YOLOV4 INT8 Calibration ERROR

[liuyang9709]

File does not exist : ../configs/yolov4-kINT8-batch2.engine
Building the TensorRT Engine...
Using cached calibration table to build the engine
Using cached calibration table to build the engine
WARNING: TensorRT was linked against cuBLAS/cuBLAS LT 11.5.1 but loaded cuBLAS/cuBLAS LT 11.4.2
WARNING: Detected invalid timing cache, setup a local cache instead
ERROR: 2: [reformatRunner.cpp::onShapeChangeNONCONST::104] Error Code 2: Internal Error (Assertion mCombinedScalesSize >= channelDst failed.)
yolo-trt: /work/app/yolo-tensorrt/modules/yolo.cpp:488: void Yolo::createYOLOEngine(nvinfer1::DataType, Int8EntropyCalibrator*): Assertion `m_Engine != nullptr' failed.

Device: RTX A4000
OS: Ubuntu 20.04
CUDA 11.3, CUDNN 8.2.1, TensorRT 8.0.1.6

[wangxudong-cq]

I had the same issue with xavier on yolov3/4/5.
ENV:
Xlib: extension "NV-GLX" missing on display "localhost:11.0".
Xlib: extension "NV-GLX" missing on display "localhost:11.0".
Xlib: extension "NV-GLX" missing on display "localhost:11.0".

• NVIDIA Jetson AGX Xavier [16GB]
  • Jetpack 4.6 [L4T 32.6.1]
  • NV Power Mode: MODE_15W_DESKTOP - Type: 7
  • jetson_stats.service: active
• Libraries:
  • CUDA: 10.2.89
  • cuDNN: 8.2.1.32
  • TensorRT: 8.0.1.6
  • Visionworks: 1.6.0.501
  • OpenCV: 3.4.3 compiled CUDA: YES
  • VPI: ii libnvvpi1 1.1.15 arm64 NVIDIA Vision Programming Interface library
  • Vulkan: 1.2.70

[wangxudong-cq]

the same issue, pls make a reply.

[Lenan22]

File does not exist : ../configs/yolov4-kINT8-batch2.engine Building the TensorRT Engine... Using cached calibration table to build the engine Using cached calibration table to build the engine WARNING: TensorRT was linked against cuBLAS/cuBLAS LT 11.5.1 but loaded cuBLAS/cuBLAS LT 11.4.2 WARNING: Detected invalid timing cache, setup a local cache instead ERROR: 2: [reformatRunner.cpp::onShapeChangeNONCONST::104] Error Code 2: Internal Error (Assertion mCombinedScalesSize >= channelDst failed.) yolo-trt: /work/app/yolo-tensorrt/modules/yolo.cpp:488: void Yolo::createYOLOEngine(nvinfer1::DataType, Int8EntropyCalibrator*): Assertion `m_Engine != nullptr' failed.

Device: RTX A4000 OS: Ubuntu 20.04 CUDA 11.3, CUDNN 8.2.1, TensorRT 8.0.1.6

Please refer to our open source quantization tool ppq, the quantization effect is better than the calibration of tensorrt, if you encounter issues, we can help you solve them.
https://github.com/openppl-public/ppq/blob/master/md_doc/deploy_trt_by_OnnxParser.md