dankernel
dankernel
Issue : https://github.com/opencv/cvat/issues/6451
@kirill-sizov Thank you! I've resolved the conflicts. While this PR works, it might not be perfect. I kindly ask the main contributors to review it in line with the intention...
@kirill-sizov Thank you. I've modified the code according to your guide comments. Please check.
@kirill-sizov Thank you. I've modified the code according to your guide comments. Please check.
In general, onnx can transform the model, but preprocessing and postprocessing does not. `tokenizer.decode` is post-processing outside of the model, so it is not converted along with the model. In...
@zhiltsov-max From the perspective of the MLOps workflow cycle, datasets can continually be added. In this case, it is more efficient to export tasks individually rather than exporting the cumulative...
same problem
https://github.com/Peterande/D-FINE/blob/39b7c469a05af26911e79ff27da25af8ba249eb5/src/solver/det_engine.py#L135-L147 I also have this problem with a custom dataset where num_classes is 1 and categories id starts from 1. In my case, the values of results["labels"] and targets["labels"] are...