Seg:69.94, XYZ:1.99, IoU:32.78, R:0.70, Dr:55.52, Cf: 13.49: 100%|██████████████████████████████████████████████████████████████████████████████| 499/499 [02:41<00:00, 3.10it/s] T-2 L:6.29, Seg:73.54, XYZ:1.50, IoU:37.30, R:1.45, Dr:60.53, Cf: 13.76: 100%|██████████████████████████████████████████████████████████████████████████████| 499/499 [02:41<00:00, 3.10it/s] T-3 L:5.79, Seg:75.41, XYZ:1.35, IoU:41.22, R:4.01, Dr:64.24, Cf: 13.34: 100%|██████████████████████████████████████████████████████████████████████████████| 499/499 [02:41<00:00, 3.08it/s] T-4 L:5.29, Seg:77.45, XYZ:1.20, IoU:44.49, R:5.97, Dr:68.00, Cf: 13.13: 100%|██████████████████████████████████████████████████████████████████████████████| 499/499 [02:41<00:00, 3.10it/s] T-5 L:4.90, Seg:78.83, XYZ:1.04, IoU:46.08, R:6.87, Dr:73.57, Cf: 12.19: 100%|██████████████████████████████████████████████████████████████████████████████| 499/499 [02:41<00:00, 3.09it/s] T-6 L:4.80, Seg:79.50, XYZ:1.08, IoU:47.87, R:9.73, Dr:74.72, Cf: 12.14: 100%|██████████████████████████████████████████████████████████████████████████████| 499/499 [02:41<00:00, 3.09it/s] T-7 L:4.61, Seg:79.85, XYZ:0.99, IoU:48.91, R:11.08, Dr:77.78, Cf: 11.99: 100%|█████████████████████████████████████████████████████████████████████████████| 499/499 [02:40<00:00, 3.11it/s] T-8 L:4.38, Seg:80.94, XYZ:0.92, IoU:50.18, R:12.24, Dr:78.23, Cf: 11.90: 100%|█████████████████████████████████████████████████████████████████████████████| 499/499 [02:39<00:00, 3.12it/s] T-9 L:4.31, Seg:81.44, XYZ:0.92, IoU:51.13, R:14.04, Dr:77.98, Cf: 11.77: 100%|█████████████████████████████████████████████████████████████████████████████| 499/499 [02:39<00:00, 3.12it/s] T-10 L:4.27, Seg:81.32, XYZ:0.89, IoU:51.78, R:15.85, Dr:77.63, Cf: 11.89: 100%|████████████████████████████████████████████████████████████████████████████| 499/499 [02:39<00:00, 3.12it/s] T-11 L:4.24, Seg:81.34, XYZ:0.94, IoU:52.46, R:17.85, Dr:80.39, Cf: 11.90: 100%|████████████████████████████████████████████████████████████████████████████| 499/499 [02:39<00:00, 3.12it/s] T-12 L:4.15, Seg:81.45, XYZ:0.85, IoU:52.64, R:18.15, Dr:79.99, Cf: 12.05: 100%|████████████████████████████████████████████████████████████████████████████| 499/499 [02:39<00:00, 3.12it/s] T-13 L:3.98, Seg:82.50, XYZ:0.84, IoU:53.88, R:20.56, Dr:81.49, Cf: 11.92: 100%|████████████████████████████████████████████████████████████████████████████| 499/499 [02:39<00:00, 3.13it/s] T-14 L:3.91, Seg:82.76, XYZ:0.82, IoU:54.59, R:21.41, Dr:82.45, Cf: 11.88: 100%|████████████████████████████████████████████████████████████████████████████| 499/499 [02:39<00:00, 3.12it/s] T-15 L:3.86, Seg:82.69, XYZ:0.82, IoU:55.13, R:23.22, Dr:82.80, Cf: 11.20: 100%|████████████████████████████████████████████████████████████████████████████| 499/499 [02:39<00:00, 3.13it/s] T-16 L:3.75, Seg:83.41, XYZ:0.79, IoU:55.82, R:22.72, Dr:83.80, Cf: 10.78: 100%|████████████████████████████████████████████████████████████████████████████| 499/499 [02:39<00:00, 3.12it/s] T-17 L:2.91, Seg:89.74, XYZ:0.81, IoU:64.15, R:0.00, Dr:75.00, Cf: 9.74: 0%|▏ | 1/499 [00:01<13:03, 1.57s/it] Traceback (most recent call last): File "train.py", line 335, in
main(cfg, args.cfg_file)
File "train.py", line 306, in main
train_one_epoch(cfg, model, training_loader, unlabeled_training_loader, optim, scheduler, counter, histo_counter, epoch, writer)
File "train.py", line 79, in train_one_epoch
unlabeled_data = next(unlabeled_iter)
File "/home/hpaat/.conda/envs/mtrans/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 517, in next
data = self._next_data()
File "/home/hpaat/.conda/envs/mtrans/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 1179, in _next_data
return self._process_data(data)
File "/home/hpaat/.conda/envs/mtrans/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 1225, in _process_data
data.reraise()
File "/home/hpaat/.conda/envs/mtrans/lib/python3.7/site-packages/torch/_utils.py", line 429, in reraise
raise self.exc_type(msg)
AssertionError: Caught AssertionError in DataLoader worker process 1.
Original Traceback (most recent call last):
File "/home/hpaat/.conda/envs/mtrans/lib/python3.7/site-packages/torch/utils/data/_utils/worker.py", line 202, in _worker_loop
data = fetcher.fetch(index)
File "/home/hpaat/.conda/envs/mtrans/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/home/hpaat/.conda/envs/mtrans/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/import/home/hpaat/my_exp/MTrans/datasets/kitti_detection.py", line 233, in getitem
obj = self.load_object_full_data(obj)
File "/import/home/hpaat/my_exp/MTrans/datasets/kitti_detection.py", line 283, in load_object_full_data
assert np.logical_and.reduce([crop_sub_cloud2d[:, 0]>0, crop_sub_cloud2d[:,0]<112, crop_sub_cloud2d[:, 1]>0, crop_sub_cloud2d[:,1]<112]).all()
AssertionError
Any way to fix this?