Error with generate_sensor_dataset_visualizations.py

[DuZzzs]

Hi, when i run python tutorials/generate_sensor_dataset_visualizations.py -d /xxx/av2, I got the error: FileNotFoundError: [Errno 2] Failed to open local file '/xxx/av2/test/0c6e62d7-bdfa-3061-8d3d-03b13aa21f68/annotations.feather'. Detail: [errno 2] No such file or directory. The test set has no label. Why is it not filtered out in the code? What is the correct command to run this py file? Thanks.

[DuZzzs]

After filtering out the test data like the following code, I got the correct boxes.

  if self.with_annotations:
      sensor_cache = sensor_cache[sensor_cache.split != 'test']

There is another problem. I want to use 3d_lane for training, but the left and right boundary types of some lanes are inconsistent, such as ls.left_mark_type: LaneMarkType.SOLID_YELLOW, ls.right_mark_type:LaneMarkType.SOLID_WHITE(val, log_id = "4c33fc38-5e59-34f8-96ba-4e5a404d3988"), which one should I choose?

[benjaminrwilson]

Hi @DuZzzs,

Hi, when i run python tutorials/generate_sensor_dataset_visualizations.py -d /xxx/av2, I got the error: FileNotFoundError: [Errno 2] Failed to open local file '/xxx/av2/test/0c6e62d7-bdfa-3061-8d3d-03b13aa21f68/annotations.feather'. Detail: [errno 2] No such file or directory. The test set has no label. Why is it not filtered out in the code? What is the correct command to run this py file? Thanks.

Yes, those files should be filtered if trying to visualize the test dataset input data. Would you mind submitting a PR with your fix?

There is another problem. I want to use 3d_lane for training, but the left and right boundary types of some lanes are inconsistent, such as ls.left_mark_type: LaneMarkType.SOLID_YELLOW, ls.right_mark_type:LaneMarkType.SOLID_WHITE(val, log_id = "4c33fc38-5e59-34f8-96ba-4e5a404d3988"), which one should I choose?

This scenario might be possible (perhaps the road was missing one of the lines). It would be good to visualize, though.

[DuZzzs]

@benjaminrwilson Thanks. I'll try later. Is there a reference code for this kind of picture in the readme?