RTDETR conversion error

[shreejalt]

Hi, I tried to use your repo but unable to convert the model to TRT. Below is the log file for RTDETR

[04/10/2024-15:24:17] [I] [TRT] [MemUsageChange] Init CUDA: CPU +221, GPU +0, now: CPU 249, GPU 16196 (MiB)
[04/10/2024-15:24:19] [I] [TRT] [MemUsageChange] Init builder kernel library: CPU +302, GPU +431, now: CPU 574, GPU 16646 (MiB)
[04/10/2024-15:24:19] [I] [TRT] ----------------------------------------------------------------
[04/10/2024-15:24:19] [I] [TRT] Input filename:   ../weights/rtdetr/rtdetr_hgnetv2_x_6x_coco.onnx
[04/10/2024-15:24:19] [I] [TRT] ONNX IR version:  0.0.8
[04/10/2024-15:24:19] [I] [TRT] Opset version:    16
[04/10/2024-15:24:19] [I] [TRT] Producer name:
[04/10/2024-15:24:19] [I] [TRT] Producer version:
[04/10/2024-15:24:19] [I] [TRT] Domain:
[04/10/2024-15:24:19] [I] [TRT] Model version:    0
[04/10/2024-15:24:19] [I] [TRT] Doc string:
[04/10/2024-15:24:19] [I] [TRT] ----------------------------------------------------------------
[04/10/2024-15:24:20] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[04/10/2024-15:24:48] [W] [TRT] DLA requests all profiles have same min, max, and opt value. All dla layers are falling back to GPU
[04/10/2024-15:24:48] [E] [TRT] 4: [network.cpp::validate::3088] Error Code 4: Internal Error (im_shape: dynamic input is missing dimensions in profile 0.)
[04/10/2024-15:24:48] [E] [TRT] 2: [builder.cpp::buildSerializedNetwork::751] Error Code 2: Internal Error (Assertion engine != nullptr failed. )
object_detection: /media/user/sdcard/orin/TensorRT_Inference_Demo/src/basemodel.cpp:57: void Model::OnnxToTRTModel(): Assertion `data' failed.
Aborted (core dumped)

Can you tell me how to disable dynamic batch in the config file?