About the error from tta_pipeline in CTRL

[SakuraRiven]

Hi, we try to use the tta_pipeline in the cfg.py, such as https://github.com/tusen-ai/SST/blob/main/configs/ctrl/ctrl_cyc_12e.py#L300.

We found the following error:

File "/usr/local/lib/python3.8/site-packages/torch/nn/parallel/distributed.py", line 886, in forward
output = self.module(*inputs[0], **kwargs[0])
File "/usr/local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/usr/local/lib/python3.8/site-packages/mmcv/runner/fp16_utils.py", line 97, in new_func
return old_func(*args, **kwargs)
File "SST/mmdet3d/models/detectors/base.py", line 60, in forward
return self.forward_test(**kwargs)
File "SST/mmdet3d/models/detectors/tracklet_detector.py", line 455, in forward_test
return self.simple_test(points, img_metas, **kwargs)
File "SST/mmdet3d/models/detectors/tracklet_detector.py", line 288, in simple_test
self.from_collate_format(tracklet, gt_tracklet_candidates)
File "SST/mmdet3d/models/detectors/tracklet_detector.py", line 405, in from_collate_format
t.from_collate_format()
AttributeError: 'list' object has no attribute 'from_collate_format'

So, how could we solve the above problem? BTW we could run the test_pipeline successfully. Thanks for your help : )

[SakuraRiven]

solved