I was able to resolve this by adding an extra dimension to the scores, boxes, and labels in pytorch.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Tensorflow object detection api 2.x model in deepstream 6.0 | 4 | 1009 | December 28, 2021 | |
| Failed to enqueue buffer in fulldims mode because binding idx 0 | 21 | 1616 | October 12, 2021 | |
| The shapes of output in .engine is 0, but the onnx works fine | 2 | 30 | July 17, 2025 | |
| Failed to allocate buffer | 9 | 842 | July 9, 2024 | |
| [defaultAllocator.cpp::deallocate::35] Error Code 1: Cuda Runtime (invalid argument) | 3 | 4012 | March 31, 2022 | |
| Application running problem | 4 | 338 | October 12, 2021 | |
| How to generate a tensorrt model that is supported by Deesptream sdk | 17 | 604 | January 29, 2024 | |
| How to handle outputs with data-dependent shapes while building a TensorRT engine? | 1 | 69 | January 31, 2025 | |
| "Engine buffer is full" | 15 | 3687 | October 12, 2021 | |
| Work with batch in TensorRT | 20 | 3851 | July 20, 2021 |