Thanh Nguyen

Based on your device setting for `onnxruntime`, I add a simple `if` block to handle the case if specific cuda device id is desirable. https://github.com/Tau-J/rtmlib/blob/875c773bce0ad0225aa9cbd894a4d333fab2575b/rtmlib/tools/base.py#L24

Support CO-DETR models

1

### Motivation CO-DETR is a SoTA model on COCO benchmark, and it is not mentioned in this list: https://github.com/open-mmlab/mmdeploy/blob/main/docs/en/03-benchmark/supported_models.md Would you consider supporting this model? ### Related resources _No response_...

Conversion to TRT failure of TensorRT 8.6.1.6 when converting CO-DETR model on GPU RTX 4090

4

## Description I tried to convert model [CO-DETR](https://github.com/marcoslucianops/DeepStream-Yolo/blob/master/docs/CODETR.md) to TRT, but it fails with error below ```bash [12/12/2024-02:17:38] [E] Error[10]: Could not find any implementation for node {ForeignNode[/0/Cast_3.../0/backbone/Reshape_3 + /0/backbone/Transpose_3]}....

Dear @TempleX98 , @Sense-X , I am trying to obtain instance segmentation results, but I have not been successful yet. I follow this [instance-segmentation-using-mmdetection-on-colab-part-1](https://medium.com/@shantnu2509/instance-segmentation-using-mmdetection-on-colab-part-1-inference-4c7bd3a3f8a1) ```Python def segm(self, img_brg:np.ndarray): result =...