ERROR with int8

[chrichard]

Using DeepStream-Yolo, when converting the yolov8 ONNX model to int8, the following error occurs. How can I solve this? The machine environment is as follows:
CUDA：12.0
OPENCV：4.8
Driver：525.125.06
GPU：Nvidia P4
Deepstream:6.3

-----------------------LOG--------------------------------------------
gstnvtracker: Loading low-level lib at /opt/nvidia/deepstream/deepstream/lib/libnvds_nvmultiobjecttracker.so
[NvMultiObjectTracker] Initialized
WARNING: [TRT]: CUDA lazy loading is not enabled. Enabling it can significantly reduce device memory usage. See CUDA_MODULE_LOADING in https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#env-vars
WARNING: ../nvdsinfer/nvdsinfer_model_builder.cpp:1487 Deserialize engine failed because file path: /AI/UnisStream/uniszoo/models/yolov8/yolov8n_traffic_b22_gpu0_int8.engine open error
WARNING: [TRT]: CUDA lazy loading is not enabled. Enabling it can significantly reduce device memory usage. See CUDA_MODULE_LOADING in https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#env-vars
WARNING: ../nvdsinfer/nvdsinfer_model_builder.cpp:659 INT8 calibration file not specified/accessible. INT8 calibration can be done through setDynamicRange API in 'NvDsInferCreateNetwork' implementation
WARNING: [TRT]: onnx2trt_utils.cpp:377: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
WARNING: [TRT]: onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
WARNING: [TRT]: Tensor DataType is determined at build time for tensors not marked as input or output.

Building the TensorRT Engine

File does not exist: /AI/models/yolov8/calib_yolov8n_traffic.table
WARNING: [TRT]: Calibration Profile is not defined. Calibrating with Profile 0
WARNING: [TRT]: CUDA lazy loading is not enabled. Enabling it can significantly reduce device memory usage. See CUDA_MODULE_LOADING in https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#env-vars
Load image: /AI/DLDataSet/calibration/00000431.jpg
Progress: 0.020004%
Load image: /AI/DLDataSet/calibration/00001011.jpg
Progress: 0.040008%
Load image: /AI/DLDataSet/calibration/00001151.jpg
Progress: 0.060012%
Load image: /AI/DLDataSet/calibration/00001601.jpg
Progress: 0.080016%
Load image: /AI/DLDataSet/calibration/00001611.jpg
Progress: 0.10002%
Cuda failure: status=700
nvbufsurface: Error(-1) in releasing cuda memory
gstnvtracker: Got -1 destroying nvbufsurface 0x5559b2dd0e80
Cuda failure: status=700
nvbufsurface: Error(-1) in releasing cuda memory
gstnvtracker: Got -1 destroying nvbufsurface 0x5559b2d34c60
[NvMultiObjectTracker] De-initialized