AssertionError when running run_test.py with DSEC-Detection dataset

[trexob1t]

When attempting to run the run_test.py script, an AssertionError is encountered. I followed the guidelines provided in the DSEC-Detection to download, so I am uncertain why this error is occurring.

Command executed

CUDA_VISIBLE_DEVICES=$DEVICE python scripts/run_test.py --config config/dagr-s-dsec.yaml --use_image --img_net resnet50 --checkpoint data/dagr_s_50.pth --batch_size 4 --dataset_directory $DSEC_ROOT --output_directory $LOG_DIR

Error message

  File "scripts/run_test.py", line 67, in <module>
    metrics = run_test_with_visualization(test_loader, ema.ema, dataset=args.dataset)
  File "/dagr/dagr/src/dagr/utils/testing.py", line 28, in run_test_with_visualization
    for i, data in enumerate(tqdm.tqdm(loader, desc=f"Testing {name}")):
  File "/miniconda3/envs/dagr/lib/python3.8/site-packages/tqdm/std.py", line 1181, in __iter__
    for obj in iterable:
  File "/miniconda3/envs/dagr/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 631, in __next__
    data = self._next_data()
  File "/miniconda3/envs/dagr/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 675, in _next_data
    data = self._dataset_fetcher.fetch(index)  # may raise StopIteration
  File "/miniconda3/envs/dagr/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py", line 51, in fetch
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "/miniconda3/envs/dagr/lib/python3.8/site-packages/torch/utils/data/_utils/fetch.py", line 51, in <listcomp>
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "/dagr/dagr/src/dagr/data/dsec_data.py", line 144, in __getitem__
    detections_1 = self.dataset.get_tracks(image_index_1, mask=track_masks, directory_name=dataset.root.name)
  File "/dagr/dagr/libs/dsec-det/src/dsec_det/dataset.py", line 110, in get_tracks
    i_0, i_1 = self.get_index_window(index, len(img_idx_to_track_idx), sync=self.sync)
  File "/dagr/dagr/libs/dsec-det/src/dsec_det/dataset.py", line 101, in get_index_window
    assert 0 <= index < num_idx - 1
AssertionError

[youkangGG]

Same error on my machine :(

[lpyQaQ]

I can run "run_test.py" successfully. However, I meet the same error when I try to run "count_flops.py".
Have you solve this problem?

[trexob1t]

@lpyQaQ unfortunately, I haven’t solved the problem. Since I only needed an event-based detection model, I switched to AEGNN, which worked out of the box for me.

[Hatins]

I can run "run_test.py" successfully. However, I meet the same error when I try to run "count_flops.py". Have you solve this problem?

Hi, I'd like to ask you which model (ResNet18/Resnet50) do you test? and if you have got the same results provided in the paper?

[lpyQaQ]

I can run "run_test.py" successfully. However, I meet the same error when I try to run "count_flops.py". Have you solve this problem?

Hi, I'd like to ask you which model (ResNet18/Resnet50) do you test? and if you have got the same results provided in the paper?

I tested resnet50. And I got results as below.

[Hatins]

I can run "run_test.py" successfully. However, I meet the same error when I try to run "count_flops.py". Have you solve this problem?

Hi, I'd like to ask you which model (ResNet18/Resnet50) do you test? and if you have got the same results provided in the paper?

I tested resnet50. And I got results as below.

Hi, @lpyQaQ , very thanks for your ripid reply, but it seems that the results are much lower than the results in the paper?